Split (2)

train · 44.6k rows

paperID stringlengths 36 36	pwc_id stringlengths 8 47	arxiv_id stringlengths 6 16 ⌀	nips_id float64	url_abs stringlengths 18 329	url_pdf stringlengths 18 742	title stringlengths 8 325	abstract stringlengths 1 7.27k ⌀	authors stringlengths 2 7.06k	published stringlengths 10 10 ⌀	conference stringlengths 12 47 ⌀	conference_url_abs stringlengths 16 198 ⌀	conference_url_pdf stringlengths 27 199 ⌀	proceeding stringlengths 6 47 ⌀	taskID stringlengths 7 1.44k	areaID stringclasses 688 values	embedding stringlengths 9.26k 12.5k	umap_embedding stringlengths 29 44
f4651457-4c87-4aea-b22f-133937a582e9	time-dependent-entity-embedding-is-not-all	null	null	https://aclanthology.org/2021.emnlp-main.639	https://aclanthology.org/2021.emnlp-main.639.pdf	Time-dependent Entity Embedding is not All You Need: A Re-evaluation of Temporal Knowledge Graph Completion Models under a Unified Framework	Various temporal knowledge graph (KG) completion models have been proposed in the recent literature. The models usually contain two parts, a temporal embedding layer and a score function derived from existing static KG modeling approaches. Since the approaches differ along several dimensions, including different score functions and training strategies, the individual contributions of different temporal embedding techniques to model performance are not always clear. In this work, we systematically study six temporal embedding approaches and empirically quantify their performance across a wide range of configurations with about 3000 experiments and 13159 GPU hours. We classify the temporal embeddings into two classes: (1) timestamp embeddings and (2) time-dependent entity embeddings. Despite the common belief that the latter is more expressive, an extensive experimental study shows that timestamp embeddings can achieve on-par or even better performance with significantly fewer parameters. Moreover, we find that when trained appropriately, the relative performance differences between various temporal embeddings often shrink and sometimes even reverse when compared to prior results. For example, TTransE (CITATION), one of the first temporal KG models, can outperform more recent architectures on ICEWS datasets. To foster further research, we provide the first unified open-source framework for temporal KG completion models with full composability, where temporal embeddings, score functions, loss functions, regularizers, and the explicit modeling of reciprocal relations can be combined arbitrarily.	['Volker Tresp', 'Yunpu Ma', 'Gengyuan Zhang', 'Zhen Han']	null	null	null	null	emnlp-2021-11	['temporal-knowledge-graph-completion']	['knowledge-base']	[-4.72682387e-01 3.06727011e-02 -5.69812715e-01 -9.11633596e-02 -3.73761147e-01 -5.37437856e-01 8.58007967e-01 4.57242221e-01 -5.02723336e-01 4.16165948e-01 3.91614825e-01 -3.26584727e-01 -4.74650621e-01 -9.69241023e-01 -5.98978281e-01 -5.23319721e-01 -5.25772631e-01 5.27478755e-01 4.83719617e-01 -2.23384023e-01 9.49226245e-02 3.91669869e-01 -1.34569204e+00 -1.91526487e-01 7.31345057e-01 1.06681085e+00 -2.14495584e-01 5.43682635e-01 -1.94365352e-01 8.47427547e-01 -3.38834912e-01 -8.39159310e-01 2.12300256e-01 2.88606107e-01 -8.31435561e-01 -6.50607407e-01 1.38455048e-01 -3.07948470e-01 -7.75051773e-01 6.61568880e-01 5.69958985e-01 1.27099901e-01 4.21518743e-01 -1.62584841e+00 -1.14184320e+00 8.90245855e-01 -3.53273869e-01 1.51904464e-01 2.14191452e-01 -6.26306282e-03 1.41142750e+00 -8.57105553e-01 6.98218942e-01 7.97030509e-01 1.16116953e+00 1.03183180e-01 -1.23322999e+00 -4.37062979e-01 2.86606699e-01 6.31144106e-01 -1.55595601e+00 -1.18500656e-02 5.37229300e-01 -2.52445132e-01 1.41110647e+00 2.78451275e-02 7.82649696e-01 1.19392073e+00 1.24371581e-01 7.42579937e-01 7.00024366e-01 -3.89798373e-01 -4.03980911e-02 -3.96149904e-02 5.56384206e-01 5.76591969e-01 5.66036701e-01 2.05939293e-01 -7.60571122e-01 -5.50844789e-01 3.26534361e-01 1.02816656e-01 -3.80833924e-01 -5.96867144e-01 -1.34652638e+00 7.85997689e-01 3.77205789e-01 4.39181656e-01 -9.64375958e-02 5.29754281e-01 6.93865776e-01 3.11555207e-01 4.62277621e-01 3.76879036e-01 -7.52489328e-01 -3.99576098e-01 -7.14920640e-01 4.92939562e-01 9.53460455e-01 1.17027676e+00 6.95108533e-01 -1.08547032e-01 -2.47787878e-01 6.22507334e-01 6.21749833e-02 -3.80961969e-02 4.57522690e-01 -5.74813724e-01 5.14687240e-01 6.48008525e-01 3.72971520e-02 -1.06404924e+00 -5.26548505e-01 -3.78484964e-01 -4.90984112e-01 -4.41583276e-01 3.80951822e-01 9.40995514e-02 -6.66327834e-01 1.92218590e+00 2.54909247e-01 5.86259544e-01 9.62665956e-03 5.48883617e-01 7.88286150e-01 4.35967326e-01 2.20094129e-01 -6.74618185e-02 1.44401574e+00 -1.16509783e+00 -8.37221026e-01 4.09139618e-02 1.08179200e+00 -5.74583590e-01 9.75927413e-01 1.46191791e-01 -9.32831109e-01 -1.98361710e-01 -1.24554896e+00 -4.87074912e-01 -1.03554761e+00 -1.76233165e-02 1.21397829e+00 5.31460881e-01 -1.30989611e+00 1.00489867e+00 -9.91768181e-01 -6.34785414e-01 -9.79522243e-02 3.05386215e-01 -2.35000163e-01 -1.11969680e-01 -1.70279527e+00 9.94948268e-01 4.97845054e-01 -1.17234863e-01 -3.34926218e-01 -1.05185091e+00 -8.38867784e-01 1.31302968e-01 3.91139388e-01 -1.04368186e+00 1.09069955e+00 -9.22441259e-02 -1.25058496e+00 6.39175534e-01 1.73728317e-01 -7.54568219e-01 4.29831535e-01 -4.52437609e-01 -6.89350247e-01 8.91361199e-03 -2.83262692e-02 2.39194021e-01 6.25683248e-01 -7.87348211e-01 -3.62618417e-01 -3.06053311e-01 3.16404879e-01 5.66386022e-02 -8.26121092e-01 -3.55213210e-02 -9.65960979e-01 -8.11895311e-01 -3.28957230e-01 -8.96700740e-01 -8.39079991e-02 -1.84765503e-01 -1.21255271e-01 -5.98661304e-01 7.58888423e-01 -4.59651738e-01 1.94061458e+00 -2.14089036e+00 2.61259973e-01 -2.03610241e-01 4.05384392e-01 2.09839582e-01 1.59878880e-02 9.82891321e-01 -1.31386563e-01 3.55551630e-01 1.06199339e-01 -4.52475101e-01 4.77778882e-01 4.21488196e-01 -4.26458091e-01 3.27537566e-01 2.78465357e-02 1.27950072e+00 -1.02063525e+00 -4.40441430e-01 3.69408280e-02 5.22416472e-01 -4.54072356e-01 -4.78930548e-02 -1.35419562e-01 -4.02200341e-01 -3.69331419e-01 5.06251931e-01 4.87196803e-01 -5.01649618e-01 4.20451611e-01 -4.38686222e-01 -7.97428191e-02 4.40994173e-01 -1.14880049e+00 1.86570406e+00 -2.31592432e-01 4.24650550e-01 -3.08665752e-01 -8.35073352e-01 4.86527920e-01 4.68153417e-01 6.68912649e-01 -6.28705084e-01 7.47777801e-03 9.25409123e-02 -1.87995866e-01 -3.29995096e-01 1.19128144e+00 9.78207141e-02 -2.06097975e-01 6.55189514e-01 2.77228266e-01 3.63508701e-01 4.16377842e-01 5.40843427e-01 1.42158866e+00 9.18115303e-02 9.54883695e-02 -1.11207716e-01 -6.61748871e-02 -6.52139336e-02 5.94875038e-01 6.47908092e-01 -2.18746051e-01 4.59074676e-01 7.10956454e-01 -5.65546453e-01 -9.42649961e-01 -1.01952064e+00 -2.27161139e-01 1.16666365e+00 1.71282679e-01 -1.17043734e+00 -1.91289201e-01 -6.93200767e-01 4.81367141e-01 5.60930610e-01 -8.34272087e-01 -3.88804495e-01 -6.34418011e-01 -1.03908110e+00 9.32875276e-01 1.19706929e+00 2.07866177e-01 -5.22812903e-01 -2.24460989e-01 3.39512467e-01 1.62680168e-02 -1.26289463e+00 -3.97593677e-01 2.34013021e-01 -9.53872740e-01 -1.05215287e+00 -3.99230212e-01 -5.03571689e-01 2.54547566e-01 2.73712367e-01 1.30493176e+00 1.49366349e-01 -1.91009924e-01 6.23127878e-01 -6.67821288e-01 -1.51869431e-01 3.00167084e-01 2.14723974e-01 3.21408093e-01 -3.28594416e-01 5.71465194e-01 -7.83702135e-01 -6.95086300e-01 8.62505436e-02 -8.71851265e-01 -2.21826211e-01 3.37041080e-01 8.13952804e-01 3.01290661e-01 -7.92803541e-02 4.19294953e-01 -8.48467231e-01 8.29265475e-01 -6.23279810e-01 -2.87191302e-01 3.75905693e-01 -1.21948183e+00 2.24135786e-01 5.74364185e-01 -4.98420984e-01 -6.74552262e-01 -7.36500800e-01 1.81646511e-01 -7.59514332e-01 5.98146439e-01 8.58528316e-01 2.79578269e-01 6.57410873e-03 4.91119266e-01 2.04949260e-01 -2.83385932e-01 -5.38451672e-01 8.89316797e-01 1.66084975e-01 3.97599578e-01 -1.00222600e+00 9.70596015e-01 3.96156222e-01 -3.17633688e-01 -3.29481453e-01 -5.53543270e-01 -4.82287765e-01 -5.29931188e-01 2.48067856e-01 6.42931521e-01 -8.32148015e-01 -6.10305667e-01 1.40302017e-01 -1.10064244e+00 -3.19479406e-01 -4.39387113e-01 5.17630219e-01 -2.82725066e-01 5.95221281e-01 -9.70219374e-01 -3.72499824e-01 -3.31353456e-01 -7.40189254e-01 8.95814121e-01 -8.77876282e-02 -2.33585432e-01 -1.29908693e+00 2.37846300e-01 6.61963448e-02 5.83826125e-01 2.32723817e-01 1.17676961e+00 -7.25333154e-01 -6.63933277e-01 -3.71630728e-01 -3.53528440e-01 1.32225692e-01 -5.76653928e-02 1.71041280e-01 -7.79393196e-01 -2.64379561e-01 -4.31743622e-01 -2.66248375e-01 8.57269704e-01 1.09592170e-01 1.03112328e+00 -2.23173514e-01 -5.62855363e-01 8.16274941e-01 1.62101781e+00 -9.23354700e-02 7.21715271e-01 5.04007220e-01 7.33191490e-01 2.26391777e-01 6.55049026e-01 4.21087414e-01 8.10460210e-01 8.01244438e-01 2.58837044e-01 2.99667150e-01 -1.85596317e-01 -3.25154096e-01 4.26368803e-01 1.26862097e+00 -4.23249990e-01 -2.65711039e-01 -8.97129416e-01 9.22996521e-01 -2.15914798e+00 -9.39111829e-01 -4.72215191e-02 2.09882569e+00 7.82262683e-01 6.61696568e-02 1.12794995e-01 8.76872092e-02 3.09717566e-01 5.70960045e-01 -3.61938775e-01 -5.45109153e-01 -6.85796663e-02 2.79231638e-01 7.63048410e-01 1.50275618e-01 -9.37023699e-01 8.75027001e-01 7.16159630e+00 8.09463084e-01 -8.73758614e-01 3.27149808e-01 -1.43095508e-01 -2.07270280e-01 -3.84222090e-01 2.02612385e-01 -8.25215042e-01 3.33876252e-01 1.16918123e+00 -7.26745844e-01 2.70010769e-01 7.75111556e-01 -3.47993970e-01 2.77326941e-01 -1.30240560e+00 8.99751484e-01 -5.20541556e-02 -1.50594735e+00 -3.48834060e-02 1.04709782e-01 6.57830477e-01 2.83720583e-01 5.07632010e-02 7.39867806e-01 6.87058449e-01 -8.76005352e-01 5.80284655e-01 4.15612698e-01 6.76036417e-01 -6.20652854e-01 6.31516457e-01 -2.61708610e-02 -1.64554441e+00 -9.67855826e-02 -2.59491295e-01 -1.63124502e-01 1.43168762e-01 5.78343153e-01 -5.13654828e-01 1.35899818e+00 9.12120581e-01 1.01764226e+00 -6.41572833e-01 8.42261016e-01 -2.74680942e-01 5.30381620e-01 -3.60614508e-01 1.01453885e-01 2.74405271e-01 -1.85385242e-01 3.54652226e-01 1.31556726e+00 3.23358595e-01 -4.75798510e-02 -2.30869174e-01 6.36935413e-01 -1.79985359e-01 -5.46646900e-02 -7.04208791e-01 -4.16468620e-01 7.29664743e-01 1.16073406e+00 -3.72843921e-01 -3.64803106e-01 -7.41000175e-01 8.07986379e-01 7.08312392e-01 5.28339624e-01 -1.23722076e+00 -4.50341731e-01 9.18911397e-01 1.44461274e-01 5.43833792e-01 -6.43527210e-01 -9.11741406e-02 -1.33577466e+00 2.55110621e-01 -3.50619674e-01 7.48703659e-01 -5.49189031e-01 -1.43160248e+00 4.13634598e-01 2.07156673e-01 -1.05224967e+00 -1.38340667e-01 -6.68055892e-01 -4.59948659e-01 5.91625094e-01 -1.75934303e+00 -1.27163386e+00 -1.40418217e-01 4.79702055e-01 -5.80145717e-02 2.77718991e-01 9.48195934e-01 7.44641066e-01 -7.40777016e-01 7.79054999e-01 2.62698442e-01 1.64283901e-01 8.62596571e-01 -1.47108018e+00 4.85935807e-01 6.97619796e-01 1.70988575e-01 1.08139241e+00 5.04923642e-01 -5.50252497e-01 -1.77763760e+00 -1.09433544e+00 1.12642455e+00 -8.03285420e-01 1.23244095e+00 -3.84016871e-01 -9.48308766e-01 1.20279467e+00 2.92093277e-01 2.45719656e-01 7.46681392e-01 7.55834281e-01 -7.88783312e-01 -1.67159140e-01 -5.82824826e-01 7.72419751e-01 1.42146122e+00 -6.26882315e-01 -5.66143453e-01 2.34905541e-01 1.19621348e+00 -2.34814301e-01 -1.53849089e+00 6.71278059e-01 6.29059672e-01 -8.08782756e-01 1.15517890e+00 -6.88927650e-01 1.62919968e-01 -3.00389290e-01 -8.37837309e-02 -1.03946936e+00 -5.21864831e-01 -6.24992490e-01 -8.06851864e-01 1.31423366e+00 2.02569291e-01 -8.80299926e-01 6.53618693e-01 7.00849295e-01 -1.82177499e-01 -1.15285611e+00 -7.49118626e-01 -1.28601956e+00 1.21479318e-01 -7.66842842e-01 8.88051629e-01 1.31144226e+00 2.18752041e-01 1.20370828e-01 -3.32732379e-01 1.69691265e-01 3.90286595e-01 3.19685459e-01 7.18459487e-01 -1.17327428e+00 -1.79035574e-01 -5.02044082e-01 -5.05702078e-01 -1.01346827e+00 -1.83339510e-02 -1.13106894e+00 -7.64690876e-01 -1.72554946e+00 1.26193628e-01 -5.89209437e-01 -6.79290116e-01 7.65027583e-01 -5.89795858e-02 -1.18635945e-01 -1.18982874e-01 3.16791058e-01 -5.91444135e-01 8.37134600e-01 8.25617790e-01 -5.77219203e-02 -8.70767422e-03 -5.95542908e-01 -6.90377712e-01 4.10750955e-01 5.06683350e-01 -3.97802502e-01 -6.42780483e-01 -6.44187629e-01 6.51263595e-01 -1.93138346e-01 2.70046532e-01 -7.25545228e-01 6.35392785e-01 1.15611136e-01 -2.06582204e-01 -5.51744103e-01 3.32188070e-01 -6.81139886e-01 3.75214815e-01 1.62664995e-01 1.69299722e-01 3.94027591e-01 3.15241933e-01 9.89956677e-01 -2.76458442e-01 1.31565303e-01 1.77443046e-02 2.64852792e-01 -9.93809938e-01 6.41694903e-01 1.38125181e-01 1.54759930e-02 1.02822506e+00 -3.91131669e-01 -6.14866376e-01 -4.90083322e-02 -6.15979671e-01 4.05506730e-01 4.63033110e-01 7.01767385e-01 4.13095981e-01 -1.63689053e+00 -3.60103548e-01 -3.16235125e-01 4.09367710e-01 -1.99220851e-01 2.03923017e-01 1.19748557e+00 -2.70041198e-01 5.11529446e-01 2.08112568e-01 -2.75628597e-01 -8.17671418e-01 8.74199986e-01 -1.77841708e-02 -8.74529541e-01 -7.70220876e-01 6.29790127e-01 -9.79277045e-02 -4.69089925e-01 2.72584319e-01 -5.34549654e-01 7.56374076e-02 3.55570614e-01 2.53669441e-01 5.17282844e-01 3.18894327e-01 -1.33449480e-01 -4.78622109e-01 5.90012729e-01 -3.48137021e-01 2.23062322e-01 1.44935739e+00 1.15105994e-01 -1.22419775e-01 4.94341075e-01 1.25972617e+00 -1.22508027e-01 -8.25446188e-01 -5.63850462e-01 1.93472117e-01 -3.11177760e-01 -1.65353969e-01 -5.30516863e-01 -1.05396283e+00 6.15761101e-01 1.73415199e-01 4.01826203e-01 1.07435441e+00 -7.12469295e-02 1.01914954e+00 7.75966495e-02 6.81734741e-01 -1.18993938e+00 8.67409855e-02 5.35555184e-01 5.91062546e-01 -8.53973866e-01 2.72570252e-01 -3.04464161e-01 -4.38953668e-01 9.79270518e-01 4.25551981e-01 -5.59408218e-02 8.31163466e-01 3.02339345e-01 -4.95935947e-01 -5.51723123e-01 -1.13596499e+00 -2.78332949e-01 3.77821140e-02 4.46510881e-01 3.46683472e-01 -2.32346021e-02 -6.89450860e-01 8.84415090e-01 -2.42000446e-01 -9.87086073e-03 1.80911064e-01 9.18896317e-01 6.97067976e-02 -1.47982776e+00 1.44937083e-01 4.21758980e-01 -4.01622206e-01 -2.11609170e-01 -5.71859665e-02 1.30149758e+00 -2.30186842e-02 6.52838945e-01 -1.05079241e-01 -6.42533779e-01 4.16729510e-01 2.84552157e-01 5.22861302e-01 -4.59608108e-01 -6.30264878e-01 -4.57734466e-01 3.03343475e-01 -6.78673387e-01 -3.05535585e-01 -4.93345290e-01 -1.13263381e+00 -6.87405646e-01 -5.67226052e-01 8.47995430e-02 3.10194224e-01 6.05912328e-01 7.91077375e-01 5.22125006e-01 1.25943169e-01 -4.36520875e-01 -4.59707558e-01 -8.92600954e-01 -8.27735841e-01 3.53381395e-01 -1.23034850e-01 -8.96094680e-01 -4.01821434e-01 -1.75897449e-01]	[8.625533103942871, 7.841413497924805]
c25667a7-5283-49c0-930c-819479e0122a	multi-agent-path-finding-with-deadlines-1	1805.04961	null	http://arxiv.org/abs/1805.04961v1	http://arxiv.org/pdf/1805.04961v1.pdf	Multi-Agent Path Finding with Deadlines: Preliminary Results	We formalize the problem of multi-agent path finding with deadlines (MAPF-DL). The objective is to maximize the number of agents that can reach their given goal vertices from their given start vertices within a given deadline, without colliding with each other. We first show that the MAPF-DL problem is NP-hard to solve optimally. We then present an optimal MAPF-DL algorithm based on a reduction of the MAPF-DL problem to a flow problem and a subsequent compact integer linear programming formulation of the resulting reduced abstracted multi-commodity flow network.	['Ariel Felner', 'Sven Koenig', 'Glenn Wagner', 'T. K. Satish Kumar', 'Jiaoyang Li', 'Hang Ma']	2018-05-13	null	null	null	null	['multi-agent-path-finding']	['playing-games']	[-2.01405719e-01 4.68276232e-01 -3.32372934e-01 -2.31215566e-01 -8.02541599e-02 -9.88752604e-01 -7.26347510e-03 5.24365246e-01 -3.98503929e-01 1.22921419e+00 -3.14065963e-01 -2.06663549e-01 -9.44784880e-01 -1.09798932e+00 -5.90433061e-01 -3.62805098e-01 -1.03599393e+00 1.27935123e+00 3.91414553e-01 -2.41011307e-01 2.19222218e-01 7.08865345e-01 -6.60961270e-01 -1.20384008e-01 7.27656841e-01 5.33312738e-01 3.36012185e-01 8.50649774e-01 7.36545771e-02 4.60234821e-01 -6.26000822e-01 -2.09044367e-01 7.24804044e-01 1.64477043e-02 -1.38160205e+00 4.15876091e-01 -2.37857521e-01 -7.25366116e-01 -4.15180951e-01 8.58667433e-01 -1.00538976e-01 2.22146422e-01 4.71401781e-01 -2.75746226e+00 -4.74907726e-01 7.45139778e-01 -9.28733706e-01 2.07462221e-01 6.85025811e-01 -1.17130488e-01 1.19564128e+00 -1.73291400e-01 1.08419394e+00 1.37903476e+00 1.23241968e-01 3.16731095e-01 -1.16779816e+00 3.86055969e-02 4.06354159e-01 1.42410502e-01 -1.02035010e+00 -3.61149199e-02 4.12262790e-02 -3.29473466e-01 1.32916963e+00 3.44332039e-01 5.88356018e-01 -1.48420081e-01 1.87386900e-01 4.98924434e-01 6.35669112e-01 -4.34825093e-01 1.10513449e-01 -5.71714565e-02 2.16877565e-01 4.56201315e-01 6.83432043e-01 1.33796006e-01 -1.67208984e-01 -2.59811103e-01 7.64905512e-01 -2.14545012e-01 -2.66683072e-01 -3.05644572e-01 -1.15255642e+00 1.02414405e+00 3.70781988e-01 -3.47379670e-02 -4.77360547e-01 4.58058089e-01 4.21579212e-01 7.00243592e-01 2.13672966e-01 3.57158124e-01 -5.50376296e-01 3.89509946e-01 -3.29299420e-01 6.97785139e-01 1.25508511e+00 1.56703460e+00 7.44604647e-01 -3.01430523e-01 6.27669543e-02 1.49609730e-01 5.17266214e-01 5.31865954e-01 -9.74312544e-01 -1.47349072e+00 7.94644654e-01 4.17189896e-01 6.36311591e-01 -9.61127877e-01 -7.69249260e-01 9.27362069e-02 -1.38552099e-01 7.00375497e-01 5.18349707e-01 -4.40744638e-01 -2.35470429e-01 1.63404965e+00 3.43369305e-01 -2.84402698e-01 1.56435356e-01 1.00704992e+00 2.94198096e-01 1.36468112e+00 -2.91563272e-01 -8.59580934e-01 1.07229674e+00 -1.63536370e+00 -4.13142622e-01 -2.90050536e-01 6.86558962e-01 -5.68786025e-01 -1.07611589e-01 1.59921870e-02 -1.45629311e+00 4.35919315e-01 -8.59035969e-01 1.48880869e-01 -1.78660288e-01 -6.79284990e-01 6.50354862e-01 2.93304473e-01 -1.39673328e+00 2.52386332e-01 -5.21506190e-01 -2.96271414e-01 2.56338920e-02 7.21593142e-01 -5.96914470e-01 -4.47015285e-01 -8.35953891e-01 8.68965685e-01 5.88105857e-01 -3.55629548e-02 -9.02251661e-01 -7.37451375e-01 -7.83479631e-01 8.74506235e-02 8.85388136e-01 -7.66951084e-01 1.33918977e+00 -6.58904076e-01 -9.83415782e-01 7.46100724e-01 -6.93944693e-02 -1.91638440e-01 3.38230878e-01 6.76401258e-01 -2.08762571e-01 4.27935421e-01 6.07983649e-01 6.08568072e-01 1.82863086e-01 -1.28347158e+00 -1.24049175e+00 -1.96064152e-02 1.13687789e+00 2.66096532e-01 3.42843048e-02 2.69390345e-01 -1.93284944e-01 -2.01242324e-02 -5.19223332e-01 -1.14736068e+00 -7.36176610e-01 2.06259742e-01 -4.52731937e-01 -5.38366199e-01 5.59585035e-01 -1.12314776e-01 9.62851286e-01 -1.49977481e+00 3.79374146e-01 4.84725058e-01 5.36564767e-01 -3.11455950e-02 -6.03088856e-01 1.09592259e+00 2.81760246e-01 6.14864193e-02 9.42724347e-02 -7.09551945e-02 2.71561354e-01 4.74612713e-01 1.35398015e-01 7.63221145e-01 -3.29951316e-01 4.83895093e-01 -1.27903402e+00 -2.95899332e-01 -1.25224873e-01 -4.09883827e-01 -6.23771608e-01 7.35445395e-02 -4.77864116e-01 -1.42470822e-01 -3.75822634e-01 5.17875016e-01 1.17284358e+00 -5.90196252e-02 4.76177067e-01 4.78251487e-01 -5.64261436e-01 -2.02927291e-01 -1.38264477e+00 1.32765234e+00 -1.25066862e-01 6.07580483e-01 9.77082133e-01 -8.49955142e-01 4.48981225e-01 1.66834176e-01 1.17285740e+00 -5.63593507e-01 -1.83021709e-01 3.54733646e-01 -1.37753263e-01 -3.01654756e-01 6.99906051e-01 -7.22746626e-02 -3.06825578e-01 1.11783648e+00 -3.82180810e-01 4.43263322e-01 9.96537805e-01 8.26244831e-01 1.27664280e+00 -5.36935210e-01 5.24130426e-02 -5.80664992e-01 3.85115713e-01 5.58263183e-01 8.29621911e-01 5.91261804e-01 -3.75540793e-01 -3.42505604e-01 1.08444595e+00 -5.62541664e-01 -9.61415529e-01 -1.01748288e+00 3.04145873e-01 8.41459155e-01 6.78960264e-01 -2.59865582e-01 -5.01028895e-01 -4.67072755e-01 5.84914267e-01 3.10808569e-01 -2.37932727e-01 6.14847004e-01 -4.50932592e-01 -4.40532386e-01 -2.07384869e-01 -8.68893638e-02 3.05776689e-02 -7.75706291e-01 -6.97993696e-01 7.42680192e-01 -1.32611409e-01 -1.37190270e+00 -8.81586850e-01 -2.44948819e-01 -2.84114361e-01 -1.44571376e+00 -5.55877686e-01 -1.15390515e+00 1.18658423e+00 9.34503853e-01 9.91502762e-01 4.10865456e-01 -3.53829533e-01 4.06922340e-01 -4.39182788e-01 -8.35677013e-02 -2.23292828e-01 1.16151117e-03 8.03548004e-03 -5.44537067e-01 -1.26800120e-01 -3.39452773e-01 -4.85875905e-01 6.02003634e-01 -7.36731410e-01 -7.30594471e-02 -5.80321252e-02 1.34718791e-01 3.99439782e-01 6.81268215e-01 9.30036128e-01 -7.09130287e-01 8.78852546e-01 -8.64277124e-01 -1.06418848e+00 5.15987456e-01 -5.10544360e-01 -3.69534224e-01 6.81679726e-01 1.81331381e-01 -5.30251741e-01 9.09042060e-02 5.00616848e-01 8.88782218e-02 3.27724129e-01 5.63476443e-01 2.03240570e-02 -4.69971418e-01 -8.34815875e-02 -4.41324174e-01 1.57155946e-01 9.03675184e-02 2.99396425e-01 2.83588827e-01 2.91387588e-01 -6.37660325e-01 6.30565464e-01 4.84899879e-01 6.89375639e-01 -1.91622853e-01 -3.17433864e-01 -3.50429386e-01 -3.56671274e-01 -5.43163598e-01 4.26246911e-01 -5.50145268e-01 -1.69476700e+00 3.38790953e-01 -1.61115503e+00 -5.31174719e-01 1.52974159e-01 8.62258077e-02 -6.88571334e-01 3.96577865e-01 -7.47925580e-01 -8.64027023e-01 -4.55009863e-02 -1.05649066e+00 2.35986486e-01 1.51205257e-01 1.56265557e-01 -1.28165865e+00 3.34784567e-01 1.17027365e-01 2.88401306e-01 7.57063329e-01 8.66321981e-01 -3.55587721e-01 -1.10464168e+00 9.92407575e-02 -6.17291927e-01 -5.71853101e-01 6.99924976e-02 9.26731676e-02 2.50100076e-01 -8.79580438e-01 -7.93340623e-01 2.26026401e-02 1.71568096e-01 5.96014977e-01 2.16456473e-01 -9.21428442e-01 -9.28011537e-01 1.87817350e-01 2.13604808e+00 5.94651937e-01 5.19265234e-02 6.93728983e-01 1.56776190e-01 6.76135659e-01 9.83360291e-01 6.76461220e-01 9.54927862e-01 6.09325767e-01 9.65113342e-01 5.64053319e-02 5.04694462e-01 3.70928586e-01 2.04017639e-01 1.45189032e-01 2.73619760e-02 -1.00549722e+00 -7.45823801e-01 1.03690624e+00 -2.46126366e+00 -8.22117627e-01 -6.63829386e-01 1.81411409e+00 1.68159395e-01 4.44809021e-03 8.01972806e-01 -5.77045567e-02 8.34023893e-01 -2.09543377e-01 -3.73592138e-01 -1.23852098e+00 1.52206033e-01 -4.85839784e-01 9.31464314e-01 1.13327026e+00 -6.79212809e-01 5.07778227e-01 7.38536453e+00 5.75753860e-02 -2.37239003e-01 5.02142794e-02 5.74583560e-02 -3.16460073e-01 -4.40363646e-01 1.68975428e-01 -5.38303494e-01 3.26690227e-01 9.04806554e-01 -1.26081479e+00 1.07162535e+00 6.02737308e-01 5.84671497e-01 -4.22604918e-01 -1.19590795e+00 3.40808392e-01 -3.63944560e-01 -1.49135613e+00 -3.42858315e-01 4.95143473e-01 1.00507951e+00 -3.45400840e-01 -4.40294385e-01 -1.93019792e-01 7.44220614e-01 -7.47412801e-01 6.85023844e-01 -6.98438510e-02 5.74002683e-01 -1.20718861e+00 3.16302747e-01 2.91186988e-01 -1.30302441e+00 -2.64475584e-01 -3.86498690e-01 -4.83356982e-01 1.09581518e+00 6.15455151e-01 -7.86653340e-01 1.10234189e+00 3.29214334e-01 3.85102719e-01 7.05950320e-01 1.57695913e+00 1.90221444e-01 -5.77985346e-01 -4.39083934e-01 2.75543898e-01 5.84202886e-01 -3.06530654e-01 9.93032396e-01 1.07447910e+00 7.13138878e-02 4.73424822e-01 1.01098871e+00 6.84356928e-01 1.21460008e-02 -2.28968859e-01 -3.99921715e-01 5.13217784e-02 7.28314936e-01 1.41499960e+00 -1.14832544e+00 5.43267466e-02 -4.75640446e-01 6.52429521e-01 3.95941585e-01 3.96063089e-01 -7.23505914e-01 -8.00574780e-01 1.18072581e+00 -6.41738921e-02 -8.89042914e-02 -5.52530348e-01 -4.80921566e-02 -4.49579954e-01 -6.58826977e-02 -2.11910799e-01 6.87673986e-01 -6.57329142e-01 -1.16378844e+00 6.28623843e-01 1.85099378e-01 -8.43931079e-01 -1.75488770e-01 -4.58399147e-01 -8.64480197e-01 8.42986941e-01 -1.92615998e+00 -8.36209357e-01 4.62388135e-02 7.23019004e-01 2.90338546e-01 8.70304555e-02 5.67284346e-01 4.41791594e-01 -4.83429223e-01 2.04057097e-02 -6.10855892e-02 -3.39551896e-01 4.26493771e-02 -1.24867010e+00 7.52393156e-02 1.17265546e+00 -9.29704070e-01 1.95053220e-02 8.32579672e-01 -6.86433017e-01 -1.78630745e+00 -1.02941918e+00 9.81536210e-01 4.06028360e-01 7.78573930e-01 1.17265910e-01 -8.87727812e-02 1.11887884e+00 5.06123066e-01 1.86641999e-02 2.90505767e-01 -3.18585694e-01 3.23673099e-01 -1.76967204e-01 -1.67124951e+00 3.04863572e-01 1.03162563e+00 2.81552225e-01 2.28255130e-02 8.84547889e-01 1.03516352e+00 -4.87435102e-01 -8.63067150e-01 -1.48091584e-01 1.23616740e-01 -2.87938356e-01 7.74199426e-01 -7.40968645e-01 1.37765378e-01 -5.79307377e-01 9.07728374e-02 -1.73638797e+00 -7.76762187e-01 -1.08432376e+00 3.24298769e-01 9.61457610e-01 5.60043573e-01 -7.88989425e-01 6.89233065e-01 7.67384112e-01 -2.55040616e-01 -4.64780122e-01 -1.23541951e+00 -1.21872294e+00 -3.15064900e-02 1.45472482e-01 6.50215864e-01 8.91182780e-01 8.11666131e-01 7.48561621e-02 -4.58251953e-01 7.07081795e-01 1.15599513e+00 5.45067430e-01 5.14109492e-01 -1.16361284e+00 -7.10833222e-02 -3.15441102e-01 -1.26334690e-02 -8.65691006e-01 4.15066183e-01 -1.03713036e+00 4.97726724e-02 -2.56848598e+00 2.27519944e-01 -1.01225340e+00 3.90469909e-01 6.55740857e-01 4.91051853e-01 -1.91774800e-01 6.92170560e-01 3.11879087e-02 -9.97409761e-01 -8.85579735e-02 1.45768976e+00 -7.32988715e-02 -3.40643644e-01 1.48457572e-01 -6.47089958e-01 -4.86879200e-02 6.67571247e-01 -7.64113784e-01 -5.55493772e-01 -7.29201138e-01 5.68381965e-01 1.11448252e+00 1.25178592e-02 -9.22035724e-02 6.11006916e-01 -1.15082633e+00 -6.33744061e-01 -4.56687361e-01 -3.77786942e-02 -1.16041577e+00 4.88715649e-01 8.68374586e-01 -5.09795584e-02 5.77168822e-01 3.10051013e-02 5.30246198e-01 1.84518397e-01 -5.39089084e-01 2.54670441e-01 -1.24725528e-01 -9.15029943e-01 6.70162380e-01 -8.56920302e-01 -8.71796012e-02 2.01002908e+00 -9.76401791e-02 -1.01226795e+00 -4.29331601e-01 -8.36687624e-01 1.38352799e+00 3.37218255e-01 2.39147350e-01 5.26292920e-01 -1.16108084e+00 -9.48983014e-01 -4.15281028e-01 -3.14220339e-01 -2.48572812e-03 3.81188601e-01 7.29337931e-01 -9.50648606e-01 7.97831595e-01 -7.57777631e-01 3.01784594e-02 -1.31975782e+00 1.09064054e+00 2.05784842e-01 -5.26453078e-01 -4.59479541e-01 4.20405835e-01 -1.75303712e-01 -1.56606823e-01 -5.42209391e-03 1.90198809e-01 2.33649224e-01 -1.62626088e-01 5.67425370e-01 1.04896569e+00 -3.15021157e-01 -3.74226958e-01 -6.67521298e-01 1.69468328e-01 -1.78605184e-01 -2.89851159e-01 1.59009731e+00 -5.35466373e-01 -8.82629275e-01 -6.39437854e-01 9.86845136e-01 -3.46908897e-01 -9.53054309e-01 -5.90303577e-02 -9.20927003e-02 -8.72243702e-01 -2.06124082e-01 -6.66469276e-01 -1.36995399e+00 -1.92950562e-01 -4.60694015e-01 8.30610991e-01 1.09364128e+00 -4.67929356e-02 7.79185891e-01 7.85300136e-02 9.91783082e-01 -1.02377725e+00 -1.92382216e-01 5.49458563e-01 8.01376283e-01 -5.45996785e-01 1.36364117e-01 -1.24915111e+00 -3.97041440e-01 1.38924170e+00 7.11006463e-01 -2.75345385e-01 6.63855910e-01 5.43724120e-01 -5.40122926e-01 -1.45021632e-01 -1.00808728e+00 -3.00140470e-01 -7.35685945e-01 8.67352426e-01 -5.83192825e-01 4.45366174e-01 -6.48966491e-01 1.57435268e-01 2.69226223e-01 1.91070229e-01 1.53022027e+00 1.19352305e+00 -7.19814599e-01 -1.47691226e+00 -3.38912070e-01 1.46129742e-01 -1.18234135e-01 4.59662050e-01 -1.22403666e-01 6.47392571e-01 -1.06971100e-01 1.46559751e+00 4.32371795e-01 3.64725627e-02 2.87707090e-01 -8.26325417e-01 6.06034458e-01 -5.71023405e-01 -2.94526011e-01 -3.92634094e-01 5.74319541e-01 -3.18913937e-01 -5.72328687e-01 -5.19070625e-01 -1.55317593e+00 -1.37835693e+00 -1.09875299e-01 2.18548819e-01 3.11232179e-01 8.46408606e-01 8.41044411e-02 4.85843420e-01 1.16796303e+00 -5.57953179e-01 3.99836376e-02 -8.54098871e-02 -8.79613698e-01 -3.57290924e-01 5.45744717e-01 -4.57642317e-01 -1.82074934e-01 -4.99989092e-01]	[4.977317810058594, 1.8222259283065796]
518666c7-e8be-484e-b56d-653a98faffea	online-and-offline-handwritten-chinese	1606.05763	null	http://arxiv.org/abs/1606.05763v1	http://arxiv.org/pdf/1606.05763v1.pdf	Online and Offline Handwritten Chinese Character Recognition: A Comprehensive Study and New Benchmark	Recent deep learning based methods have achieved the state-of-the-art performance for handwritten Chinese character recognition (HCCR) by learning discriminative representations directly from raw data. Nevertheless, we believe that the long-and-well investigated domain-specific knowledge should still help to boost the performance of HCCR. By integrating the traditional normalization-cooperated direction-decomposed feature map (directMap) with the deep convolutional neural network (convNet), we are able to obtain new highest accuracies for both online and offline HCCR on the ICDAR-2013 competition database. With this new framework, we can eliminate the needs for data augmentation and model ensemble, which are widely used in other systems to achieve their best results. This makes our framework to be efficient and effective for both training and testing. Furthermore, although directMap+convNet can achieve the best results and surpass human-level performance, we show that writer adaptation in this case is still effective. A new adaptation layer is proposed to reduce the mismatch between training and test data on a particular source layer. The adaptation process can be efficiently and effectively implemented in an unsupervised manner. By adding the adaptation layer into the pre-trained convNet, it can adapt to the new handwriting styles of particular writers, and the recognition accuracy can be further improved consistently and significantly. This paper gives an overview and comparison of recent deep learning based approaches for HCCR, and also sets new benchmarks for both online and offline HCCR.	['Cheng-Lin Liu', 'Xu-Yao Zhang', 'Yoshua Bengio']	2016-06-18	null	null	null	null	['offline-handwritten-chinese-character', 'offline-handwritten-chinese-character']	['computer-vision', 'natural-language-processing']	[-2.77073495e-02 -4.83661681e-01 6.30026981e-02 -6.21979833e-01 -5.64553916e-01 -4.42948043e-01 6.59698427e-01 -1.67247444e-01 -7.27442503e-01 5.53078413e-01 -5.59156835e-02 -6.87476844e-02 4.50640917e-02 -7.16501176e-01 -5.29355109e-01 -7.82556117e-01 3.18651736e-01 5.73580444e-01 2.87259132e-01 -3.74302298e-01 2.97014564e-01 1.11143541e+00 -1.31454146e+00 4.31482702e-01 8.64553332e-01 9.28743899e-01 3.79803389e-01 7.12954819e-01 -4.36187647e-02 8.87628317e-01 -9.14403558e-01 -5.70928395e-01 1.01582065e-01 -3.58247250e-01 -6.44001365e-01 1.80113852e-01 3.13599050e-01 -3.89122337e-01 -5.96814692e-01 6.69346154e-01 6.99750304e-01 6.74837828e-02 6.22329533e-01 -5.25509894e-01 -9.08544362e-01 5.26043832e-01 -2.83382952e-01 -1.98963836e-01 -2.34806091e-02 7.33534917e-02 4.28435832e-01 -1.20304501e+00 6.43849611e-01 6.85289204e-01 6.44841790e-01 8.65858495e-01 -8.53606939e-01 -4.19198811e-01 1.87812641e-01 5.71292996e-01 -1.27717280e+00 -3.05349410e-01 8.15529585e-01 -1.38505444e-01 9.01445925e-01 2.14314178e-01 5.48700809e-01 1.24925160e+00 -1.53950021e-01 1.24612439e+00 1.10380888e+00 -5.78245163e-01 9.65901241e-02 1.17148273e-01 2.78791696e-01 4.69036937e-01 -1.47573873e-02 -1.41373217e-01 -4.06597614e-01 4.45789635e-01 7.36780882e-01 4.44309413e-02 -3.54137212e-01 -7.44087202e-03 -1.00251734e+00 5.42068005e-01 3.70336682e-01 8.07195187e-01 -3.84541661e-01 -6.62475452e-02 5.79494894e-01 1.88352212e-01 1.88810050e-01 6.30074680e-01 -5.61996639e-01 -3.91117215e-01 -1.13723719e+00 9.53862146e-02 7.65160322e-01 9.58230495e-01 5.19231498e-01 3.45876455e-01 -3.89248788e-01 1.22048771e+00 -1.43840313e-01 4.73021328e-01 8.05904508e-01 -4.18396205e-01 2.77230531e-01 6.15927756e-01 -2.37468451e-01 -8.87002468e-01 -3.25402051e-01 -7.09779501e-01 -1.16865218e+00 8.74278098e-02 4.93831635e-01 -6.23654909e-02 -1.23861432e+00 1.15289855e+00 -3.63733321e-01 -1.12287655e-01 1.58074468e-01 9.35690701e-01 6.35378361e-01 5.98964691e-01 -3.12818646e-01 5.28616682e-02 1.02511406e+00 -1.14241600e+00 -8.23980868e-01 -1.42541230e-01 4.75060403e-01 -8.38378549e-01 1.08028817e+00 7.42108047e-01 -7.41948485e-01 -7.53653109e-01 -1.32375884e+00 -1.18390083e-01 -6.19386375e-01 8.00226688e-01 5.53593218e-01 6.71787024e-01 -8.29877615e-01 7.15410531e-01 -8.13810229e-01 -2.76373088e-01 5.44721067e-01 3.82384628e-01 -6.26040697e-01 -4.77346808e-01 -1.00780463e+00 1.05679631e+00 5.08014798e-01 6.46049321e-01 -7.94248641e-01 -3.14588904e-01 -3.81493092e-01 1.00626690e-04 3.22419405e-01 1.38785437e-01 9.48948324e-01 -1.01589692e+00 -1.97874212e+00 5.74563086e-01 1.20741770e-01 -5.02106965e-01 8.20017278e-01 -3.05459768e-01 -5.51147759e-01 7.13261664e-02 -5.59855461e-01 3.27836841e-01 7.16348886e-01 -1.08824980e+00 -4.08562005e-01 -2.11623818e-01 -4.54058558e-01 -1.21434093e-01 -8.68681729e-01 -3.70218456e-02 -1.01054490e+00 -7.63138473e-01 -4.71697282e-03 -7.37217665e-01 -2.86628902e-02 -2.67926395e-01 -2.52529532e-01 -2.65019774e-01 9.73067284e-01 -1.00170660e+00 1.08504474e+00 -2.09917998e+00 7.70298988e-02 2.18702301e-01 -1.08018667e-01 1.08766055e+00 -3.30883026e-01 3.86863738e-01 2.47344717e-01 -1.82211339e-01 -1.65676177e-01 -4.16784048e-01 -2.07102254e-01 2.25118771e-01 -2.33203411e-01 4.30230021e-01 4.03553098e-01 9.28316176e-01 -6.05247140e-01 -7.37196431e-02 3.60965073e-01 5.78689039e-01 -2.68204868e-01 3.59604716e-01 -8.21008533e-02 3.01988751e-01 -1.49474427e-01 5.80034673e-01 8.12130809e-01 -9.13138017e-02 4.32555437e-01 -1.14099756e-01 -2.06061780e-01 -5.04483320e-02 -1.04786396e+00 1.51678514e+00 -4.87192959e-01 9.48813260e-01 -3.89481723e-01 -1.17871201e+00 1.67717946e+00 1.47086203e-01 7.33370930e-02 -8.42969477e-01 2.38181539e-02 6.19863510e-01 1.60529450e-01 -2.32740179e-01 6.02141082e-01 2.93764025e-01 1.29045248e-01 2.28279289e-02 2.98350126e-01 6.03526607e-02 2.15591401e-01 -1.89650282e-01 9.34579372e-01 1.36789784e-01 -9.28752124e-03 -2.13472500e-01 9.99619603e-01 -7.04719126e-02 4.69782054e-01 9.10744846e-01 5.55101633e-02 9.53073442e-01 3.11512738e-01 -8.25958908e-01 -1.11670971e+00 -5.63954473e-01 -1.30872652e-01 8.90983939e-01 -2.05826133e-01 -1.27105117e-01 -6.74231768e-01 -7.64346778e-01 -1.33125216e-01 4.81861144e-01 -6.14605010e-01 -8.08092281e-02 -1.04643190e+00 -8.46416652e-01 1.00167215e+00 9.76864815e-01 9.65053678e-01 -1.03341925e+00 -1.99727044e-01 4.78119165e-01 1.66968510e-01 -1.17945218e+00 -1.88872859e-01 4.41208661e-01 -7.34151900e-01 -7.89288580e-01 -1.33094680e+00 -8.73412669e-01 6.44178987e-01 -2.07597524e-01 6.87983692e-01 2.09934369e-01 -2.65999019e-01 5.85305877e-02 -7.99479604e-01 -2.44251132e-01 -5.98092139e-01 5.28906584e-01 -1.48129344e-01 2.38957152e-01 4.56246674e-01 -5.07923141e-02 -2.85666138e-01 3.81774306e-01 -7.82857955e-01 -3.23819630e-02 7.67696857e-01 1.31129313e+00 3.29282522e-01 -1.69845432e-01 6.69233739e-01 -7.25747406e-01 5.98149598e-01 2.18351468e-01 -6.56879246e-01 5.78760624e-01 -7.78779626e-01 1.08365454e-01 1.07020116e+00 -6.32721186e-01 -1.19085383e+00 1.95361972e-01 -5.28666258e-01 -3.39720666e-01 -2.36992940e-01 6.11342967e-01 -1.98790789e-01 -2.79346198e-01 6.47625923e-01 7.04533041e-01 -7.14601576e-02 -8.73003423e-01 1.53302357e-01 1.01025414e+00 7.76547194e-01 -5.63703001e-01 5.26088119e-01 2.02908933e-01 -2.07144752e-01 -8.05474818e-01 -4.46141213e-01 -1.69979885e-01 -1.01306117e+00 -2.70592481e-01 6.84381306e-01 -6.74870610e-01 -6.16677880e-01 1.28227878e+00 -1.14048564e+00 -4.90861565e-01 -1.29004404e-01 4.18613374e-01 -1.67892665e-01 3.97706836e-01 -7.80739009e-01 -6.48910701e-01 -2.17976600e-01 -1.23200917e+00 6.49018288e-01 3.80278438e-01 3.79490286e-01 -1.14318943e+00 -1.37523755e-01 4.11853045e-02 7.50430822e-01 3.30140591e-02 6.91505671e-01 -1.15759087e+00 -4.87105757e-01 -6.53978288e-01 -2.86753386e-01 1.09741843e+00 1.21907011e-01 1.43356755e-01 -1.16301858e+00 -3.57996732e-01 -2.22762600e-01 -4.31178689e-01 1.15868354e+00 -1.15848608e-01 1.44351971e+00 4.12302315e-02 -5.22914939e-02 6.38423502e-01 1.42289138e+00 3.21863472e-01 1.03144073e+00 4.42174524e-01 7.34538198e-01 2.98661768e-01 4.39286917e-01 4.08438176e-01 7.52357841e-02 7.93746293e-01 3.68766896e-02 -2.66849369e-01 -3.64401907e-01 -6.92082718e-02 2.96128333e-01 1.03030622e+00 -6.33383989e-01 -2.30158046e-01 -1.10939515e+00 3.50161999e-01 -1.79271722e+00 -7.43199944e-01 -7.68648759e-02 2.22576094e+00 8.92110884e-01 -1.48713095e-02 -1.46617547e-01 2.71183103e-01 5.88793933e-01 3.73701490e-02 -4.04365063e-01 -5.41385591e-01 -5.63011885e-01 4.22551304e-01 6.14457667e-01 1.88922942e-01 -1.15017486e+00 1.09671915e+00 6.11772251e+00 1.04873705e+00 -1.61604905e+00 -3.35701592e-02 6.57585621e-01 3.64134252e-01 1.73385546e-01 -3.02888602e-01 -1.02149236e+00 2.46460959e-01 7.67128646e-01 1.87111795e-01 5.34474075e-01 8.35187078e-01 -1.46895662e-01 2.29526386e-01 -1.14103854e+00 9.67186511e-01 4.54128116e-01 -1.61817491e+00 -2.22598631e-02 6.98491558e-02 8.12305748e-01 -6.44088313e-02 7.72065222e-02 4.81768459e-01 -1.20200902e-01 -1.12246406e+00 5.14401197e-01 6.90209508e-01 8.47846985e-01 -7.33596146e-01 1.26331592e+00 3.31242204e-01 -8.92632186e-01 -2.27430061e-01 -5.96898437e-01 1.60763979e-01 -2.23840624e-01 4.99606669e-01 -8.84438455e-01 7.84298480e-01 3.78840834e-01 8.79068911e-01 -9.62201178e-01 1.01337147e+00 -3.48006338e-01 5.73182821e-01 -4.08259109e-02 -4.27041709e-01 3.24303180e-01 5.11549152e-02 1.45880058e-01 1.48022103e+00 1.83416367e-01 -2.03557268e-01 -1.29542068e-01 6.63159907e-01 -2.20660001e-01 1.39029145e-01 -1.53102994e-01 -2.59635329e-01 2.50274092e-01 1.14514959e+00 -6.28588259e-01 -4.85812396e-01 -3.68033677e-01 1.25488687e+00 4.99413580e-01 3.41468811e-01 -6.21004522e-01 -7.07483649e-01 2.73619264e-01 -2.70747066e-01 5.75859249e-01 -3.36384177e-01 -3.94317001e-01 -1.34936571e+00 1.17925882e-01 -1.17023659e+00 7.24485889e-02 -3.31142247e-01 -1.30809355e+00 9.13174510e-01 -4.06684220e-01 -1.19429660e+00 -2.65409261e-01 -1.22518659e+00 -4.55996931e-01 8.70965302e-01 -1.44614089e+00 -1.23889494e+00 -3.35131466e-01 6.21293604e-01 6.21311188e-01 -7.34956145e-01 9.91246879e-01 4.37449962e-01 -6.74877882e-01 1.09507692e+00 5.70452452e-01 7.14005709e-01 8.27777982e-01 -1.27454638e+00 2.92973220e-01 1.02534640e+00 1.82902917e-01 5.36851168e-01 3.07730407e-01 -4.10497695e-01 -1.52627456e+00 -9.92565215e-01 7.57980883e-01 -1.78603709e-01 3.65641534e-01 -5.66009521e-01 -1.26227462e+00 2.52448291e-01 5.26467897e-02 1.27366647e-01 4.63034481e-01 1.72691479e-01 -5.16555011e-01 -4.65531379e-01 -8.60163629e-01 4.58767563e-01 6.80153191e-01 -4.22278792e-01 -4.57197517e-01 2.38234803e-01 1.23747312e-01 -5.00403404e-01 -8.95510495e-01 4.63119417e-01 7.03432560e-01 -8.66224110e-01 5.71974158e-01 -5.47995269e-01 4.73672718e-01 -1.46635890e-01 -1.28267705e-01 -1.12728643e+00 -3.03923726e-01 -2.25843519e-01 -8.05647746e-02 1.35206342e+00 4.56655800e-01 -5.71571946e-01 6.80466056e-01 5.04909217e-01 -2.66539454e-01 -8.52681220e-01 -8.33775699e-01 -9.99020338e-01 3.92031878e-01 -3.26782286e-01 6.41706347e-01 8.02226186e-01 -2.50761956e-01 -1.00288294e-01 -6.87823176e-01 -2.72937100e-02 2.48227149e-01 7.35767111e-02 7.88239121e-01 -1.04696321e+00 -2.75466442e-01 -4.30459112e-01 -6.68472111e-01 -1.04622006e+00 1.22437753e-01 -7.53860235e-01 1.87670961e-01 -1.34773552e+00 1.21703625e-01 -3.71488333e-01 -4.30693150e-01 6.67141020e-01 -8.40502530e-02 3.91682833e-01 5.32955706e-01 2.22721711e-01 -3.24951679e-01 5.17445087e-01 1.27849555e+00 -1.52698502e-01 -1.99468568e-01 -3.93842980e-02 -2.79251516e-01 4.80613023e-01 7.72922397e-01 -1.13271140e-01 1.47652835e-01 -5.89502394e-01 -2.52901435e-01 -3.68371099e-01 1.21988058e-01 -1.06960225e+00 4.89378333e-01 2.37401992e-01 9.20393527e-01 -4.67652291e-01 1.56931072e-01 -7.28447199e-01 -2.52350271e-01 4.51957822e-01 -3.88892233e-01 -2.57763714e-01 2.76317954e-01 2.02111989e-01 -4.94139135e-01 -3.12916666e-01 9.36053038e-01 5.22388034e-02 -9.53503132e-01 1.08901128e-01 -5.60841143e-01 -4.38825011e-01 6.95463359e-01 -1.92266136e-01 -4.17924047e-01 -1.93772569e-01 -5.78294873e-01 -1.61200806e-01 2.90440142e-01 5.62443972e-01 7.54189253e-01 -1.22418547e+00 -8.68366897e-01 4.19342309e-01 9.25191641e-02 -2.80330718e-01 3.52999300e-01 7.39609540e-01 -7.61272609e-01 5.23167014e-01 -5.36844492e-01 -5.07525623e-01 -1.09443271e+00 4.40355659e-01 5.11583090e-01 -4.65768784e-01 -6.04129136e-01 6.88289881e-01 -4.33452576e-01 -5.09975135e-01 4.08446640e-01 -1.36668116e-01 -3.35684627e-01 -1.40406549e-01 7.16963589e-01 4.57576156e-01 4.65754539e-01 -4.20427054e-01 -4.64347929e-01 4.90274698e-01 -5.42825997e-01 7.83285312e-03 1.52230561e+00 4.52792138e-01 7.76638687e-02 2.14628175e-01 1.18366158e+00 -4.41321395e-02 -1.25748122e+00 -7.43246078e-02 6.91465512e-02 -3.85313690e-01 -1.19337410e-01 -1.17478192e+00 -1.32276440e+00 1.13838398e+00 6.67888582e-01 -1.76763564e-01 1.40033376e+00 -2.93335170e-01 4.71222550e-01 8.12107205e-01 2.80330300e-01 -1.46675396e+00 2.08964050e-01 7.91630685e-01 1.05567133e+00 -1.17456365e+00 -9.05610770e-02 -2.19362918e-02 -6.84980631e-01 1.78642035e+00 7.20677197e-01 -3.23366672e-01 4.49991971e-01 4.76243287e-01 2.45313600e-01 3.05292189e-01 -5.26719570e-01 -6.62476197e-02 4.03072804e-01 6.34953916e-01 5.66081107e-01 2.49420200e-02 -3.89837921e-01 8.01796854e-01 2.50437617e-01 1.07306339e-01 6.19262874e-01 7.82735586e-01 -1.80529103e-01 -1.61600661e+00 -2.20089048e-01 3.97675782e-01 -3.91583443e-01 1.22237146e-01 -4.31732267e-01 9.61403787e-01 2.14472204e-03 5.92762947e-01 -9.51817445e-03 -6.66561306e-01 4.87723082e-01 3.10447197e-02 5.50427914e-01 -3.13868105e-01 -7.46776640e-01 1.24335457e-02 -1.46383986e-01 -3.22322756e-01 -2.50448316e-01 -4.73173261e-01 -9.42557931e-01 -1.04367942e-01 -3.97183985e-01 -1.28297076e-01 7.70714521e-01 1.02888143e+00 2.67092615e-01 5.22470057e-01 6.36127412e-01 -6.79118276e-01 -8.76475513e-01 -1.13995588e+00 -5.62522888e-01 2.70976841e-01 1.49738029e-01 -2.88792819e-01 -6.51475638e-02 1.12042874e-01]	[11.823775291442871, 2.591913938522339]
aa3216ba-3f78-4f51-a3e5-89dafedb231d	a-comparison-of-data-augmentation-techniques	2003.13502	null	https://arxiv.org/abs/2003.13502v2	https://arxiv.org/pdf/2003.13502v2.pdf	An Open-source Tool for Hyperspectral Image Augmentation in Tensorflow	Satellite imagery allows a plethora of applications ranging from weather forecasting to land surveying. The rapid development of computer vision systems could open new horizons to the utilization of satellite data due to the abundance of large volumes of data. However, current state-of-the-art computer vision systems mainly cater to applications that mainly involve natural images. While useful, those images exhibit a different distribution from satellite images in addition to having more spectral channels. This allows the use of pretrained deep learning models only in a subset of spectral channels that are equivalent to natural images thus discarding valuable information from other spectral channels. This calls for research effort to optimize deep learning models for satellite imagery to enable the assessment of their utility in the domain of remote sensing. Tensorflow tool allows for rapid prototyping and testing of deep learning models, however, its built-in image generator is designed to handle a maximum of four spectral channels. This manuscript introduces an open-source tool that allows the implementation of image augmentation for hyperspectral images in Tensorflow. Given how accessible and easy-to-use Tensorflow is, this tool would provide many researchers with the means to implement, test, and deploy deep learning models for remote sensing applications.	['Mohamed Abdelhack']	2020-03-30	null	null	null	null	['satellite-image-classification']	['computer-vision']	[ 2.29550168e-01 -2.76543438e-01 6.79909736e-02 -4.31917161e-01 -1.60126597e-01 -4.82638448e-01 4.22300667e-01 8.57309066e-03 -4.93013293e-01 4.77174163e-01 -1.77078187e-01 -6.42747819e-01 -1.38681531e-01 -1.04189932e+00 -3.38020563e-01 -8.61503720e-01 -5.46164811e-01 2.63313174e-01 -1.38642922e-01 -5.09610116e-01 -2.03461960e-01 9.12325919e-01 -1.85925984e+00 2.03816086e-01 7.33060718e-01 7.37043917e-01 5.23187816e-01 5.89892566e-01 -2.25692779e-01 2.92079538e-01 -2.42733702e-01 9.76236388e-02 6.28302515e-01 -3.65217999e-02 -6.85986757e-01 4.09283459e-01 6.25262082e-01 -5.83310187e-01 -1.68226942e-01 1.07979000e+00 4.50539947e-01 1.85688078e-01 2.68037975e-01 -1.05158424e+00 -4.91412580e-01 4.61057484e-01 -4.09674019e-01 2.55265564e-01 -3.35775137e-01 3.91592056e-01 9.71692979e-01 -5.32434523e-01 3.72832060e-01 9.45703566e-01 5.84793150e-01 2.28054859e-02 -1.18375647e+00 -3.56615841e-01 -5.08838855e-02 3.02181035e-01 -1.26260173e+00 -7.78325871e-02 5.80859482e-01 -6.31755888e-01 1.04843974e+00 4.70982373e-01 7.88089514e-01 6.24518394e-01 -2.20651627e-01 4.39336985e-01 1.37833047e+00 -2.37366423e-01 1.49889767e-01 1.60148874e-01 2.05818519e-01 3.91931325e-01 2.20054284e-01 2.91484475e-01 5.63456416e-02 1.50143638e-01 6.94439888e-01 1.32958993e-01 -3.13548028e-01 -1.32301584e-01 -1.14018619e+00 9.63373482e-01 7.26832092e-01 4.51825738e-01 -6.95713282e-01 -3.21910948e-01 3.98370296e-01 4.83140439e-01 4.98612165e-01 4.72826749e-01 -6.38121068e-01 2.37980768e-01 -1.07961309e+00 1.54013351e-01 4.82769489e-01 2.92347670e-01 1.13243747e+00 4.25641090e-01 9.25217792e-02 9.66111779e-01 2.72325397e-01 5.14924228e-01 5.75597823e-01 -7.57160842e-01 7.46160299e-02 8.03765595e-01 -1.57272264e-01 -7.91034758e-01 -6.21960342e-01 -7.30522990e-01 -1.01499975e+00 4.54721481e-01 2.60731876e-01 -4.86399233e-02 -9.90913212e-01 1.08443654e+00 1.85442209e-01 -1.69628590e-01 5.88314086e-02 1.20218551e+00 7.40977228e-01 8.11025500e-01 1.39348172e-02 7.87140504e-02 1.13562822e+00 -7.80126154e-01 -3.02327454e-01 -3.90023351e-01 6.46913111e-01 -5.74155092e-01 1.36037481e+00 3.30841660e-01 -2.44549021e-01 -5.51336765e-01 -9.85891163e-01 8.54597613e-02 -7.25157261e-01 2.91887045e-01 9.39524770e-01 6.86971843e-01 -9.00839388e-01 6.29331827e-01 -6.77357674e-01 -6.42706811e-01 4.54349279e-01 3.24099779e-01 -5.08634090e-01 -3.61728035e-02 -9.53453064e-01 9.96871769e-01 7.23467588e-01 4.75145221e-01 -6.38230741e-01 -5.62892795e-01 -7.56368697e-01 2.34450161e-01 -5.41373007e-02 -7.39668161e-02 9.06681955e-01 -1.44949675e+00 -1.08126497e+00 9.09810841e-01 4.23073441e-01 -5.52307963e-01 2.89361566e-01 1.52238280e-01 -4.01501298e-01 4.37282957e-02 1.37236029e-01 7.46989787e-01 7.10977077e-01 -8.80141556e-01 -5.25966525e-01 -3.84050786e-01 4.17171493e-02 9.91662219e-02 -7.05091298e-01 -1.21375732e-01 -5.84962107e-02 -4.14275974e-01 1.33961484e-01 -1.01661742e+00 -2.79222786e-01 1.73217326e-01 -1.21636607e-01 3.51713061e-01 1.11066008e+00 -7.37607598e-01 7.44448960e-01 -2.17077494e+00 -1.59555972e-01 1.76140711e-01 -2.46821091e-01 6.83449924e-01 -5.14032781e-01 4.14722472e-01 -5.22839248e-01 3.12386394e-01 -3.97137463e-01 1.39860407e-01 -1.94201693e-01 4.27625030e-01 -1.56217143e-01 5.89780450e-01 1.65086851e-01 4.24510300e-01 -6.57605529e-01 -7.54648447e-02 7.52355814e-01 5.00911593e-01 -9.20303613e-02 -3.93517613e-02 -2.61936218e-01 5.12469113e-01 -9.35194269e-02 6.97899342e-01 1.00811374e+00 5.40170893e-02 5.51436469e-03 -2.12066531e-01 -5.49472094e-01 -6.17644638e-02 -1.13023615e+00 1.37368166e+00 -2.91397542e-01 9.92863357e-01 1.20357350e-01 -1.16613233e+00 9.22982514e-01 6.91136792e-02 4.24083352e-01 -5.50173938e-01 -2.72400361e-02 9.91838276e-02 2.37991080e-01 -6.69594228e-01 5.30746639e-01 -1.69784710e-01 5.96001863e-01 3.25722814e-01 -3.24939899e-02 -3.08569759e-01 3.68158519e-01 -3.37417573e-01 4.37995344e-01 3.99658754e-02 9.96863022e-02 -1.87422663e-01 5.33069730e-01 2.43451640e-01 1.32230967e-01 5.33889055e-01 -9.42699015e-02 3.32794428e-01 4.95082289e-02 -9.84082162e-01 -1.21456277e+00 -5.47461808e-01 -4.96579945e-01 1.11477864e+00 -4.87064362e-01 1.39121130e-01 -4.14070666e-01 -2.38964930e-01 3.99627462e-02 5.36698520e-01 -4.82172698e-01 2.26823866e-01 1.67812690e-01 -1.11748183e+00 4.48850781e-01 3.58669870e-02 7.94642270e-01 -1.18830156e+00 -6.30643725e-01 1.27242073e-01 -4.67919111e-02 -9.57379699e-01 2.70532727e-01 1.71094880e-01 -1.31398726e+00 -9.77937818e-01 -6.27763927e-01 -4.25548553e-01 4.72288549e-01 6.75769925e-01 7.64776349e-01 1.37753338e-01 -5.04268050e-01 1.17242999e-01 -5.10857463e-01 -5.23872852e-01 -3.84102046e-01 2.69461155e-01 -2.95117587e-01 1.15865454e-01 4.04456258e-01 -5.24807751e-01 -4.23275918e-01 5.96550219e-02 -1.46634293e+00 2.41371221e-03 7.27869630e-01 8.20717692e-01 4.63346064e-01 4.17088479e-01 -5.55594265e-02 -4.67313796e-01 3.52308720e-01 -5.12565434e-01 -8.12879920e-01 4.56275791e-02 -5.35081625e-01 2.48008948e-02 5.08717775e-01 -1.03568085e-01 -7.61548936e-01 3.11544925e-01 -7.89815262e-02 -6.20197095e-02 -5.18207610e-01 1.23968077e+00 2.23405421e-01 -1.95238546e-01 9.03160691e-01 2.55749077e-01 1.85489774e-01 -5.79903483e-01 1.76426664e-01 9.31416154e-01 1.66713834e-01 -1.14243545e-01 6.72114372e-01 5.86411417e-01 6.56419946e-03 -1.66790771e+00 -6.74153328e-01 -6.53738737e-01 -5.97342610e-01 -2.67853409e-01 5.08065045e-01 -9.18704808e-01 -2.52686650e-01 6.44922316e-01 -7.49320686e-01 -3.19291294e-01 -1.47332489e-01 5.53785443e-01 1.23153165e-01 4.35997128e-01 -6.35787621e-02 -7.59360135e-01 -4.11924064e-01 -1.14462769e+00 7.66547561e-01 2.47225732e-01 1.32717103e-01 -8.90968442e-01 3.12891714e-02 3.91540706e-01 5.49749196e-01 1.10077493e-01 6.73630953e-01 -4.15452003e-01 -5.94633520e-01 -3.79711181e-01 -4.07640576e-01 9.12357807e-01 2.20345020e-01 4.09511685e-01 -1.25768816e+00 -3.38764101e-01 -2.75216699e-01 -3.03019702e-01 1.05178452e+00 6.02083147e-01 1.20912564e+00 -8.40072110e-02 9.10349265e-02 9.16522861e-01 1.64604330e+00 -2.94937659e-02 8.05459261e-01 9.13801074e-01 5.91294050e-01 8.64296377e-01 3.88585955e-01 4.01780814e-01 2.07327493e-02 4.31717992e-01 1.00529850e+00 -5.76686919e-01 3.84426594e-01 4.14248377e-01 9.30051878e-02 9.85800326e-02 -3.58101726e-01 2.27307215e-01 -1.12048304e+00 5.87513983e-01 -1.53728259e+00 -1.20561206e+00 -5.64017415e-01 2.24084616e+00 4.54089701e-01 -2.88730532e-01 -1.59336448e-01 2.97960997e-01 5.10129154e-01 2.82990932e-01 -2.84982413e-01 -3.41189086e-01 -3.23763639e-01 3.55809242e-01 9.28056180e-01 3.34659755e-01 -1.52357626e+00 8.80838573e-01 6.02399778e+00 2.63759583e-01 -1.89548969e+00 -9.95670557e-02 4.66646165e-01 1.11542083e-01 -6.69589117e-02 2.68193036e-02 -4.40966785e-01 2.44108289e-01 9.24839735e-01 2.18086734e-01 4.69927251e-01 9.19677258e-01 7.33998477e-01 -2.54615873e-01 -5.51576376e-01 8.12375009e-01 -3.82950813e-01 -1.37429285e+00 1.67290896e-01 2.14348018e-01 6.54606998e-01 7.56277561e-01 5.23018558e-03 6.30425662e-02 -2.46815775e-02 -1.06436944e+00 3.90627086e-01 2.68837303e-01 5.41290641e-01 -6.07218862e-01 9.74071145e-01 3.30538094e-01 -7.87720621e-01 -2.74105012e-01 -6.22601330e-01 -3.60023260e-01 -4.28859293e-01 7.67656386e-01 -1.01250553e+00 3.71602565e-01 9.09477830e-01 6.29114568e-01 -6.63564265e-01 1.27948809e+00 -1.10598942e-02 6.67871654e-01 -4.11202192e-01 2.79248506e-01 4.63959217e-01 -5.37991107e-01 3.61962885e-01 1.27995682e+00 4.43868726e-01 -1.34898752e-01 2.67175585e-01 5.41094422e-01 3.38909060e-01 1.92041576e-01 -6.39439106e-01 -4.06712681e-01 1.86104685e-01 1.47413373e+00 -7.22305119e-01 -3.70613821e-02 -4.73719627e-01 7.95606852e-01 -2.09140673e-01 3.62631738e-01 -4.71522510e-01 -2.16779411e-01 8.52513313e-01 9.91082340e-02 2.21177146e-01 -5.29776752e-01 -3.30409765e-01 -9.70168114e-01 -1.41057000e-01 -1.07442451e+00 1.23599507e-01 -8.73306394e-01 -8.80781949e-01 7.03103185e-01 -2.40625329e-02 -1.29384983e+00 -1.96108576e-02 -9.48336363e-01 -4.89253342e-01 1.11980152e+00 -1.98738539e+00 -1.32374132e+00 -8.32497239e-01 4.34592098e-01 3.48976374e-01 -3.67362678e-01 1.11042738e+00 2.53879726e-01 -5.34536958e-01 -2.83639818e-01 3.78641218e-01 1.85332909e-01 4.62307453e-01 -1.04667139e+00 2.61412323e-01 9.81082261e-01 1.03313871e-01 2.24444866e-01 6.87658191e-01 -2.75347441e-01 -1.10467935e+00 -1.23229134e+00 5.69565296e-01 3.02891135e-01 7.80502737e-01 3.23088706e-01 -9.53353167e-01 5.91921568e-01 8.24036300e-02 2.82135084e-02 8.43776882e-01 7.46875182e-02 -2.48419568e-01 -5.82290411e-01 -8.71193230e-01 3.45455110e-01 7.49930516e-02 -4.98109251e-01 -2.26679429e-01 4.68336254e-01 -2.15047467e-02 -2.31037587e-02 -5.82324386e-01 1.95229411e-01 3.86974990e-01 -1.12934887e+00 8.57102811e-01 -3.83742452e-01 4.74085778e-01 -4.81545657e-01 -1.49274170e-01 -1.38270879e+00 -3.89627069e-01 -1.31489456e-01 5.68971395e-01 7.63263047e-01 5.22259772e-01 -6.47936881e-01 6.86395168e-01 3.84821504e-01 -1.84303492e-01 -6.34591579e-02 -6.20066762e-01 -6.37772441e-01 -1.58360660e-01 -4.93863493e-01 5.78857899e-01 1.16547287e+00 -6.35567784e-01 -2.59369966e-02 -1.99701548e-01 4.97888923e-01 4.50291306e-01 1.80107430e-01 6.30775511e-01 -1.67450154e+00 -1.95560902e-01 -5.09398460e-01 -5.04506111e-01 -4.13729161e-01 4.89307046e-02 -7.92099595e-01 -2.48624444e-01 -1.62681425e+00 -4.00835760e-02 -4.41390097e-01 -5.46962321e-02 9.62260842e-01 1.10720210e-01 5.27651072e-01 1.53569177e-01 3.56906503e-01 4.20724303e-01 2.82228589e-01 1.08423007e+00 -5.15245616e-01 -3.66072029e-01 7.41489828e-02 -5.45809686e-01 5.10809839e-01 1.09335279e+00 -3.19144696e-01 -2.68361002e-01 -6.60557508e-01 1.78876102e-01 -4.66363221e-01 7.18138695e-01 -9.71631408e-01 -2.04631105e-01 -2.47819975e-01 3.51616025e-01 -4.68866140e-01 2.52973467e-01 -9.74777699e-01 3.19347829e-01 4.00316805e-01 5.94620518e-02 -3.36404085e-01 5.24717093e-01 -5.55731505e-02 -4.77880359e-01 -4.44810003e-01 9.14556980e-01 -4.76589441e-01 -1.25222361e+00 3.64035815e-01 -6.44762635e-01 -6.68075919e-01 8.98078859e-01 -5.75143099e-01 -1.73275292e-01 -2.36497596e-01 -7.48731136e-01 -1.86206363e-02 4.62353140e-01 3.13244581e-01 2.93018341e-01 -7.90104628e-01 -9.32796955e-01 4.84123230e-01 1.49717480e-01 2.39570960e-02 3.12438577e-01 6.17059767e-01 -1.18045998e+00 3.64446223e-01 -7.72073090e-01 -7.05005407e-01 -1.26783741e+00 9.29457471e-02 6.24351561e-01 -9.30893607e-03 -4.90197867e-01 7.01719105e-01 -2.35949308e-01 -5.40490746e-01 -1.37297586e-01 -3.51212740e-01 -3.54409099e-01 4.35201019e-01 5.29889822e-01 2.31606588e-01 5.12120068e-01 -6.85976744e-01 -1.65980712e-01 7.46816620e-02 2.52974421e-01 -1.08127082e-02 1.77747154e+00 1.82098582e-01 -5.03089011e-01 2.29398683e-01 9.61537659e-01 -4.32912767e-01 -1.20991504e+00 1.21078141e-01 -1.30598739e-01 -4.59679544e-01 8.35090339e-01 -8.02827656e-01 -1.24069035e+00 1.02267039e+00 1.03290641e+00 4.15566206e-01 1.37324440e+00 -5.69700301e-01 -4.17946391e-02 6.15532875e-01 2.29445156e-02 -1.04833591e+00 -4.89912301e-01 7.24597752e-01 8.25780928e-01 -1.52521551e+00 1.62842914e-01 -2.80207414e-02 -5.72499037e-01 1.50855899e+00 1.90595135e-01 1.41751379e-01 5.55647314e-01 1.55778239e-02 3.21666598e-01 -1.57368734e-01 -1.65581524e-01 -6.53350592e-01 2.27154404e-01 9.05784547e-01 7.44906723e-01 3.32007021e-01 -1.77929088e-01 -1.31809980e-01 -1.14147052e-01 1.26840904e-01 7.15746820e-01 6.39011860e-01 -6.47837400e-01 -1.07401466e+00 -7.14369237e-01 5.78945518e-01 -3.43582779e-01 -4.47043329e-01 -3.58752102e-01 7.74051785e-01 1.77205712e-01 8.04028869e-01 1.63492441e-01 -8.07281509e-02 4.54895571e-02 7.52383173e-02 1.30524114e-01 -5.53530216e-01 -5.09097874e-01 1.50985606e-02 -5.89870196e-03 -2.32975066e-01 -6.67873621e-01 -6.28877878e-01 -8.07235658e-01 -3.62697124e-01 -2.74929464e-01 5.37548140e-02 1.30535531e+00 9.25315261e-01 1.88947722e-01 2.41925985e-01 5.44260204e-01 -1.20120049e+00 -5.58809161e-01 -1.27022028e+00 -9.76892531e-01 2.04327136e-01 3.94972920e-01 -3.11218768e-01 -2.24093348e-01 1.61142007e-01]	[9.609736442565918, -1.5363012552261353]
8e06f89c-b667-4776-8d94-6c2debb41e7c	iterative-potts-minimization-for-the-recovery	1812.00862	null	https://arxiv.org/abs/1812.00862v2	https://arxiv.org/pdf/1812.00862v2.pdf	Iterative Potts minimization for the recovery of signals with discontinuities from indirect measurements -- the multivariate case	Signals and images with discontinuities appear in many problems in such diverse areas as biology, medicine, mechanics, and electrical engineering. The concrete data are often discrete, indirect and noisy measurements of some quantities describing the signal under consideration. A frequent task is to find the segments of the signal or image which corresponds to finding the discontinuities or jumps in the data. Methods based on minimizing the piecewise constant Mumford-Shah functional -- whose discretized version is known as Potts functional -- are advantageous in this scenario, in particular, in connection with segmentation. However, due to their non-convexity, minimization of such functionals is challenging. In this paper we propose a new iterative minimization strategy for the multivariate Potts functional dealing with indirect, noisy measurements. We provide a convergence analysis and underpin our findings with numerical experiments.	['Lukas Kiefer', 'Andreas Weinmann', 'Martin Storath']	2018-12-03	null	null	null	null	['electrical-engineering']	['miscellaneous']	[ 4.10178483e-01 -1.95866019e-01 2.04186052e-01 -2.50533074e-01 -7.58388221e-01 -1.20595604e-01 9.11582112e-02 3.39652240e-01 -5.52010953e-01 1.00149500e+00 -2.63906986e-01 5.22022061e-02 -3.06346178e-01 -6.38067842e-01 -7.39272892e-01 -7.80315757e-01 -1.92495912e-01 2.75517493e-01 -2.23174952e-02 -6.89209392e-03 3.91854465e-01 5.04378617e-01 -1.08170116e+00 -3.10231417e-01 9.77681220e-01 1.05944538e+00 1.39927492e-01 3.90098959e-01 -1.16650008e-01 3.05233821e-02 -3.40822279e-01 3.64170992e-03 1.20739855e-01 -6.01822734e-01 -6.29724920e-01 5.71522772e-01 -1.16783366e-01 3.66678722e-02 2.84621447e-01 1.39596164e+00 1.88300177e-01 3.49247038e-01 7.72521019e-01 -9.13469553e-01 -2.26727873e-01 2.57066309e-01 -9.26316261e-01 1.86593413e-01 2.29265943e-01 -1.45600066e-01 7.21139967e-01 -9.38542008e-01 5.69170654e-01 9.37416375e-01 8.83430779e-01 1.80787623e-01 -1.54978776e+00 -9.67158228e-02 -9.29776430e-02 7.22726211e-02 -1.59128523e+00 -1.88983932e-01 8.55307400e-01 -4.94610786e-01 8.72799680e-02 4.35634553e-01 4.42029685e-01 7.42021561e-01 1.90989584e-01 6.01276278e-01 1.33358836e+00 -2.06439331e-01 2.43671775e-01 1.36468798e-01 2.09473178e-01 4.71670598e-01 2.48711601e-01 -4.20144767e-01 1.53322905e-01 -2.45561391e-01 8.33241582e-01 3.45360599e-02 -4.81115729e-01 -1.69964984e-01 -1.04535067e+00 8.15820754e-01 2.06268445e-01 6.43614292e-01 -6.46692336e-01 2.38871528e-03 1.91669211e-01 2.07938939e-01 7.54899561e-01 3.37642908e-01 2.01109629e-02 8.98327399e-03 -1.01454818e+00 -1.51948370e-02 8.65573525e-01 7.08808661e-01 8.34318757e-01 -1.03637993e-01 2.34330878e-01 6.10779285e-01 1.34650782e-01 3.13221902e-01 8.74918327e-02 -8.58489215e-01 1.57177299e-01 2.13911876e-01 2.44981512e-01 -1.09903693e+00 -4.07654315e-01 -4.62588310e-01 -1.11411202e+00 2.59829983e-02 7.23590910e-01 -4.62258197e-02 -5.60253382e-01 1.43936908e+00 6.10982060e-01 5.59073389e-01 -2.98049957e-01 1.02986181e+00 2.90825397e-01 6.62004232e-01 -1.08556606e-01 -7.93550670e-01 9.84370470e-01 -2.56990165e-01 -7.37428844e-01 2.56631941e-01 2.79381633e-01 -6.80582941e-01 9.73690927e-01 5.87358594e-01 -1.20072258e+00 -3.17150831e-01 -6.86254740e-01 1.49329782e-01 6.10757098e-02 3.00728306e-02 7.98364803e-02 3.17507714e-01 -7.18881249e-01 8.00789654e-01 -7.37476230e-01 -2.34625638e-01 1.80824310e-01 2.71880269e-01 -1.78019047e-01 2.32601568e-01 -7.85595834e-01 7.10040510e-01 -7.89835528e-02 5.05886495e-01 -4.22644764e-01 -4.44303244e-01 -5.54739237e-01 6.71757832e-02 5.14013529e-01 -3.02890599e-01 8.57094765e-01 -1.10745120e+00 -1.38538408e+00 8.50627184e-01 -2.21253335e-01 -3.55304778e-01 1.00482798e+00 1.35809138e-01 -3.20897135e-03 9.30892378e-02 1.13636561e-01 -6.49189427e-02 1.14036047e+00 -1.28739548e+00 -3.39376964e-02 -4.21343207e-01 -1.96672857e-01 1.52395274e-02 1.73622042e-01 -2.99792290e-01 -5.59229329e-02 -6.58504546e-01 4.78316873e-01 -7.17131972e-01 -6.26570225e-01 4.61965464e-02 -7.46840894e-01 1.85634911e-01 6.16629541e-01 -6.60684764e-01 8.75934958e-01 -2.38742924e+00 3.10567021e-01 5.77442646e-01 4.10955288e-02 -7.30521008e-02 2.61870712e-01 4.18700457e-01 1.90394267e-01 3.97466496e-02 -9.46282387e-01 -4.16683018e-01 -4.17640120e-01 8.54563713e-02 -7.01689497e-02 1.20658553e+00 2.08607525e-01 4.35355306e-01 -6.89119637e-01 -6.41101718e-01 2.69080639e-01 5.14183879e-01 -2.85006493e-01 -1.59877867e-01 -1.56544492e-01 1.09034073e+00 -7.95270741e-01 3.16884845e-01 8.00108254e-01 -2.12192059e-01 -2.68755611e-02 -9.25443918e-02 -4.43956256e-01 -3.48752141e-01 -1.40493619e+00 1.54789758e+00 -3.56174648e-01 5.94630837e-01 6.66889071e-01 -1.70522773e+00 7.98623741e-01 2.99576432e-01 8.60602200e-01 -2.24147797e-01 3.98249269e-01 4.58217800e-01 -2.33038917e-01 -6.34206116e-01 2.37500891e-01 -6.47055805e-01 1.19378783e-01 2.09788326e-02 -4.58540916e-01 -2.79935479e-01 2.29099467e-01 -1.35358110e-01 7.31242716e-01 -1.73985004e-01 3.45419228e-01 -6.51467621e-01 8.65624726e-01 -4.16773558e-02 4.81725901e-01 4.07215625e-01 -4.78419438e-02 7.24332571e-01 6.64027572e-01 7.82460570e-02 -8.58364820e-01 -6.55268610e-01 -6.12531483e-01 2.88681425e-02 3.93391043e-01 3.65257591e-01 -8.47795069e-01 -1.90226451e-01 4.90683727e-02 3.75500619e-01 -5.74095726e-01 1.48480773e-01 -5.38336277e-01 -6.14981174e-01 1.22671351e-01 9.65905860e-02 5.33574700e-01 -7.17002332e-01 -5.50262630e-01 6.09469235e-01 -1.78794801e-01 -1.22512817e+00 -5.00567913e-01 -1.04133561e-01 -1.27018535e+00 -1.00589728e+00 -1.12608480e+00 -7.10740089e-01 7.37135231e-01 5.07497452e-02 6.81423068e-01 9.67036858e-02 -5.20510077e-01 4.27306294e-01 -8.65667239e-02 2.21079513e-02 -4.32912230e-01 -2.23127156e-01 -1.51140168e-01 5.90448618e-01 -3.58406782e-01 -5.90156734e-01 -5.22097230e-01 5.40376365e-01 -9.96679246e-01 -3.61553907e-01 3.94991398e-01 7.05587626e-01 1.07567382e+00 1.41437277e-01 6.44515634e-01 -1.03597367e+00 6.48264587e-01 -5.10149240e-01 -8.14430356e-01 8.00603107e-02 -2.41493553e-01 -5.69178388e-02 7.25197613e-01 -5.17430544e-01 -7.80184805e-01 1.29256085e-01 -2.38794684e-02 -3.81068468e-01 -1.71835080e-01 7.14750946e-01 -1.30883113e-01 -2.95616984e-01 3.91188622e-01 2.68130839e-01 1.03584923e-01 -6.82173431e-01 8.41872692e-02 5.28159738e-01 4.87970471e-01 -6.35711491e-01 3.86271805e-01 8.88774157e-01 5.28390765e-01 -1.47674429e+00 -5.41299522e-01 -7.80730665e-01 -4.15869147e-01 -3.92736524e-01 7.83671319e-01 -1.75577417e-01 -7.97286451e-01 4.24274206e-01 -1.08608973e+00 -1.55554414e-01 -5.34798801e-01 6.65096819e-01 -7.24513113e-01 7.11188018e-01 -6.47534609e-01 -1.14492595e+00 1.20649055e-01 -1.34104002e+00 8.73424947e-01 2.39766225e-01 4.64589000e-02 -1.24826121e+00 7.82700256e-02 9.22730565e-02 3.14967275e-01 6.24575794e-01 6.63880885e-01 -1.45421669e-01 -3.34156871e-01 -2.79765487e-01 -2.38843143e-01 4.99255657e-01 1.74468681e-01 -5.93053810e-02 -4.94681120e-01 -1.15246832e-01 6.16466641e-01 1.08773150e-01 6.48667395e-01 1.03510010e+00 1.23497200e+00 -1.05397729e-02 -2.39091277e-01 5.08688629e-01 1.58221018e+00 -1.06424419e-02 3.68694782e-01 -7.83591941e-02 3.56083304e-01 7.66090930e-01 6.37216330e-01 6.23985350e-01 -2.01968446e-01 5.92762530e-01 3.91851813e-01 -9.63351503e-02 3.27948570e-01 2.74797261e-01 -2.64953561e-02 7.52158105e-01 -1.80071637e-01 -1.19335882e-01 -5.09781361e-01 5.86150944e-01 -1.81738997e+00 -7.79397011e-01 -7.40088761e-01 2.39460254e+00 7.94177234e-01 3.43780443e-02 9.13589001e-02 5.51392138e-01 9.83750165e-01 -1.07369997e-01 -3.58805835e-01 -2.75221586e-01 -1.14147894e-01 1.35352984e-01 5.18200696e-01 6.97959900e-01 -9.25442338e-01 2.35071540e-01 5.56520557e+00 7.40728974e-01 -1.25488484e+00 1.05430640e-01 5.82542419e-01 4.20352131e-01 -1.57984033e-01 -1.98776983e-02 -4.25239563e-01 6.55664325e-01 5.90092897e-01 -1.37557406e-02 2.28115246e-01 3.88495594e-01 5.97310185e-01 -6.31019771e-01 -7.95633376e-01 9.19322073e-01 -3.24329197e-01 -1.05288541e+00 -4.95480657e-01 2.48132139e-01 7.52230406e-01 -4.79149759e-01 5.02270162e-02 -3.59077841e-01 -5.91909647e-01 -8.00530314e-01 5.92903316e-01 7.49439836e-01 7.82143652e-01 -6.59023643e-01 4.90223587e-01 4.74309325e-01 -9.98786449e-01 3.86456162e-01 -2.28805616e-01 1.00990228e-01 6.14817679e-01 1.15244889e+00 -4.16519403e-01 4.10741508e-01 1.53750286e-01 7.15406656e-01 1.50114268e-01 1.42347085e+00 1.11764990e-01 6.64099991e-01 -5.91166556e-01 -1.22235142e-01 2.53548294e-01 -1.06584728e+00 9.07622457e-01 1.09999645e+00 4.95710880e-01 4.15950626e-01 2.25271925e-01 1.08137488e+00 -6.18110932e-02 3.56520534e-01 -4.38864529e-01 6.57685399e-02 6.66047330e-04 1.11813581e+00 -1.24713790e+00 -1.72938332e-01 -3.95562768e-01 8.94612074e-01 -2.27339804e-01 5.05952597e-01 -6.09233320e-01 -1.89216346e-01 3.11849982e-01 4.42942858e-01 -5.65546900e-02 -4.67982858e-01 -4.97737646e-01 -9.85802948e-01 1.45225719e-01 -3.88064146e-01 1.83904976e-01 -9.91327614e-02 -1.19227481e+00 1.79134592e-01 2.04993889e-01 -1.12321270e+00 1.25325337e-01 -4.02670026e-01 -6.66616678e-01 8.22699189e-01 -1.20180666e+00 -5.86384833e-01 -2.48321697e-01 7.57599711e-01 5.72278082e-01 5.92026114e-01 1.74665362e-01 5.24601221e-01 -5.69016576e-01 -8.77305418e-02 3.82988185e-01 -2.09962875e-01 1.55213773e-01 -1.16917396e+00 -1.47738740e-01 6.89468861e-01 -8.43120068e-02 3.16746712e-01 1.13392746e+00 -5.41358113e-01 -1.26816487e+00 -5.37709355e-01 5.96608102e-01 3.91908139e-01 7.28835404e-01 7.54952282e-02 -1.25640798e+00 4.58299816e-01 -1.99647203e-01 2.20277712e-01 1.58504084e-01 -4.43969458e-01 6.04512751e-01 5.14134355e-02 -1.46697271e+00 2.54660875e-01 5.64060152e-01 -2.74075300e-01 -1.21832885e-01 3.48600090e-01 7.80234486e-02 -3.51577431e-01 -8.84456694e-01 2.01551363e-01 2.10104957e-01 -8.75183940e-01 9.08011734e-01 -3.69597316e-01 6.85796365e-02 -1.30377010e-01 1.61347948e-02 -1.23362827e+00 1.52922720e-01 -7.60322213e-01 3.66079867e-01 9.20884132e-01 2.63665438e-01 -8.61534476e-01 7.51740813e-01 3.15401584e-01 -5.84934130e-02 -7.91416049e-01 -1.30559313e+00 -7.90409505e-01 -1.96057811e-01 -5.38558602e-01 4.59893830e-02 9.51434672e-01 -1.82017282e-01 -1.57640651e-02 -2.66945004e-01 1.30313665e-01 1.10569024e+00 4.53337133e-02 3.55098963e-01 -1.34994936e+00 -3.33910108e-01 -4.14505959e-01 -5.55642009e-01 -9.88873124e-01 1.12963999e-02 -5.70296228e-01 4.14166689e-01 -1.36915493e+00 -1.83276042e-01 -6.73045993e-01 3.20490561e-02 -3.36522877e-01 -1.95070505e-02 8.31108540e-02 -1.30687714e-01 2.77376682e-01 -1.64624795e-01 4.87302601e-01 1.45806885e+00 -2.24482566e-02 -4.14438695e-01 5.92041969e-01 6.30967133e-03 8.34972620e-01 6.58251107e-01 -4.81166989e-01 -2.88165808e-01 -1.21577136e-01 2.45918259e-02 5.08558512e-01 4.25180554e-01 -8.62499356e-01 5.35112917e-02 -1.98202953e-01 -1.43823668e-01 -3.15919608e-01 4.69720066e-01 -8.78763378e-01 2.39105389e-01 4.41685706e-01 -2.00490028e-01 -2.02265382e-01 -1.35146335e-01 7.38642931e-01 -4.31488186e-01 -6.99127913e-01 1.13989782e+00 -8.60543698e-02 -3.76394331e-01 1.79434389e-01 -3.31911057e-01 3.07573050e-01 9.99812722e-01 -2.34752417e-01 2.68850714e-01 -5.71914971e-01 -1.13772309e+00 3.76653038e-02 3.85666221e-01 -4.03227627e-01 7.09952950e-01 -9.25514340e-01 -5.93526006e-01 1.15853153e-01 -4.23387229e-01 1.82177410e-01 2.40318671e-01 1.66234481e+00 -6.55641615e-01 4.69764844e-02 1.16871260e-01 -7.75349915e-01 -9.78306234e-01 2.48196334e-01 3.19157600e-01 -2.87723839e-02 -6.83819413e-01 6.96062565e-01 5.27851731e-02 2.04422027e-01 6.47859946e-02 -6.63581312e-01 1.58568341e-02 1.77606910e-01 7.04096109e-02 8.05384696e-01 6.12090416e-02 -8.28864753e-01 -2.55935699e-01 7.54236698e-01 5.46074390e-01 -3.99255812e-01 1.18535185e+00 -2.84828037e-01 -3.19954932e-01 8.50421309e-01 1.53188801e+00 5.91621324e-02 -1.27240443e+00 -2.07354635e-01 2.54830003e-01 -4.15498227e-01 8.64887908e-02 1.06586620e-01 -1.00501859e+00 8.42290401e-01 2.68446684e-01 7.77034521e-01 1.02921093e+00 -1.30247757e-01 9.08969641e-01 3.16172875e-02 4.41882282e-01 -1.12101054e+00 -2.77926117e-01 9.60584655e-02 7.80834496e-01 -1.07539117e+00 -1.11789443e-02 -5.87945104e-01 -3.67734879e-01 1.04285872e+00 -1.09276056e-01 -4.87516522e-01 9.67754781e-01 1.80298209e-01 -1.23365708e-01 -4.09769407e-03 -2.22687460e-02 -3.31094533e-01 2.84347206e-01 9.05229226e-02 3.55855763e-01 6.05123937e-02 -9.14181173e-01 9.45759863e-02 3.39239448e-01 1.01308495e-01 5.41718483e-01 9.43655312e-01 -3.78660470e-01 -8.28416049e-01 -6.15285397e-01 5.09169877e-01 -6.00776494e-01 3.28082651e-01 -4.88755628e-02 6.15232468e-01 -8.15430954e-02 8.36667657e-01 -2.64314622e-01 1.97257310e-01 4.77284551e-01 -2.48320729e-01 5.37341535e-01 -3.87174159e-01 -3.43594283e-01 4.66613114e-01 -2.14267775e-01 -4.55671191e-01 -5.96796036e-01 -9.46835577e-01 -1.42653155e+00 -1.38167590e-01 -5.63839674e-01 3.67476583e-01 9.29614007e-01 1.02713585e+00 -3.23364168e-01 2.93428600e-01 7.71590829e-01 -1.03601468e+00 -8.97775054e-01 -7.99524784e-01 -1.16202831e+00 4.70550269e-01 5.86535811e-01 -5.38618505e-01 -5.70715189e-01 -7.09989667e-02]	[7.233721733093262, 3.973675489425659]
03987820-2690-4789-8fc3-8a50ae860aa7	beyond-the-camera-neural-networks-in-world	2003.05614	null	https://arxiv.org/abs/2003.05614v1	https://arxiv.org/pdf/2003.05614v1.pdf	Beyond the Camera: Neural Networks in World Coordinates	Eye movement and strategic placement of the visual field onto the retina, gives animals increased resolution of the scene and suppresses distracting information. This fundamental system has been missing from video understanding with deep networks, typically limited to 224 by 224 pixel content locked to the camera frame. We propose a simple idea, WorldFeatures, where each feature at every layer has a spatial transformation, and the feature map is only transformed as needed. We show that a network built with these WorldFeatures, can be used to model eye movements, such as saccades, fixation, and smooth pursuit, even in a batch setting on pre-recorded video. That is, the network can for example use all 224 by 224 pixels to look at a small detail one moment, and the whole scene the next. We show that typical building blocks, such as convolutions and pooling, can be adapted to support WorldFeatures using available tools. Experiments are presented on the Charades, Olympic Sports, and Caltech-UCSD Birds-200-2011 datasets, exploring action recognition, fine-grained recognition, and video stabilization.	['Karteek Alahari', 'Cordelia Schmid', 'Gunnar A. Sigurdsson', 'Abhinav Gupta']	2020-03-12	null	null	null	null	['video-stabilization']	['computer-vision']	[-5.91493025e-02 -2.72950321e-01 1.27750263e-01 -3.54307562e-01 3.74582827e-01 -4.63590950e-01 3.07241350e-01 -3.54813367e-01 -7.46619284e-01 5.45309663e-01 7.37347975e-02 1.60141125e-01 -8.26818645e-02 -3.80768865e-01 -1.07000613e+00 -6.01540923e-01 -1.10292487e-01 -3.64584565e-01 6.56452060e-01 -1.32051840e-01 3.74950677e-01 6.68333590e-01 -2.06844711e+00 4.79604185e-01 9.53988433e-02 9.87637520e-01 4.15592998e-01 7.37334132e-01 2.60029763e-01 1.06084132e+00 -4.40803438e-01 1.53313786e-01 4.75008279e-01 -3.07051927e-01 -5.14448404e-01 3.42598110e-01 1.01308763e+00 -5.77797949e-01 -8.02082777e-01 1.10814142e+00 1.42309695e-01 5.13442159e-01 1.28586948e-01 -1.18944895e+00 -6.22591972e-01 2.59750426e-01 -6.77198887e-01 7.93089509e-01 1.87759846e-02 6.46459758e-01 5.13836682e-01 -9.03040648e-01 6.68668807e-01 1.26307106e+00 3.81942779e-01 7.49152243e-01 -1.18043911e+00 -2.74017394e-01 2.65993088e-01 4.64606792e-01 -1.02369726e+00 -7.63009787e-01 2.58917928e-01 -4.22791958e-01 1.40543056e+00 1.19214579e-01 9.62716877e-01 7.50664175e-01 3.18914980e-01 6.22486651e-01 9.38021541e-01 -3.37843329e-01 2.14981940e-02 -2.44351819e-01 5.26809692e-01 7.28366077e-01 1.82018578e-01 3.36111128e-01 -1.00444818e+00 5.32951236e-01 1.18164670e+00 2.16624841e-01 -6.44724131e-01 -4.03292567e-01 -1.50694692e+00 3.42036724e-01 6.95403099e-01 -3.75651792e-02 -4.22085941e-01 2.95145512e-01 4.76715341e-02 6.05020940e-01 2.74290740e-01 4.14892256e-01 -3.89080346e-01 1.47419777e-02 -9.29839551e-01 2.26771042e-01 4.58523303e-01 8.26029420e-01 9.02999699e-01 1.57784671e-01 -3.78694624e-01 5.06988168e-01 2.70226687e-01 4.53475416e-01 6.74316883e-01 -1.44443536e+00 8.88760015e-02 3.46845776e-01 1.18535712e-01 -7.95441270e-01 -7.55235493e-01 -8.83930996e-02 -5.93642592e-01 7.68855751e-01 6.41800225e-01 -3.19265395e-01 -1.24930835e+00 1.76908314e+00 1.77874625e-01 4.08909529e-01 -1.78626463e-01 1.20634425e+00 8.42579484e-01 5.97276568e-01 -1.70339122e-01 1.07245119e-02 1.29279912e+00 -1.20679235e+00 -4.82549518e-01 -6.52862251e-01 2.99855202e-01 -3.79395902e-01 8.74780536e-01 3.31373692e-01 -1.38838732e+00 -8.11199069e-01 -1.08387280e+00 -3.10318410e-01 -3.62512380e-01 1.92045167e-01 7.45285749e-01 1.99649334e-01 -1.55911708e+00 8.91144216e-01 -1.05136561e+00 -6.75763011e-01 3.55559409e-01 4.68690991e-01 -6.83580279e-01 1.10692397e-01 -8.35725963e-01 9.58999038e-01 2.03557089e-01 3.31755072e-01 -9.76460338e-01 -5.47095060e-01 -8.87520134e-01 3.04686069e-01 3.33161652e-01 -7.89489686e-01 1.03731430e+00 -1.11727071e+00 -1.44842136e+00 8.76436174e-01 -1.39513761e-01 -7.83741593e-01 3.79823558e-02 -2.73563474e-01 -2.53400952e-01 2.63130635e-01 -1.47848248e-01 1.14324117e+00 1.06054294e+00 -3.81358236e-01 -6.60354972e-01 -6.93640649e-01 3.05411875e-01 2.98698366e-01 -1.05715372e-01 3.51150990e-01 -6.37493849e-01 -3.83155555e-01 1.12142667e-01 -9.79483783e-01 -2.15165347e-01 4.77107257e-01 -2.08314043e-02 6.53307438e-02 5.86218774e-01 -2.97851592e-01 8.72780025e-01 -2.49640608e+00 2.81317711e-01 -1.50403276e-01 3.08905125e-01 3.99306357e-01 -4.61309046e-01 3.59291919e-02 -4.01751131e-01 -2.85570562e-01 1.23588629e-01 -5.83293028e-02 -3.16296428e-01 1.05256811e-01 -1.03935614e-01 8.03572237e-01 3.25658172e-01 7.44951010e-01 -5.19681036e-01 -1.13372736e-01 3.96200299e-01 3.78765196e-01 -8.00437450e-01 -2.02998698e-01 5.41659892e-02 1.35588437e-01 -9.52247903e-02 3.31335217e-01 6.87781692e-01 -2.57161647e-01 -2.86774248e-01 -6.00715280e-02 -6.38365030e-01 -2.78280228e-02 -1.21426010e+00 1.87480533e+00 1.37751639e-01 1.36204433e+00 2.38621578e-01 -8.35918665e-01 5.45390189e-01 -1.55371308e-01 1.70209289e-01 -7.89366782e-01 2.26121575e-01 -3.07890058e-01 3.36809307e-01 -5.32856464e-01 4.05116826e-01 3.95254105e-01 4.09530103e-01 3.17536831e-01 3.45536441e-01 3.11619610e-01 5.94995022e-01 -3.33586931e-02 1.19327223e+00 2.98005015e-01 2.06449658e-01 -2.43783906e-01 2.86154807e-01 -1.08560078e-01 2.70238936e-01 8.00654411e-01 -5.08760929e-01 7.00809360e-01 5.38517952e-01 -8.06737542e-01 -7.49926984e-01 -7.36971736e-01 2.84637846e-02 1.23049104e+00 2.34613091e-01 -2.69957125e-01 -9.05494630e-01 -1.29194945e-01 -1.88565496e-02 3.06230843e-01 -8.43947649e-01 -2.95870930e-01 -5.58950245e-01 -3.63101780e-01 1.75277710e-01 4.28633392e-01 6.39246345e-01 -1.47038591e+00 -1.28823316e+00 1.35901764e-01 2.20413432e-01 -8.89784694e-01 -5.06653011e-01 4.45381701e-01 -7.18017459e-01 -1.16006386e+00 -6.25457048e-01 -7.15615153e-01 4.98449057e-01 7.96866834e-01 8.25163007e-01 -1.77422408e-02 -4.44920957e-01 2.32005388e-01 -1.06589422e-02 -1.71337530e-01 3.49495620e-01 -4.63815123e-01 3.49122494e-01 1.34036645e-01 5.90075850e-01 -5.21929741e-01 -7.39739776e-01 2.30420128e-01 -7.94789970e-01 7.53835365e-02 4.37989712e-01 7.90260196e-01 3.90438259e-01 -2.90363908e-01 -4.11916375e-02 -4.10749286e-01 1.12592682e-01 -9.70322713e-02 -7.85270810e-01 1.02794327e-01 1.29751801e-01 9.64481756e-02 4.33581173e-01 -3.50515217e-01 -7.59346604e-01 2.48438701e-01 1.43639788e-01 -7.15798438e-01 -3.93429786e-01 2.37654611e-01 1.38458818e-01 -4.73963290e-01 9.22458351e-01 2.44919583e-01 2.24548027e-01 -2.70954281e-01 3.47388864e-01 3.33719999e-01 7.03481019e-01 9.77468714e-02 4.52749372e-01 6.88658834e-01 1.49650229e-02 -1.03802097e+00 -7.78261185e-01 -4.89058524e-01 -8.82377684e-01 -3.50943565e-01 1.05577934e+00 -1.06555247e+00 -9.88586962e-01 7.96782553e-01 -1.24189115e+00 -5.09526312e-01 -4.69541878e-01 7.38478243e-01 -5.55900753e-01 9.25290734e-02 -3.56897712e-01 -3.40845197e-01 1.55406550e-01 -9.10109699e-01 7.89799035e-01 7.92641759e-01 5.05481400e-02 -4.69097167e-01 -6.90632239e-02 -2.58205682e-01 3.92135262e-01 -5.62934950e-02 3.68005276e-01 -2.56334364e-01 -7.93313265e-01 -2.87347306e-02 -2.67162263e-01 3.61420035e-01 -6.08443953e-02 3.23325694e-01 -1.08255255e+00 -1.41604558e-01 7.43732825e-02 -3.40891659e-01 1.36287451e+00 9.24173057e-01 1.35366583e+00 1.13435220e-02 -2.06183985e-01 1.12495637e+00 1.04241335e+00 1.27734721e-01 9.30114150e-01 4.86715049e-01 4.77379590e-01 5.56855440e-01 3.25460762e-01 5.16957045e-01 1.80209443e-01 5.97051144e-01 5.97624779e-01 -5.19682579e-02 -4.07828808e-01 1.77088767e-01 4.24018115e-01 1.67465165e-01 -3.39830190e-01 -1.25196591e-01 -4.85981882e-01 3.78851593e-01 -1.82386565e+00 -1.40084672e+00 -4.06539403e-02 2.34672093e+00 4.17465031e-01 1.44302353e-01 -4.85706590e-02 -3.14485848e-01 7.91600823e-01 3.95478606e-01 -7.59871185e-01 -2.26395503e-01 -3.52472603e-01 1.91383168e-01 6.93747878e-01 1.95813388e-01 -1.20579076e+00 1.02846789e+00 7.00720978e+00 2.79942840e-01 -1.33841193e+00 -2.40594104e-01 3.65956306e-01 -6.71968520e-01 3.92407566e-01 -1.37316778e-01 -8.62121582e-01 4.24342036e-01 9.20543611e-01 -6.71927631e-02 8.74945164e-01 8.53940189e-01 4.15196121e-01 -3.41113001e-01 -1.30721307e+00 1.32216215e+00 1.90449104e-01 -1.50171232e+00 -2.67878145e-01 -1.07540466e-01 4.48057652e-01 4.79486048e-01 1.50806293e-01 2.80559838e-01 2.19008818e-01 -8.51524234e-01 6.49747014e-01 7.82008111e-01 4.71118867e-01 -4.31644201e-01 3.51147577e-02 3.83213133e-01 -7.94246256e-01 -2.41187304e-01 -8.58664036e-01 -3.26645732e-01 -1.12122111e-01 -6.26484007e-02 -9.04248878e-02 -1.67497605e-01 1.20846176e+00 1.06212461e+00 -8.52259457e-01 1.35737598e+00 -2.40759291e-02 2.21282616e-01 -3.78807276e-01 -4.32390124e-02 3.79585534e-01 3.76082398e-02 4.89178419e-01 8.86985838e-01 3.23630452e-01 1.98878378e-01 -1.81167737e-01 6.16539896e-01 -4.33441950e-03 -4.18490708e-01 -6.68965042e-01 1.67088926e-01 7.22195059e-02 1.29596484e+00 -6.27666056e-01 -3.17970693e-01 -6.45045519e-01 1.05003500e+00 4.75625604e-01 5.75375915e-01 -8.16845894e-01 -5.64638317e-01 1.05322683e+00 1.59891531e-01 6.79501593e-01 -2.78700918e-01 1.78054124e-01 -1.32684958e+00 -6.55637160e-02 -8.23867083e-01 6.03396855e-02 -1.37164760e+00 -6.73534513e-01 4.85619694e-01 -1.76706836e-01 -1.41017389e+00 -2.07862109e-01 -1.11358881e+00 -5.13655126e-01 8.18260252e-01 -1.46904063e+00 -5.60172319e-01 -4.68654126e-01 1.11744058e+00 6.59174621e-01 -1.56011358e-01 4.28579837e-01 5.45674823e-02 -6.16737843e-01 2.28303537e-01 1.33124664e-01 -6.86187344e-03 8.20632815e-01 -9.40504730e-01 5.82328916e-01 1.14890528e+00 3.80203068e-01 7.28336275e-01 6.53289795e-01 -1.39377341e-01 -1.42171049e+00 -9.69655752e-01 4.84073460e-01 -3.73013794e-01 6.88056767e-01 -2.63486564e-01 -9.20236528e-01 8.09126139e-01 3.65542769e-01 6.06057644e-01 2.80680329e-01 -1.29186332e-01 -1.79102615e-01 -2.52154380e-01 -8.97295833e-01 7.58149564e-01 1.24124587e+00 -2.30796292e-01 -6.03912354e-01 3.15376490e-01 5.26838362e-01 -6.48852229e-01 -1.57129854e-01 -2.35726982e-01 6.99325383e-01 -1.36363459e+00 8.79663289e-01 -1.05899370e+00 2.89643884e-01 -6.14988327e-01 4.09214683e-02 -1.50023615e+00 -8.15937400e-01 -6.34426355e-01 -9.29877982e-02 5.13148487e-01 1.23538300e-01 -4.89254862e-01 5.59778273e-01 6.48152888e-01 -4.04186279e-01 1.53059121e-02 -1.05786061e+00 -5.35581589e-01 -4.04651433e-01 -2.02654690e-01 3.03204089e-01 4.03425008e-01 -1.03709944e-01 1.34753287e-01 -5.19359171e-01 2.54806578e-01 5.12372911e-01 -1.04759827e-01 8.67899418e-01 -1.13921833e+00 -2.73333430e-01 -2.97066569e-01 -9.31159556e-01 -1.29912615e+00 -4.81151510e-03 -3.76183271e-01 1.23680793e-01 -1.28983927e+00 -5.44183105e-02 4.69796330e-01 -4.00083452e-01 7.56398439e-01 7.88285285e-02 4.10229981e-01 3.51444244e-01 1.47380427e-01 -7.01246083e-01 2.66023844e-01 1.45999396e+00 5.31708673e-02 -1.10970318e-01 -2.32647918e-02 -6.38226509e-01 1.03354144e+00 4.95050907e-01 -2.52465278e-01 -2.19174132e-01 -8.54935408e-01 -6.69504628e-02 -1.79833900e-02 7.68620491e-01 -1.40019560e+00 8.16855073e-01 -1.00452110e-01 7.96504796e-01 -4.96345669e-01 5.34011960e-01 -7.68619955e-01 -8.77125859e-02 2.85280496e-01 -3.44989479e-01 8.45867302e-03 4.75162953e-01 5.54859579e-01 -2.40946144e-01 -7.31399432e-02 9.51108754e-01 -4.09654915e-01 -1.42028117e+00 4.32787716e-01 -4.72121298e-01 -1.04544990e-01 1.10843587e+00 -4.64132786e-01 -6.84079528e-01 -1.34135380e-01 -1.10355234e+00 3.35674196e-01 4.44189578e-01 4.17679042e-01 6.02021754e-01 -1.19069552e+00 -5.59670389e-01 7.13093817e-01 -5.49094491e-02 -3.34905416e-01 7.30367720e-01 9.37291205e-01 -7.33621538e-01 5.16461253e-01 -9.51567888e-01 -8.54481578e-01 -1.34833837e+00 8.45852315e-01 5.56526780e-01 4.53076154e-01 -7.41694272e-01 1.02725947e+00 6.87914312e-01 2.86482275e-01 2.48321012e-01 -7.63675153e-01 -5.56167126e-01 5.78513034e-02 1.08927357e+00 7.99208581e-02 2.04211399e-02 -5.95967352e-01 -4.56925929e-01 7.14728832e-01 -1.00407094e-01 4.29398902e-02 1.40205133e+00 -3.91975462e-01 -2.22457021e-01 3.71874452e-01 7.60607600e-01 -3.97983104e-01 -1.95469773e+00 -1.87453553e-01 -3.17429632e-01 -7.36661494e-01 1.98582545e-01 -4.09241170e-01 -1.31396902e+00 9.86930549e-01 6.96873426e-01 3.72681320e-01 1.26577640e+00 -2.12466836e-01 -1.90012474e-02 8.62916112e-01 2.53916264e-01 -9.21682119e-01 -1.82047606e-01 7.39633620e-01 9.24330294e-01 -1.19729590e+00 -1.98147684e-01 1.09124586e-01 -7.62956023e-01 1.35676098e+00 9.81759131e-01 -5.39853573e-01 7.08351076e-01 1.92514911e-01 -6.93971142e-02 -2.47974396e-01 -1.14447021e+00 -4.46276695e-01 1.25583172e-01 6.97044730e-01 1.66136041e-01 -4.29106623e-01 3.40240180e-01 2.32967362e-01 4.42358479e-02 1.48551971e-01 8.00389886e-01 7.94350505e-01 -9.60214019e-01 -3.31681043e-01 -2.64192820e-01 3.82863581e-01 -3.25691611e-01 -3.89328510e-01 -1.69471860e-01 9.23240125e-01 1.53319523e-01 6.76050663e-01 5.97472310e-01 -1.63001761e-01 4.18324590e-01 -6.91133216e-02 6.18391991e-01 -7.67471910e-01 -3.95655185e-01 1.19110897e-01 -1.69372186e-01 -1.14489555e+00 -6.59506500e-01 -7.63332427e-01 -9.42935407e-01 -3.45393151e-01 -2.37437971e-02 -3.61949563e-01 5.19708991e-01 6.57412767e-01 4.62013036e-01 6.35641694e-01 4.11845535e-01 -1.25857401e+00 -1.92203268e-01 -1.01506901e+00 -9.38398063e-01 -2.07225699e-03 7.15343177e-01 -7.31655896e-01 -4.44958121e-01 4.35936064e-01]	[9.2272367477417, 1.5494073629379272]
485cdace-4642-4dc3-a6c5-0dfa9dcd499f	beyond-instance-level-image-retrieval	null	null	http://openaccess.thecvf.com/content_cvpr_2017/html/Gordo_Beyond_Instance-Level_Image_CVPR_2017_paper.html	http://openaccess.thecvf.com/content_cvpr_2017/papers/Gordo_Beyond_Instance-Level_Image_CVPR_2017_paper.pdf	Beyond Instance-Level Image Retrieval: Leveraging Captions to Learn a Global Visual Representation for Semantic Retrieval	Querying with an example image is a simple and intuitive interface to retrieve information from a visual database. Most of the research in image retrieval has focused on the task of instance-level image retrieval, where the goal is to retrieve images that contain the same object instance as the query image. In this work we move beyond instance-level retrieval and consider the task of semantic image retrieval in complex scenes, where the goal is to retrieve images that share the same semantics as the query image. We show that, despite its subjective nature, the task of semantically ranking visual scenes is consistently implemented across a pool of human annotators. We also show that a similarity based on human-annotated region-level captions is highly correlated with the human ranking and constitutes a good computable surrogate. Following this observation, we learn a visual embedding of the images where the similarity in the visual space is correlated with their semantic similarity surrogate. We further extend our model to learn a joint embedding of visual and textual cues that allows one to query the database using a text modifier in addition to the query image, adapting the results to the modifier. Finally, our model can ground the ranking decisions by showing regions that contributed the most to the similarity between pairs of images, providing a visual explanation of the similarity.	['Diane Larlus', 'Albert Gordo']	2017-07-01	null	null	null	cvpr-2017-7	['semantic-retrieval']	['natural-language-processing']	[ 2.21927404e-01 6.14571348e-02 -3.14235777e-01 -5.38071036e-01 -9.39176202e-01 -7.84264088e-01 7.94905603e-01 7.15576768e-01 -4.50246632e-01 1.12865552e-01 3.97768974e-01 1.29029542e-01 -1.69608295e-01 -6.38314068e-01 -8.84322822e-01 -5.21012127e-01 1.65957913e-01 4.03119147e-01 2.81439602e-01 -2.22645864e-01 3.73122692e-01 5.12546420e-01 -1.88125682e+00 6.72567964e-01 2.13390380e-01 1.08469605e+00 6.27771556e-01 4.79257822e-01 -8.38303566e-02 6.65766120e-01 -6.88610733e-01 -3.09249729e-01 2.68113852e-01 -3.77150148e-01 -1.10706043e+00 2.15524957e-01 8.77844632e-01 -2.13370144e-01 -2.55762070e-01 1.03425682e+00 2.18395069e-01 3.32492679e-01 8.50075245e-01 -1.44809496e+00 -1.02415216e+00 2.41643190e-01 -2.31421292e-01 8.85680877e-03 7.02136755e-01 -2.03127459e-01 1.36072302e+00 -1.01361346e+00 1.15267956e+00 1.41800106e+00 -7.39289671e-02 3.29589933e-01 -1.27224159e+00 -1.49586841e-01 1.84475526e-01 4.20936435e-01 -1.58794808e+00 -2.36867115e-01 7.54533827e-01 -5.63334286e-01 6.44276202e-01 4.38525915e-01 5.64464808e-01 6.23651028e-01 -2.61643976e-01 7.31704414e-01 8.34565580e-01 -5.59716582e-01 1.93826854e-01 4.65070516e-01 -2.18404103e-02 5.89499414e-01 -6.04956178e-03 4.51911837e-02 -5.89087009e-01 -2.30208039e-02 5.99672198e-01 3.44881594e-01 -2.79037058e-01 -8.82022679e-01 -1.37827921e+00 9.03391898e-01 1.06653547e+00 5.64982235e-01 -3.61441404e-01 4.57613647e-01 3.37815076e-01 1.92770109e-01 2.57464230e-01 7.97389030e-01 -1.03895374e-01 5.34165502e-01 -8.57980847e-01 1.84283152e-01 4.87872839e-01 1.01499355e+00 1.15061033e+00 -5.04626930e-01 -4.77256149e-01 7.97478259e-01 3.64668727e-01 5.10515630e-01 3.96260649e-01 -1.29139566e+00 1.91341177e-01 6.74181163e-01 2.78305471e-01 -1.26737285e+00 1.37431487e-01 -1.31296977e-01 -3.61679971e-01 3.54878575e-01 1.97905034e-01 8.44009280e-01 -8.63305926e-01 1.91239059e+00 1.44185886e-01 -2.90463150e-01 9.02517214e-02 1.18711257e+00 8.14427257e-01 6.26371026e-01 3.54605108e-01 1.47119418e-01 1.68703640e+00 -8.57886672e-01 -4.13514972e-01 -3.51776838e-01 5.27363420e-01 -7.69695580e-01 1.32287145e+00 -1.94082797e-01 -8.77162755e-01 -6.35603368e-01 -1.07072878e+00 -5.05159914e-01 -8.84322882e-01 1.77801266e-01 1.44388109e-01 1.01393133e-01 -1.28038621e+00 4.20229495e-01 -1.93514436e-01 -8.36554289e-01 1.80499509e-01 -7.13950098e-02 -7.68431067e-01 -2.04781979e-01 -1.08675563e+00 1.02891362e+00 5.57670414e-01 -2.45856047e-01 -1.03882635e+00 -5.34571409e-01 -9.55173969e-01 3.43752466e-02 2.76959866e-01 -8.81464481e-01 1.03924417e+00 -1.42021096e+00 -7.45366395e-01 1.64098060e+00 -3.10308903e-01 -1.60214812e-01 1.57069772e-01 -2.50104326e-03 -1.50436133e-01 5.90385675e-01 6.61987841e-01 1.18234646e+00 9.24900532e-01 -1.84254682e+00 -6.58696771e-01 -5.64616024e-01 4.97162700e-01 3.74494851e-01 -2.02593639e-01 -1.06650189e-01 -8.82392406e-01 -6.20090961e-01 1.01296104e-01 -8.46902966e-01 -7.93936476e-03 4.69742686e-01 -2.09634393e-01 -2.99289197e-01 7.24145472e-01 -4.68259692e-01 9.62269068e-01 -2.31715512e+00 2.58531213e-01 3.89089704e-01 2.23694786e-01 -2.53609747e-01 -4.23156559e-01 7.09735572e-01 -1.89270243e-01 3.02060694e-01 -8.24803784e-02 -1.70108452e-01 1.13204181e-01 2.31771246e-01 -4.80397016e-01 3.17775041e-01 1.09201632e-01 1.15143991e+00 -1.25246453e+00 -7.18240142e-01 1.91978633e-01 3.59105498e-01 -2.69776165e-01 4.78081882e-01 -1.36894464e-01 1.21658906e-01 -4.58843946e-01 5.02261400e-01 1.38981745e-01 -4.65508968e-01 -3.71814705e-02 -7.25448906e-01 2.92976052e-01 -1.93752885e-01 -1.00002801e+00 1.86946380e+00 -4.38297838e-01 8.64219010e-01 -1.52715519e-01 -8.50089252e-01 7.27833927e-01 8.49837288e-02 3.87356669e-01 -1.03022575e+00 -3.02475303e-01 1.85494065e-01 -5.91107428e-01 -5.91858566e-01 6.98506474e-01 -9.97421816e-02 -2.70822376e-01 6.73476398e-01 -2.47455314e-02 -4.14787561e-01 1.63370445e-01 6.79425895e-01 8.03022742e-01 7.91005045e-02 4.32652295e-01 -1.22476280e-01 4.38037694e-01 1.37575611e-01 -2.16051206e-01 1.02793133e+00 -1.38424681e-02 7.43965447e-01 2.07615808e-01 -3.08434397e-01 -1.27205408e+00 -1.37567449e+00 3.30173858e-02 1.44796860e+00 7.47355282e-01 -5.10291815e-01 -5.26598573e-01 -4.38654155e-01 7.63560906e-02 4.17525858e-01 -8.95726979e-01 -1.54178336e-01 -1.95394576e-01 -1.84501652e-02 1.96952850e-01 4.79019970e-01 2.76136070e-01 -1.13687766e+00 -8.13032389e-01 -2.88120776e-01 -3.49887878e-01 -1.00758564e+00 -7.36154497e-01 1.14168867e-03 -6.25027895e-01 -1.13692057e+00 -8.61534476e-01 -1.06545055e+00 9.08323884e-01 5.16631186e-01 1.53827810e+00 4.12851959e-01 -4.81939405e-01 1.03703177e+00 -4.68263984e-01 -1.21569067e-01 -4.23035085e-01 -2.81747431e-01 -3.17351222e-01 1.39688477e-01 6.10463247e-02 4.99067791e-02 -9.46128607e-01 3.13228399e-01 -1.54945445e+00 5.63232973e-02 4.78704363e-01 6.48455679e-01 8.91677439e-01 -4.37538892e-01 1.34834930e-01 -7.09188461e-01 5.83286583e-01 -3.68888885e-01 -2.62130648e-01 7.17802942e-01 -4.37849820e-01 4.45117354e-01 1.91470131e-01 -3.08062732e-01 -5.78023612e-01 2.87015855e-01 4.38446820e-01 -6.01168692e-01 -6.25212789e-02 4.18615460e-01 -7.39285797e-02 2.31681734e-01 6.90389335e-01 1.59689143e-01 -1.47326306e-01 -2.69970477e-01 1.05639660e+00 4.31787193e-01 6.54969454e-01 -5.10827184e-01 7.11613655e-01 7.16831326e-01 1.30133680e-03 -6.03081346e-01 -9.97915685e-01 -8.34943652e-01 -7.37173080e-01 -3.53971779e-01 1.18654609e+00 -9.05197442e-01 -6.35558844e-01 -1.90120384e-01 -1.21680391e+00 -5.41655049e-02 -4.91263509e-01 2.05249473e-01 -8.09215426e-01 3.25486988e-01 -1.08399540e-01 -5.00243068e-01 -7.88526163e-02 -1.16473365e+00 1.71799791e+00 -8.24360698e-02 -2.84663647e-01 -8.41484129e-01 -1.23148412e-01 1.77627355e-01 2.58980393e-01 3.02613020e-01 1.32962406e+00 -5.80533147e-01 -9.25475121e-01 -1.63854614e-01 -5.96658170e-01 1.56394213e-01 7.86285698e-02 -1.93512201e-01 -9.73007798e-01 -2.61278719e-01 -2.58685648e-01 -3.72299433e-01 7.78241873e-01 1.51864707e-01 1.18289936e+00 -2.26722628e-01 -4.48108196e-01 2.61729270e-01 1.76260471e+00 2.41062697e-02 5.47730088e-01 3.64165932e-01 6.79257274e-01 9.97141182e-01 7.30121732e-01 5.77531941e-02 3.04616511e-01 9.82210159e-01 5.79148471e-01 -3.01984847e-01 -2.55330294e-01 -6.10114753e-01 9.94757861e-02 2.81536043e-01 2.88531661e-01 -2.18513325e-01 -8.15362573e-01 7.41383553e-01 -1.96929622e+00 -1.06168830e+00 2.25172400e-01 2.40628982e+00 6.72355235e-01 -3.90566707e-01 6.21687109e-03 -2.49249116e-01 7.43117988e-01 2.61408508e-01 -3.87598544e-01 -2.22381741e-01 -1.34613201e-01 -1.21491246e-01 3.77442360e-01 5.92118919e-01 -9.79381859e-01 8.84770274e-01 6.63666725e+00 8.21901023e-01 -9.17874038e-01 2.84625012e-02 6.58102393e-01 1.15326390e-01 -6.27765715e-01 1.77152872e-01 -3.29773724e-01 1.95797145e-01 4.25320387e-01 -3.83592486e-01 4.47400331e-01 7.70726204e-01 8.02445263e-02 -1.97133914e-01 -1.75793362e+00 1.22023380e+00 5.71635187e-01 -1.23316336e+00 7.91645050e-01 -5.90087436e-02 5.42449713e-01 -3.77159625e-01 2.36005768e-01 -2.98858993e-02 -2.70913132e-02 -1.26634383e+00 1.01663387e+00 7.83840418e-01 8.12649965e-01 -3.16134870e-01 4.32253063e-01 -7.39068389e-02 -1.29205883e+00 3.11445240e-02 -3.56846064e-01 3.54595274e-01 2.18161121e-02 4.71821055e-02 -9.30385172e-01 3.97578806e-01 8.78920496e-01 8.19956720e-01 -1.09498298e+00 8.86001885e-01 -1.13783732e-01 -2.08181500e-01 -2.42931023e-02 9.39996466e-02 1.88939288e-01 -6.63988814e-02 4.07145351e-01 1.11371636e+00 2.38300204e-01 -5.39173037e-02 2.16162115e-01 1.07640266e+00 -2.10405305e-01 3.13321352e-01 -9.81598854e-01 9.64973494e-02 3.95517319e-01 1.12739420e+00 -7.67705262e-01 -5.95046401e-01 -1.36733249e-01 1.25355744e+00 4.17189449e-01 6.33306623e-01 -4.98589128e-01 -3.23671937e-01 5.40683746e-01 1.61583215e-01 1.61794379e-01 1.30274281e-01 1.83679596e-01 -8.56167197e-01 1.81228161e-01 -6.62389398e-01 5.23133039e-01 -1.60572672e+00 -1.34732950e+00 5.93521118e-01 3.21949542e-01 -1.38373923e+00 -3.86571378e-01 -6.19047344e-01 -1.98367044e-01 8.89925480e-01 -1.31869984e+00 -1.15336573e+00 -5.70899487e-01 7.20165074e-01 5.27805924e-01 1.75613597e-01 8.21801007e-01 1.91414684e-01 1.67932659e-01 2.77089447e-01 -6.82094172e-02 1.43815130e-01 8.17303836e-01 -1.28999531e+00 -1.44219458e-01 4.68830138e-01 5.93124330e-01 7.94183493e-01 8.15401316e-01 -2.53716201e-01 -1.22414744e+00 -1.03632700e+00 9.48589385e-01 -7.59569943e-01 5.59232533e-01 -2.34598979e-01 -8.46927106e-01 4.51571256e-01 2.55124182e-01 6.84919134e-02 4.97574568e-01 -3.81580681e-01 -8.04066658e-01 -1.97425008e-01 -8.93196762e-01 7.19794631e-01 9.69660997e-01 -1.23866451e+00 -7.16030240e-01 4.09064978e-01 7.59500504e-01 -3.88334803e-02 -7.37648308e-01 1.67731136e-01 6.42297208e-01 -7.10839152e-01 1.37330341e+00 -7.88245201e-01 4.44477826e-01 -6.75427139e-01 -6.10957742e-01 -1.18478489e+00 -1.40437439e-01 1.48057520e-01 3.92448574e-01 1.02567244e+00 3.19390416e-01 -7.24477321e-02 4.25775379e-01 6.29760623e-01 3.52149844e-01 -3.90093058e-01 -8.33207130e-01 -7.36658394e-01 -1.98339924e-01 -1.89305499e-01 3.72341156e-01 6.40761673e-01 -3.28556150e-01 4.03192729e-01 -1.18093379e-01 1.86010718e-01 4.32623565e-01 5.23343563e-01 6.79274261e-01 -1.09465885e+00 -9.95374024e-02 -4.68971372e-01 -7.87007987e-01 -1.00326645e+00 2.03551441e-01 -1.33706450e+00 2.39822239e-01 -1.73737192e+00 6.08784616e-01 -1.42542079e-01 -4.94963348e-01 2.98403144e-01 -1.01593502e-01 6.62685990e-01 5.92975795e-01 4.83381748e-01 -1.07928252e+00 2.53634691e-01 1.17294025e+00 -5.57318509e-01 8.93339440e-02 -5.47795296e-01 -7.22208917e-01 3.53387505e-01 3.86682868e-01 -5.94425440e-01 -5.12574434e-01 -5.40412784e-01 4.78061199e-01 -7.50207156e-03 7.74366081e-01 -5.80538273e-01 1.36385977e-01 -8.97843242e-02 3.87826622e-01 -2.47549877e-01 5.07087111e-01 -1.12881196e+00 1.89562574e-01 2.31369361e-01 -9.05946732e-01 3.90861481e-01 -8.86625871e-02 7.52119482e-01 -5.71463823e-01 -4.69209194e-01 5.32812059e-01 -2.62720942e-01 -1.14718926e+00 2.97819495e-01 -1.21106274e-01 1.15061738e-01 9.43944633e-01 -2.72399068e-01 -3.23880255e-01 -6.81447566e-01 -8.50618660e-01 7.60896280e-02 9.11633372e-01 7.49550581e-01 8.71968567e-01 -1.62474942e+00 -4.88608032e-01 -1.48223355e-01 1.06727004e+00 -5.26218653e-01 -6.60778880e-02 2.66738266e-01 -4.60710227e-01 4.38599944e-01 -1.48421675e-01 -8.33501995e-01 -1.36042655e+00 1.01902604e+00 3.59845579e-01 7.11593777e-02 -3.47736180e-01 4.35695976e-01 6.46998882e-01 -6.77084401e-02 3.30192536e-01 -1.94705352e-01 -1.85328111e-01 2.73478389e-01 4.34298158e-01 -6.01528361e-02 -1.39718518e-01 -1.01963806e+00 -4.55203533e-01 9.45097804e-01 2.72598028e-01 -4.19333577e-01 9.74259198e-01 -3.85453820e-01 -3.17993820e-01 7.18320727e-01 1.76703191e+00 -1.94455236e-01 -7.63549924e-01 -3.69175166e-01 2.34228313e-01 -7.25384295e-01 -2.41615161e-01 -7.05350637e-01 -9.28527534e-01 8.10197651e-01 8.68033648e-01 1.76468313e-01 1.10383201e+00 6.91319466e-01 1.70014992e-01 6.17417276e-01 3.36449206e-01 -9.43413615e-01 6.55111194e-01 8.90476406e-02 1.26204407e+00 -1.50116968e+00 1.35538979e-02 -1.69993326e-01 -7.32208371e-01 9.71450448e-01 3.05456549e-01 -1.32472381e-01 5.56422114e-01 -6.20309532e-01 2.14070871e-01 -6.28491461e-01 -5.92225671e-01 -6.26280904e-01 8.08886170e-01 6.42677844e-01 3.25486511e-01 6.48314208e-02 -1.10039629e-01 -2.34749634e-03 6.37096018e-02 -4.77945417e-01 1.46014705e-01 7.31530428e-01 -5.57329595e-01 -1.00510371e+00 -3.03945690e-01 2.88046986e-01 -1.30292878e-01 -1.54785797e-01 -7.01841235e-01 7.00113118e-01 -1.33825585e-01 9.42605853e-01 2.27725223e-01 -2.10774075e-02 4.17230666e-01 8.38079378e-02 4.31120396e-01 -8.33725095e-01 -4.82217342e-01 -3.21164310e-01 -3.67023081e-01 -7.96415329e-01 -7.49603868e-01 -1.37709990e-01 -1.10110009e+00 2.96571732e-01 1.01418227e-01 2.26473674e-01 7.86921442e-01 6.63735330e-01 3.29558223e-01 1.19754642e-01 5.19549429e-01 -8.30690145e-01 -2.74924189e-02 -5.21140635e-01 -5.96743882e-01 1.42251718e+00 4.19246793e-01 -5.81586421e-01 -4.15485799e-01 4.46682483e-01]	[10.777669906616211, 1.3182439804077148]
f5aef0f3-fdb1-46a5-827d-092e308d61e6	rethinking-the-objectives-of-vector-quantized	2212.03185	null	https://arxiv.org/abs/2212.03185v2	https://arxiv.org/pdf/2212.03185v2.pdf	Rethinking the Objectives of Vector-Quantized Tokenizers for Image Synthesis	Vector-Quantized (VQ-based) generative models usually consist of two basic components, i.e., VQ tokenizers and generative transformers. Prior research focuses on improving the reconstruction fidelity of VQ tokenizers but rarely examines how the improvement in reconstruction affects the generation ability of generative transformers. In this paper, we surprisingly find that improving the reconstruction fidelity of VQ tokenizers does not necessarily improve the generation. Instead, learning to compress semantic features within VQ tokenizers significantly improves generative transformers' ability to capture textures and structures. We thus highlight two competing objectives of VQ tokenizers for image synthesis: semantic compression and details preservation. Different from previous work that only pursues better details preservation, we propose Semantic-Quantized GAN (SeQ-GAN) with two learning phases to balance the two objectives. In the first phase, we propose a semantic-enhanced perceptual loss for better semantic compression. In the second phase, we fix the encoder and codebook, but enhance and finetune the decoder to achieve better details preservation. The proposed SeQ-GAN greatly improves VQ-based generative models and surpasses the GAN and Diffusion Models on both unconditional and conditional image generation. Our SeQ-GAN (364M) achieves Frechet Inception Distance (FID) of 6.25 and Inception Score (IS) of 140.9 on 256x256 ImageNet generation, a remarkable improvement over VIT-VQGAN (714M), which obtains 11.2 FID and 97.2 IS.	['Mike Zheng Shou', 'XiaoHu Qie', 'Ying Shan', 'Yixiao Ge', 'Xintao Wang', 'YuChao Gu']	2022-12-06	null	null	null	null	['conditional-image-generation']	['computer-vision']	[ 3.58518273e-01 1.81023985e-01 -9.93053019e-02 -2.24417225e-01 -8.31070781e-01 -3.54793370e-01 7.40762830e-01 -3.90200078e-01 2.65032314e-02 7.18793273e-01 5.66641033e-01 -2.34331876e-01 1.03926353e-01 -1.19018197e+00 -8.34316194e-01 -8.22274148e-01 2.64954001e-01 7.90493861e-02 1.18810050e-01 -2.06968099e-01 1.65968239e-01 1.48603931e-01 -1.37675858e+00 4.26903307e-01 1.02830660e+00 1.00180566e+00 4.73728389e-01 8.79064083e-01 -2.59345561e-01 1.12101746e+00 -6.95946217e-01 -5.84639132e-01 1.71911359e-01 -9.39572632e-01 -6.05157495e-01 6.86588418e-03 1.58741161e-01 -4.32972670e-01 -4.51966017e-01 1.05267501e+00 5.54397941e-01 -1.11067660e-01 7.89387882e-01 -1.15366316e+00 -1.01430225e+00 8.07155669e-01 -5.39498687e-01 8.92550424e-02 5.05944900e-02 2.96675295e-01 1.07727695e+00 -6.70070291e-01 5.14774263e-01 1.45428371e+00 6.25799656e-01 6.62796080e-01 -1.03573132e+00 -8.84141743e-01 -2.29793414e-01 6.88452693e-03 -1.38459694e+00 -4.68936294e-01 7.41210043e-01 -1.94576412e-01 8.44994783e-01 2.29035020e-01 6.91651642e-01 9.55306172e-01 4.01392549e-01 6.48469627e-01 1.03101873e+00 -2.74754077e-01 3.18909794e-01 -1.12874478e-01 -6.66007400e-01 5.88184059e-01 -4.27257195e-02 3.33101988e-01 -5.60319126e-01 2.89301634e-01 1.21455884e+00 -1.32295266e-01 -1.62855685e-01 1.02188073e-01 -1.01659334e+00 1.06279409e+00 5.44330835e-01 1.95031822e-01 -4.46492404e-01 7.21725523e-01 2.08302721e-01 2.10077569e-01 2.01605663e-01 5.16012609e-01 -3.72974202e-02 -3.18290234e-01 -1.25060081e+00 8.73050466e-02 2.22816393e-01 1.06527328e+00 7.86823392e-01 7.46959448e-01 -5.14411509e-01 8.87300372e-01 2.91590005e-01 8.00658464e-01 5.45603991e-01 -1.13414836e+00 2.38441646e-01 2.23279849e-01 -3.84486109e-01 -9.33737099e-01 4.05222952e-01 -7.30118334e-01 -1.11215794e+00 2.86445208e-02 -1.49583921e-01 -4.31315079e-02 -1.21838701e+00 1.79088914e+00 -1.43669382e-01 1.50561258e-01 1.05158031e-01 6.71310484e-01 8.95358920e-01 1.00618899e+00 9.30184945e-02 4.68653068e-02 1.06372726e+00 -9.69632328e-01 -7.95257568e-01 2.22615171e-02 3.46370637e-01 -9.69035804e-01 1.08429790e+00 2.31408700e-01 -1.37897027e+00 -7.94005036e-01 -1.14983058e+00 1.20366395e-01 6.70942143e-02 -1.13987029e-01 4.57854629e-01 7.91120768e-01 -1.26781988e+00 4.84406918e-01 -8.03255975e-01 -4.87192832e-02 6.74750984e-01 1.71156541e-01 2.19747294e-02 -2.45612174e-01 -1.18908298e+00 4.09706533e-01 2.68478096e-01 -4.83802676e-01 -1.42950010e+00 -8.08514714e-01 -7.37048924e-01 2.14725435e-01 4.62808013e-02 -9.96862948e-01 1.03315377e+00 -1.08757532e+00 -1.70566797e+00 3.86347383e-01 -7.66624063e-02 -5.17396212e-01 4.28628147e-01 1.02695353e-01 -2.19635159e-01 2.62530625e-01 5.90401329e-02 1.15829802e+00 1.12920928e+00 -1.44662905e+00 -6.59329712e-01 4.78382148e-02 2.63012256e-02 9.63616595e-02 -3.26570690e-01 -4.46678877e-01 -5.48976123e-01 -1.13316727e+00 -8.73841792e-02 -6.07391536e-01 -8.74975920e-02 -2.25242719e-01 -4.93115246e-01 8.74489769e-02 8.65418673e-01 -6.86553359e-01 1.07749283e+00 -2.18063855e+00 -1.30149499e-01 6.90393448e-02 2.19839036e-01 3.40046138e-01 -4.55317885e-01 4.23700333e-01 1.41220897e-01 3.10500443e-01 -3.20262671e-01 -4.51414764e-01 -1.25876232e-03 5.57698190e-01 -4.26014692e-01 1.50371371e-02 4.17851359e-01 1.31473231e+00 -7.97562301e-01 -5.30981183e-01 3.36134970e-01 9.13722932e-01 -1.05871499e+00 1.42504483e-01 -1.10041939e-01 3.86286497e-01 -3.17716032e-01 5.56069195e-01 8.04178655e-01 -1.92464575e-01 6.90443581e-03 -1.71776682e-01 1.57833636e-01 1.96212545e-01 -7.85831690e-01 1.67960286e+00 -6.80990875e-01 5.67107677e-01 -2.88525999e-01 -1.02246296e+00 1.07483864e+00 2.95945019e-01 4.17012304e-01 -1.14123857e+00 2.84052137e-02 5.87926619e-02 -2.47897521e-01 5.97179346e-02 7.77755558e-01 -4.80349272e-01 1.70746427e-02 1.11892216e-01 4.32257384e-01 -3.35308403e-01 -1.90608472e-01 3.86607528e-01 9.97611463e-01 -2.17637159e-02 8.82695019e-02 -1.57640606e-01 2.60525942e-01 -4.35610533e-01 5.14603138e-01 7.09918380e-01 1.16815276e-01 8.32818091e-01 5.55320621e-01 1.83999538e-01 -1.53383946e+00 -1.43139231e+00 1.40832365e-01 6.66867554e-01 1.98254943e-01 -6.44449234e-01 -8.95018637e-01 -4.49208468e-01 -3.02372843e-01 8.99398744e-01 -5.20941615e-01 -5.32564163e-01 -3.62538278e-01 -5.77217996e-01 7.70636141e-01 7.05773056e-01 9.58353519e-01 -8.38975370e-01 -4.81744677e-01 1.55708581e-01 -3.10650617e-01 -9.00805175e-01 -6.17779434e-01 -8.96243080e-02 -8.92420650e-01 -4.72794980e-01 -8.03052425e-01 -6.35740876e-01 4.63947594e-01 4.70146388e-02 1.22982907e+00 -4.22035791e-02 5.28955348e-02 1.98823541e-01 -7.67414927e-01 -1.91589475e-01 -7.62585938e-01 3.00563127e-02 -4.17416960e-01 -9.87976789e-02 -1.86631769e-01 -8.09259295e-01 -7.85790563e-01 2.14309141e-01 -1.20596874e+00 1.82780713e-01 8.28542531e-01 9.79126155e-01 7.68036902e-01 3.86254638e-01 5.50833881e-01 -5.93800306e-01 4.76231784e-01 -2.12898776e-01 -1.94226995e-01 7.36344829e-02 -7.92628407e-01 2.47606114e-01 8.62143874e-01 -2.77303785e-01 -9.55060244e-01 -2.46756047e-01 -6.08045638e-01 -6.67099774e-01 2.30390444e-01 2.90021539e-01 -3.09117109e-01 6.05260134e-02 3.86692703e-01 6.55747712e-01 -1.39265180e-01 -1.46315008e-01 5.76667130e-01 3.70807707e-01 6.32388532e-01 -4.64973927e-01 7.53978193e-01 3.47517788e-01 -1.15040570e-01 -8.00282240e-01 -2.96554744e-01 6.54491633e-02 2.86607686e-02 -1.01090997e-01 9.47161674e-01 -9.72043514e-01 -4.22287583e-01 4.15772706e-01 -8.21001053e-01 -4.95739818e-01 -7.82488167e-01 2.57199317e-01 -8.14724386e-01 2.17039824e-01 -5.82262576e-01 -5.95957458e-01 -3.84732574e-01 -1.17881691e+00 1.07622409e+00 1.89700183e-02 7.26025999e-02 -9.38174784e-01 -1.45832896e-01 2.15881303e-01 7.65305936e-01 2.82962084e-01 8.96045506e-01 8.64605010e-02 -7.13454008e-01 2.41764233e-01 -3.04550856e-01 5.81376791e-01 2.68370628e-01 -2.14934811e-01 -8.88181388e-01 -3.32987398e-01 -9.38196182e-02 -1.69789448e-01 1.20442855e+00 6.00327492e-01 1.17549789e+00 -6.03186309e-01 4.50741164e-02 1.05224216e+00 1.65699303e+00 5.88864148e-01 1.29250431e+00 -1.53425010e-02 7.06509888e-01 6.48245513e-02 2.23132104e-01 6.07023239e-01 2.91917890e-01 5.48536956e-01 5.75816274e-01 -2.61585534e-01 -8.73285830e-01 -8.47496629e-01 5.07746577e-01 1.08716691e+00 -2.28092581e-01 -5.45508623e-01 -5.06091118e-01 6.09867871e-01 -1.34334743e+00 -9.55897033e-01 3.26029986e-01 1.83432138e+00 8.59997392e-01 8.16831216e-02 -5.08631952e-02 2.43771583e-01 5.26497126e-01 2.71430582e-01 -4.88135606e-01 -3.97802889e-01 -2.89230168e-01 6.48679256e-01 5.72020888e-01 6.87800884e-01 -6.44487739e-01 1.02291763e+00 6.24966574e+00 1.34424400e+00 -1.32757056e+00 3.17281812e-01 8.54440272e-01 6.02819072e-03 -8.91267955e-01 -2.14125738e-02 -5.01023412e-01 7.27703214e-01 8.97888899e-01 -4.25060652e-02 5.44710815e-01 6.33920074e-01 -2.12255958e-02 2.56879538e-01 -6.77875042e-01 9.82631803e-01 1.32272597e-02 -1.59998429e+00 6.43547177e-01 2.10611299e-01 1.00985241e+00 -3.25012892e-01 5.95038950e-01 2.57384658e-01 4.91987199e-01 -1.34210598e+00 1.10210931e+00 4.73197252e-01 1.30645251e+00 -1.02725184e+00 6.99920475e-01 -5.45704388e-04 -1.25984693e+00 -1.49587486e-02 -4.30688709e-01 2.89458424e-01 2.78974950e-01 6.90000176e-01 -7.16963351e-01 5.65023839e-01 6.24419451e-01 8.05407584e-01 -3.71425122e-01 5.27924597e-01 -3.52691591e-01 8.77366245e-01 1.04262710e-01 2.02908546e-01 4.40084845e-01 -7.57044330e-02 4.67140615e-01 1.07306349e+00 6.54137552e-01 9.74157006e-02 -1.16717532e-01 1.11422360e+00 -1.94158271e-01 -1.90007314e-01 -3.75763625e-01 -1.55979216e-01 4.67333257e-01 7.04712927e-01 -6.77422762e-01 -4.25764441e-01 -1.03686854e-01 1.18834674e+00 -2.40395829e-01 3.34743142e-01 -1.16183078e+00 -4.04398292e-01 6.20152414e-01 3.06683183e-01 8.51545393e-01 -2.01539740e-01 -3.43523055e-01 -1.05994308e+00 -3.96941453e-01 -9.62138832e-01 8.95785983e-04 -7.40509748e-01 -9.25110340e-01 6.55206859e-01 -1.45307481e-01 -9.43337083e-01 -2.55175382e-01 -4.02453095e-02 -4.54094499e-01 8.27490389e-01 -1.58703697e+00 -1.52729189e+00 -2.80069500e-01 6.30031049e-01 7.09080517e-01 -2.39967763e-01 5.92385828e-01 3.35869849e-01 -1.89105943e-01 1.07040858e+00 2.94370681e-01 -2.79322192e-02 3.87894809e-01 -1.05427551e+00 5.10246098e-01 9.93517578e-01 1.29978374e-01 4.50689852e-01 5.55090427e-01 -6.26125753e-01 -1.30492604e+00 -1.54630983e+00 5.92591405e-01 -6.78125620e-02 1.25219658e-01 -2.44873196e-01 -5.58336854e-01 5.03587544e-01 4.08583492e-01 -3.11365783e-01 3.66794348e-01 -4.68146443e-01 -3.94852370e-01 -2.15341568e-01 -1.19683576e+00 3.85523736e-01 1.03826976e+00 -5.04598796e-01 -1.04494072e-01 -1.51725516e-01 9.45644915e-01 -9.85171795e-02 -9.14249659e-01 3.95094782e-01 4.18665498e-01 -1.13699234e+00 1.13972449e+00 -6.14267513e-02 1.06542826e+00 -1.17399111e-01 -6.66691005e-01 -1.38859868e+00 -5.79840124e-01 -6.03047013e-01 -2.93656420e-02 1.36455607e+00 1.87655672e-01 -5.81729591e-01 8.25254560e-01 -3.37936521e-01 -3.69362563e-01 -8.58178437e-01 -7.07326472e-01 -9.33940470e-01 3.98791075e-01 -4.14675474e-01 9.57643807e-01 7.49716878e-01 -6.52442634e-01 2.74655372e-01 -5.87104738e-01 -1.31318182e-01 6.75146639e-01 -6.91871867e-02 5.69261193e-01 -5.22724926e-01 -4.69630450e-01 -5.60181856e-01 -4.30991113e-01 -1.30105960e+00 -1.85742006e-01 -9.94807243e-01 -1.06650971e-01 -1.63038063e+00 3.10671300e-01 -6.76562309e-01 -2.20328614e-01 3.39766920e-01 -4.63984944e-02 6.12856507e-01 3.37154806e-01 2.08264142e-01 -2.87116289e-01 1.02171981e+00 1.62585545e+00 -3.01775545e-01 -4.36569415e-02 -4.49318796e-01 -9.22877192e-01 3.08766156e-01 7.26268947e-01 -2.93061286e-01 -8.77771854e-01 -5.64062178e-01 1.08162321e-01 1.42821744e-01 4.36156034e-01 -1.14243376e+00 -7.78831691e-02 -9.94283482e-02 4.18984979e-01 -4.79565710e-01 4.22089726e-01 -3.03860754e-01 5.36897361e-01 6.77356243e-01 -2.98393160e-01 -7.77779147e-02 -3.25652733e-02 5.14801443e-01 -4.63142276e-01 3.47517841e-02 1.10429502e+00 -2.88656175e-01 -6.01056635e-01 4.18124706e-01 -2.91439325e-01 1.24425150e-01 8.14782858e-01 -4.18779552e-01 -2.31185913e-01 -7.70116627e-01 -3.41054857e-01 -2.56856620e-01 6.55772567e-01 2.59018987e-01 1.09518683e+00 -1.53894687e+00 -8.90095890e-01 4.56600636e-01 -2.06340149e-01 -1.34021372e-01 3.35372567e-01 4.37638074e-01 -6.03447914e-01 4.52031046e-01 -3.75839919e-01 -6.14495516e-01 -9.13704336e-01 4.14524108e-01 1.28133118e-01 -3.70050102e-01 -5.76222956e-01 1.14529467e+00 5.86980939e-01 -5.22629172e-02 -1.55907512e-01 -2.90110916e-01 6.88311681e-02 -2.80356377e-01 3.21263224e-01 2.97424704e-01 -8.95973966e-02 -5.72710991e-01 -8.17843303e-02 5.33997774e-01 7.49478415e-02 -2.26536900e-01 1.32121909e+00 -1.93065312e-02 8.26473162e-02 -5.49339131e-02 1.27335227e+00 4.39735642e-03 -1.39601588e+00 -2.47741323e-02 -7.87029982e-01 -6.42620385e-01 2.51942277e-01 -7.30123878e-01 -1.60518360e+00 1.02526248e+00 5.44505179e-01 9.08564869e-03 1.54468513e+00 1.14734761e-01 1.30678177e+00 -4.40827549e-01 3.17205459e-01 -6.94144011e-01 6.56141818e-01 4.76539880e-01 8.32806885e-01 -6.49261236e-01 -1.36237532e-01 -4.63083267e-01 -8.44741046e-01 6.30632818e-01 3.35032791e-01 -2.09168032e-01 4.62120622e-01 4.06009167e-01 -2.50442058e-01 -1.32347494e-01 -7.79854238e-01 2.21666973e-02 2.13156506e-01 8.12302351e-01 3.08153689e-01 1.91495955e-01 -1.00639202e-01 4.63998914e-01 -7.36386597e-01 -1.69014797e-01 3.83708000e-01 6.37605965e-01 -3.71057123e-01 -1.11607397e+00 -2.43106902e-01 5.08350790e-01 -4.77718890e-01 -3.42634469e-01 -1.12603597e-01 4.29506987e-01 4.53506917e-01 9.09323215e-01 2.59271711e-01 -8.41592073e-01 1.31289005e-01 -3.61563861e-01 5.52252352e-01 -2.49573022e-01 -3.48887295e-01 2.74100840e-01 -2.17110902e-01 -5.19721389e-01 -4.83878821e-01 -4.38082725e-01 -1.04012465e+00 -6.66792691e-01 -3.18653852e-01 1.75042540e-01 5.79116344e-01 5.83389521e-01 3.99674743e-01 1.07270002e+00 7.73118556e-01 -2.54261762e-01 -3.18699628e-01 -7.24975348e-01 -6.73856497e-01 3.26104045e-01 2.48837307e-01 -4.44375306e-01 -1.89119741e-01 4.47151095e-01]	[11.515533447265625, -0.6174464225769043]
eeb3903a-e9a0-447a-92c8-4208fbab6e75	text-based-sentiment-analysis-and-music	1810.03031	null	http://arxiv.org/abs/1810.03031v1	http://arxiv.org/pdf/1810.03031v1.pdf	Text-based Sentiment Analysis and Music Emotion Recognition	Sentiment polarity of tweets, blog posts or product reviews has become highly attractive and is utilized in recommender systems, market predictions, business intelligence and more. Deep learning techniques are becoming top performers on analyzing such texts. There are however several problems that need to be solved for efficient use of deep neural networks on text mining and text polarity analysis. First, deep neural networks need to be fed with data sets that are big in size as well as properly labeled. Second, there are various uncertainties regarding the use of word embedding vectors: should they be generated from the same data set that is used to train the model or it is better to source them from big and popular collections? Third, to simplify model creation it is convenient to have generic neural network architectures that are effective and can adapt to various texts, encapsulating much of design complexity. This thesis addresses the above problems to provide methodological and practical insights for utilizing neural networks on sentiment analysis of texts and achieving state of the art results. Regarding the first problem, the effectiveness of various crowdsourcing alternatives is explored and two medium-sized and emotion-labeled song data sets are created utilizing social tags. To address the second problem, a series of experiments with large text collections of various contents and domains were conducted, trying word embeddings of various parameters. Regarding the third problem, a series of experiments involving convolution and max-pooling neural layers were conducted. Combining convolutions of words, bigrams, and trigrams with regional max-pooling layers in a couple of stacks produced the best results. The derived architecture achieves competitive performance on sentiment polarity analysis of movie, business and product reviews.	['Erion Çano']	2018-10-06	null	null	null	null	['music-emotion-recognition']	['music']	[-1.20324880e-01 -6.23658262e-02 -1.49553716e-01 -6.16936862e-01 -9.30848271e-02 -5.24076521e-01 4.97711748e-01 4.53980744e-01 -6.87583208e-01 5.75352013e-01 3.77531886e-01 -1.59169674e-01 1.41170949e-01 -8.80604029e-01 -2.99400747e-01 -5.51015377e-01 1.76377445e-01 3.76977146e-01 8.58217925e-02 -8.46937597e-01 3.89281571e-01 3.13792586e-01 -1.72633183e+00 4.74617600e-01 4.19397563e-01 1.22899759e+00 1.59296747e-02 6.05031371e-01 -6.05565369e-01 7.76532531e-01 -6.67595208e-01 -7.99838960e-01 2.63617754e-01 1.36421593e-02 -5.54130673e-01 -1.45725831e-01 -1.16024921e-02 -1.32163065e-02 3.41307551e-01 8.34829807e-01 8.56531441e-01 2.08270565e-01 5.58383286e-01 -9.51480448e-01 -1.05257583e+00 7.01503575e-01 -4.55793709e-01 3.67243201e-01 2.62272537e-01 -3.74036878e-01 1.13777900e+00 -9.97442544e-01 5.07106483e-01 8.49012136e-01 6.63238108e-01 4.08847958e-01 -8.29581618e-01 -4.91021246e-01 8.36571679e-02 -6.21929690e-02 -7.83525169e-01 -1.54072016e-01 7.80875146e-01 -5.34496069e-01 1.28701937e+00 4.86797048e-03 7.37691998e-01 1.10847890e+00 2.50181973e-01 5.19461393e-01 9.85185325e-01 -5.25653660e-01 2.89587528e-01 1.03783822e+00 4.83754396e-01 1.51367441e-01 2.11422220e-01 -3.62351358e-01 -5.18361926e-01 -1.08260989e-01 -5.94552271e-02 2.40717437e-02 1.44128278e-01 -8.82927105e-02 -8.37954640e-01 1.36252117e+00 1.14232972e-01 7.03613877e-01 -5.25340199e-01 -2.93490976e-01 7.76135981e-01 5.00966668e-01 9.19891894e-01 7.88891733e-01 -7.71602333e-01 -7.10547715e-02 -6.85271740e-01 2.99617082e-01 1.11577189e+00 5.31845391e-01 6.00911915e-01 3.11648697e-01 2.18495041e-01 1.19158745e+00 2.38835245e-01 4.67527896e-01 1.16163647e+00 -9.84471589e-02 4.23314840e-01 8.30141604e-01 -5.39921373e-02 -1.45914328e+00 -7.16069996e-01 -2.36629695e-01 -5.11840463e-01 5.43104410e-02 2.11636662e-01 -5.58551133e-01 -6.87905252e-01 1.28932703e+00 3.70325536e-01 -3.57251018e-01 3.55739653e-01 7.48021722e-01 1.17936575e+00 8.98905158e-01 1.23034358e-01 2.65491139e-02 1.59088206e+00 -7.73453653e-01 -6.63434744e-01 -5.01642406e-01 6.87776744e-01 -9.26947832e-01 1.09094882e+00 3.82643551e-01 -9.16844726e-01 -6.29178405e-01 -1.20493138e+00 -1.80278316e-01 -1.28847063e+00 1.46247983e-01 7.01388240e-01 8.66505444e-01 -1.04089594e+00 5.71973562e-01 -2.52395660e-01 -3.90959144e-01 1.02154665e-01 7.50684798e-01 -4.22727793e-01 3.50565821e-01 -1.48045421e+00 1.02022636e+00 1.73822641e-01 2.89120823e-01 -1.00051880e-01 -3.41202855e-01 -8.73203039e-01 6.96238279e-02 -1.59292102e-01 -1.76996633e-01 1.03223014e+00 -1.59823024e+00 -1.53336322e+00 8.97843838e-01 -2.57725101e-02 -4.81777728e-01 -1.00300476e-01 -8.73713195e-02 -7.29690015e-01 -1.74044788e-01 5.57431765e-02 3.80180955e-01 7.07532346e-01 -7.72727668e-01 -7.25542128e-01 -3.97722572e-01 1.69452965e-01 1.65002897e-01 -1.10933638e+00 3.87804389e-01 2.02142633e-02 -5.54007292e-01 -2.95972675e-01 -1.08764052e+00 -2.46523425e-01 -7.68463492e-01 -6.19084165e-02 -3.76664907e-01 5.88357329e-01 -4.27626818e-01 1.06070244e+00 -2.00860405e+00 -2.21194848e-01 2.53290504e-01 4.01980728e-02 4.92838860e-01 -2.17577502e-01 4.58574444e-01 -2.75265992e-01 3.29386622e-01 2.81461120e-01 -2.65539289e-01 4.41489071e-02 8.07965454e-03 -4.32143688e-01 2.90267438e-01 2.73030668e-01 7.74409711e-01 -5.76926410e-01 -1.24638468e-01 7.12678805e-02 5.35764992e-01 -4.82272506e-01 -8.76111239e-02 -5.91732264e-02 -9.59784761e-02 -3.29869181e-01 3.61832619e-01 3.43494177e-01 -1.23283267e-01 8.53502005e-02 -1.21493392e-01 -6.00700192e-02 3.71548831e-01 -1.11874449e+00 8.64351630e-01 -5.28646111e-01 9.59105670e-01 -1.93580657e-01 -1.10306931e+00 1.27447140e+00 3.84389102e-01 5.87006807e-01 -8.36523890e-01 6.61777079e-01 3.15084845e-01 5.77690974e-02 -7.08838522e-01 1.13832867e+00 -5.91895044e-01 -1.92534670e-01 5.43632269e-01 2.08751976e-01 -1.18854254e-01 4.06589448e-01 -1.93197176e-01 6.53237224e-01 -3.83091420e-01 -1.34265730e-02 -3.10761988e-01 6.23639166e-01 1.82960927e-01 2.81783164e-01 1.05871707e-01 -5.83701208e-02 5.18338919e-01 7.25109994e-01 -8.49647820e-01 -9.71507072e-01 -2.53854811e-01 -1.01248957e-01 1.61017704e+00 -6.14413060e-02 -2.30900005e-01 -5.05074978e-01 -2.69710153e-01 -1.92996934e-01 4.64123130e-01 -8.81118894e-01 1.39522314e-01 -3.57454419e-01 -1.15634620e+00 2.22072080e-01 5.43904901e-01 5.20352926e-03 -1.49301386e+00 -4.47887361e-01 3.63920689e-01 1.45811275e-01 -1.03124046e+00 1.35130942e-01 5.53811193e-01 -6.68381035e-01 -7.88892031e-01 -6.89793050e-01 -1.08323979e+00 3.40988755e-01 2.91540682e-01 1.19631541e+00 -2.33699635e-01 1.70894861e-01 6.59319609e-02 -6.13613665e-01 -1.01452374e+00 -2.51253456e-01 4.71006811e-01 1.31473675e-01 1.95544347e-01 1.11416364e+00 -5.15788794e-01 -4.33618993e-01 1.68507531e-01 -1.00644982e+00 -4.98534739e-01 2.29657337e-01 6.69409156e-01 3.05256039e-01 6.29698858e-03 9.51419413e-01 -1.02921820e+00 1.46670115e+00 -6.19208395e-01 -4.14780647e-01 -1.27204373e-01 -6.03111506e-01 -2.80678779e-01 9.92212236e-01 -4.88865584e-01 -8.54377687e-01 -2.28135467e-01 -4.11075324e-01 8.77007097e-02 -2.80023180e-02 7.28586555e-01 3.84841204e-01 1.79732978e-01 9.91780281e-01 -1.45524308e-01 1.76402535e-02 -8.75099748e-02 2.50521600e-01 9.47240889e-01 -2.36393332e-01 -1.05467357e-01 1.81353211e-01 4.25443590e-01 -5.34602046e-01 -1.09931469e+00 -7.69722819e-01 -6.37656391e-01 -3.49261969e-01 -7.14218840e-02 1.12546825e+00 -9.28693891e-01 -4.85542715e-01 3.63537878e-01 -9.80192959e-01 -3.98880057e-02 -3.78478408e-01 4.40857857e-01 1.95473880e-02 1.01716062e-02 -6.12068117e-01 -6.47016346e-01 -6.48160338e-01 -1.14898849e+00 7.10998654e-01 4.93930131e-01 -5.30688882e-01 -1.19170332e+00 3.97989661e-01 3.40490997e-01 6.67082965e-01 2.16119178e-02 7.49260843e-01 -1.48264706e+00 3.84784818e-01 -9.04739380e-01 1.59875434e-02 8.02703738e-01 -5.77362403e-02 1.63938835e-01 -1.21004832e+00 1.85166933e-02 1.53821185e-01 -5.65260887e-01 4.92237955e-01 4.78892565e-01 7.04695642e-01 -1.40684798e-01 7.88473114e-02 9.47135612e-02 1.25697803e+00 2.96137214e-01 5.87847888e-01 6.09238863e-01 5.21105170e-01 1.08508885e+00 3.67316902e-01 3.34709108e-01 5.29455364e-01 1.60959154e-01 2.12259635e-01 -8.11603144e-02 6.23805583e-01 2.51896024e-01 4.42661315e-01 1.12700295e+00 -7.54833892e-02 -2.93511093e-01 -7.26524711e-01 8.01302552e-01 -1.47055805e+00 -9.78257000e-01 -1.92319512e-01 1.85160971e+00 5.15201569e-01 3.18339854e-01 1.87892795e-01 3.15361947e-01 5.46187401e-01 3.48497301e-01 -2.17496231e-01 -1.26159704e+00 -2.50929296e-01 3.48958343e-01 3.36758763e-01 2.03580886e-01 -1.19633901e+00 7.01638103e-01 5.81545067e+00 5.15524983e-01 -1.70494032e+00 1.93295747e-01 7.51668751e-01 -2.95657098e-01 -3.32916498e-01 -3.97449702e-01 -9.66882825e-01 4.49966431e-01 1.30903482e+00 1.20362956e-02 1.71744511e-01 1.16001880e+00 1.60055622e-01 -1.68056771e-01 -7.47543514e-01 7.93134093e-01 2.90591240e-01 -1.38792872e+00 -3.05006325e-01 -2.49569323e-02 1.01160264e+00 4.67686296e-01 1.56939998e-01 5.47021210e-01 1.71387166e-01 -9.44700062e-01 5.23947775e-01 5.20754047e-02 1.73535094e-01 -6.97513700e-01 1.38180315e+00 1.73291378e-02 -8.10540318e-01 -3.33704799e-01 -6.83354497e-01 -2.87098378e-01 9.01903883e-02 6.58262491e-01 -7.81039059e-01 8.88478905e-02 7.73498654e-01 5.88934481e-01 -5.27231395e-01 5.02628207e-01 3.33038270e-02 4.53591526e-01 -3.47056299e-01 -8.56770039e-01 4.77922678e-01 -3.33090067e-01 6.84235319e-02 1.39337444e+00 1.78977221e-01 -2.89439797e-01 -2.47359902e-01 3.76067430e-01 -1.24624103e-01 7.33179569e-01 -7.84853697e-01 -3.83284450e-01 6.36033118e-02 1.76502669e+00 -9.52204406e-01 -2.56698877e-01 -5.40403426e-01 4.64510918e-01 1.61277577e-01 2.23304048e-01 -7.17636526e-01 -5.48695266e-01 5.09278536e-01 1.91305235e-01 2.97560394e-01 1.45929173e-01 -4.71290529e-01 -9.97673750e-01 1.10496975e-01 -1.00790179e+00 8.55166987e-02 -6.71999514e-01 -1.37621605e+00 1.07264113e+00 -3.19058716e-01 -1.12376380e+00 -3.50687057e-01 -1.01508117e+00 -6.94755316e-01 9.98035729e-01 -1.50259817e+00 -9.57226515e-01 -7.03776926e-02 4.62161958e-01 4.54290003e-01 -5.61600566e-01 9.25288260e-01 5.20400465e-01 -4.20661867e-01 2.98337251e-01 1.68289050e-01 1.57854527e-01 5.82249939e-01 -1.16433001e+00 1.24420613e-01 4.33093339e-01 1.14591517e-01 6.26951814e-01 5.94101012e-01 -2.47718886e-01 -1.12599635e+00 -6.86279118e-01 1.27830231e+00 -4.33566689e-01 9.28601682e-01 -4.59618419e-01 -6.97312236e-01 2.20369399e-01 4.18685764e-01 -1.76537305e-01 1.21136868e+00 2.70932525e-01 2.56948769e-02 -3.43363941e-01 -9.55072701e-01 5.63330412e-01 2.46685650e-02 -5.83495796e-01 -6.48712993e-01 4.12026256e-01 5.11727273e-01 -9.48828086e-02 -8.05021346e-01 1.30519569e-01 6.86290145e-01 -1.08477688e+00 6.52312994e-01 -6.97915137e-01 9.55188036e-01 1.32517656e-02 -1.98525101e-01 -1.35838211e+00 -1.11773508e-02 -1.96528703e-01 4.53726828e-01 1.21945000e+00 9.26465750e-01 -8.32813263e-01 8.84883761e-01 9.20520246e-01 1.20596461e-01 -1.06236446e+00 -4.56201702e-01 -1.50520757e-01 2.61927217e-01 -4.93074387e-01 5.47399700e-01 9.91539299e-01 1.85268819e-01 8.69440019e-01 -2.81982034e-01 -3.31173152e-01 -3.98217171e-01 7.74649978e-02 7.85734892e-01 -1.27428579e+00 2.77891271e-02 -4.99531269e-01 -3.35986733e-01 -4.99975860e-01 1.72604397e-01 -6.46246493e-01 -1.52988181e-01 -1.58656454e+00 -3.28610480e-01 -2.48597369e-01 -2.69218296e-01 2.11768687e-01 1.39963344e-01 4.23458368e-01 1.36789978e-01 -3.67655866e-02 -1.96588203e-01 3.57771516e-01 9.21132207e-01 -1.06127523e-01 -4.00707603e-01 1.12635540e-02 -1.06053150e+00 8.93434227e-01 1.02162778e+00 -4.24096853e-01 -4.39312667e-01 -2.65844077e-01 1.21107495e+00 -3.44402343e-01 -2.72310168e-01 -6.86292350e-01 1.65529445e-01 2.20646188e-01 3.98847789e-01 -4.53523844e-01 4.91955727e-01 -9.38601375e-01 -2.99135923e-01 -8.79317522e-02 -4.29226965e-01 4.82788086e-01 3.02066565e-01 3.95287871e-01 -4.60889190e-01 -6.20224953e-01 5.30887425e-01 -1.34371310e-01 -6.92883074e-01 -4.48828451e-02 -6.75728738e-01 -3.95162441e-02 9.03722107e-01 -4.19486195e-01 -5.17632402e-02 -5.04622757e-01 -5.82066119e-01 4.54344079e-02 1.12675488e-01 7.91774631e-01 3.21815908e-01 -1.01522565e+00 -4.90178168e-01 2.33390480e-01 6.49193972e-02 -2.75889307e-01 1.11391336e-01 7.87278414e-01 -5.91714978e-01 3.51153404e-01 -1.73221380e-01 -1.79115906e-01 -1.11453819e+00 4.99158233e-01 1.19899459e-01 -4.71355289e-01 6.24138750e-02 8.52892756e-01 -1.18471295e-01 -7.56173134e-01 -8.59607663e-03 -3.58311415e-01 -9.70392585e-01 8.58035505e-01 6.18628561e-01 1.23047344e-01 4.03822929e-01 -7.75638044e-01 -1.72945336e-01 5.24839163e-01 -1.02741838e-01 -5.36662433e-03 1.67893374e+00 3.28922942e-02 -2.54177421e-01 5.53741276e-01 1.28909039e+00 2.81544775e-01 -4.43978608e-01 5.34649715e-02 1.07663572e-01 8.71689692e-02 2.01540440e-02 -4.81970668e-01 -1.23871768e+00 9.38796461e-01 5.21926761e-01 9.95733321e-01 8.61556828e-01 -3.75510782e-01 6.79461658e-01 5.64093173e-01 -1.05527751e-01 -1.69122934e+00 -7.74938334e-03 7.51142621e-01 6.80717766e-01 -1.46388495e+00 -2.04399023e-02 2.73538321e-01 -1.08861220e+00 1.30595517e+00 4.19809163e-01 -4.51428980e-01 1.01880443e+00 2.05045193e-01 3.90716970e-01 -5.19768953e-01 -4.99631554e-01 -5.37010394e-02 3.01176846e-01 3.60180438e-01 8.70203316e-01 -2.81840432e-02 -6.13979757e-01 9.69982028e-01 -4.66480792e-01 6.72825426e-03 6.19366586e-01 7.40779459e-01 -4.31756049e-01 -1.04570138e+00 -2.47247472e-01 7.28035748e-01 -1.12582636e+00 -2.31370136e-01 -4.94338810e-01 6.76832020e-01 2.19782963e-01 1.22185063e+00 -3.61487381e-02 -6.55787289e-01 3.27652484e-01 1.35270506e-01 -3.17501009e-01 -8.00605655e-01 -1.13766503e+00 -4.84359860e-02 4.68985260e-01 -6.34410158e-02 -7.70331979e-01 -2.72105545e-01 -9.83918369e-01 -2.93555439e-01 -6.17119789e-01 2.45937392e-01 1.27498412e+00 9.25607860e-01 3.79756778e-01 3.57455999e-01 6.80391252e-01 -1.04222786e+00 -4.03449953e-01 -1.16006935e+00 -7.38064289e-01 5.04014730e-01 -4.41671312e-02 -3.94585729e-01 -3.64959687e-01 -8.48621654e-04]	[11.138564109802246, 7.050936222076416]
dd948ead-2b61-43e7-a14a-53e31fb6ba5b	classification-of-goods-using-text	2111.01663	null	https://arxiv.org/abs/2111.01663v1	https://arxiv.org/pdf/2111.01663v1.pdf	Classification of Goods Using Text Descriptions With Sentences Retrieval	The task of assigning and validating internationally accepted commodity code (HS code) to traded goods is one of the critical functions at the customs office. This decision is crucial to importers and exporters, as it determines the tariff rate. However, similar to court decisions made by judges, the task can be non-trivial even for experienced customs officers. The current paper proposes a deep learning model to assist this seemingly challenging HS code classification. Together with Korea Customs Service, we built a decision model based on KoELECTRA that suggests the most likely heading and subheadings (i.e., the first four and six digits) of the HS code. Evaluation on 129,084 past cases shows that the top-3 suggestions made by our model have an accuracy of 95.5% in classifying 265 subheadings. This promising result implies algorithms may reduce the time and effort taken by customs officers substantially by assisting the HS code classification task.	['Heeja Kim', 'Minsoo Song', 'Sungdae Ji', 'Yeonsoo Choi', 'Suyoung Yang', 'Soyeon Jung', 'Meeyoung Cha', 'Sungwon Park', 'Sihyun Kim', 'Sundong Kim', 'Eunji Lee']	2021-11-02	null	null	null	null	['code-classification']	['computer-code']	[-4.56723660e-01 -2.21290246e-01 -5.06721139e-01 -6.56495154e-01 -7.00094223e-01 -1.22227335e+00 2.29111925e-01 3.00272971e-01 -3.25175494e-01 3.68339717e-01 1.35621905e-01 -1.22112930e+00 -1.78289145e-01 -6.66636407e-01 -5.61939657e-01 -3.65740776e-01 -3.30508575e-02 5.50133884e-01 -4.15644765e-01 -2.15720013e-01 5.58204830e-01 5.63988864e-01 -8.04771125e-01 2.35665724e-01 1.02620125e+00 1.24302888e+00 1.59415767e-01 2.25852773e-01 -1.54134676e-01 6.84680402e-01 -4.03005153e-01 -1.03991973e+00 8.51087987e-01 -3.31685394e-02 -7.44307458e-01 -1.45386979e-01 1.32598355e-01 -7.96878397e-01 -1.52377799e-01 1.35642982e+00 -1.59213543e-01 -1.07370712e-01 8.96014392e-01 -1.05655217e+00 -1.27190125e+00 9.22310233e-01 -6.47823215e-01 1.16664171e-01 1.85054958e-01 -9.70510617e-02 1.57322240e+00 -8.58447254e-01 4.42912102e-01 7.01457918e-01 8.28431904e-01 -4.03557420e-02 -9.58513916e-01 -1.08851838e+00 1.43332392e-01 1.53049618e-01 -1.53100526e+00 -2.17050239e-01 4.33764309e-01 -7.73274660e-01 7.19350457e-01 2.62344271e-01 4.78147715e-01 4.34717327e-01 5.66119194e-01 4.85640079e-01 9.33793128e-01 -1.94796026e-01 2.48022035e-01 3.37425083e-01 4.64111380e-02 4.71403748e-01 6.13308430e-01 -2.31976256e-01 3.55456509e-02 -2.14439437e-01 4.40143168e-01 4.06590030e-02 2.86766022e-01 -6.17235824e-02 -7.29211628e-01 1.21379459e+00 5.24160147e-01 2.63630092e-01 -4.38336730e-01 -1.96862355e-01 4.25569713e-01 4.44652110e-01 3.32929969e-01 7.27371037e-01 -9.90885615e-01 -1.30973607e-01 -1.00483644e+00 1.60087585e-01 9.30923939e-01 9.54182744e-01 5.41973293e-01 -8.80169943e-02 3.82680714e-01 7.80896962e-01 4.32479709e-01 3.81537825e-01 -1.66429963e-03 -7.85720885e-01 6.12041175e-01 5.18856049e-01 3.61365736e-01 -1.26975536e+00 -3.72276038e-01 -7.77291417e-01 -7.67018735e-01 1.39291003e-01 5.58275998e-01 -2.73530602e-01 -8.74273062e-01 1.01397383e+00 -2.58501768e-01 -6.89835310e-01 -2.23929703e-01 1.01141918e+00 3.69029969e-01 7.30885208e-01 2.09642842e-01 2.69714236e-01 1.35043168e+00 -6.35875046e-01 -4.04223979e-01 -2.17034951e-01 6.33570492e-01 -1.07743621e+00 4.75640476e-01 5.26107728e-01 -5.83889425e-01 -5.27831793e-01 -8.27877164e-01 6.21438771e-02 -5.06771803e-01 6.29066110e-01 1.31871116e+00 5.38999677e-01 -5.69753408e-01 6.44904971e-01 -3.45175356e-01 -1.24045998e-01 3.52497727e-01 4.30806816e-01 -4.19626445e-01 -1.85510535e-02 -1.15583897e+00 1.09562528e+00 1.80356160e-01 4.38238919e-01 -5.15571952e-01 -4.41170782e-01 -9.14448917e-01 2.55261153e-01 3.18782255e-02 5.12451716e-02 1.19882035e+00 -1.12204933e+00 -8.05033743e-01 9.48192298e-01 2.28455052e-01 -5.48148870e-01 2.92869717e-01 -2.22135670e-02 -8.89374256e-01 -1.43601045e-01 7.39802718e-01 4.45803136e-01 3.59791428e-01 -1.09832752e+00 -1.15777206e+00 -2.70635933e-02 2.68200070e-01 -2.55985737e-01 1.18508907e-02 2.86515862e-01 -1.35225013e-01 -8.83582711e-01 4.11735654e-01 -1.13843954e+00 -1.01799712e-01 -2.78594699e-02 -4.70535755e-01 -4.50785249e-01 1.02795176e-01 -1.40517926e+00 1.35810781e+00 -2.30330038e+00 -4.82401550e-01 6.72829926e-01 -1.61512226e-01 2.34199598e-01 2.01924339e-01 3.61976385e-01 -1.99099407e-01 2.35971868e-01 -1.71784475e-01 3.20112616e-01 3.72384638e-01 5.04699051e-02 -7.40095019e-01 5.82605541e-01 2.93967336e-01 5.08563638e-01 -8.81341934e-01 2.17520725e-02 -7.86834732e-02 -8.63762796e-02 -4.71754730e-01 -2.05243468e-01 3.24270815e-01 2.20559448e-01 -3.53742450e-01 1.23852801e+00 9.25545931e-01 -1.82344228e-01 4.72623140e-01 -1.08753905e-01 -2.85851389e-01 5.90741634e-01 -1.08356810e+00 1.41428709e+00 -3.05677891e-01 8.16346467e-01 1.59621596e-01 -1.00781941e+00 1.18932426e+00 -1.31668031e-01 3.28993827e-01 -7.01986969e-01 1.15702286e-01 6.22823417e-01 4.46183145e-01 -3.77038687e-01 8.64313245e-01 -1.79313198e-01 -6.95190072e-01 3.11947733e-01 -2.50489891e-01 6.90619051e-02 2.80358493e-01 1.16730519e-01 6.70994818e-01 1.23291284e-01 4.95538712e-01 -4.23605204e-01 1.99770004e-01 4.71203029e-01 9.64410305e-01 6.65286005e-01 -2.95426965e-01 1.97579324e-01 6.72054529e-01 -9.40945983e-01 -1.15264547e+00 -7.34408081e-01 -2.89911419e-01 1.20567238e+00 2.36835945e-02 -1.75740406e-01 -2.27876633e-01 -8.03411543e-01 5.26676595e-01 8.25224280e-01 -5.94653845e-01 7.50042051e-02 -2.14870661e-01 -4.31352496e-01 5.72409689e-01 5.81036031e-01 3.76686364e-01 -3.21416914e-01 -2.24336669e-01 4.13413823e-01 6.60003349e-02 -9.23548043e-01 -5.21980703e-01 4.63428557e-01 -1.21464223e-01 -1.04585481e+00 -7.02557862e-01 -9.14539397e-01 7.45114088e-01 1.56219661e-01 8.98867071e-01 2.49768525e-01 9.11914706e-02 -4.72553045e-01 -5.67794979e-01 -3.96076113e-01 -5.49444973e-01 3.67900938e-01 1.55150950e-01 -9.45036300e-03 6.86917543e-01 -6.82433695e-02 -3.47239345e-01 2.60688126e-01 -3.94627631e-01 -9.60155055e-02 7.78481901e-01 4.95979816e-01 8.60538855e-02 3.65681231e-01 4.41877633e-01 -7.12668657e-01 4.27690178e-01 -6.74812138e-01 -9.20289814e-01 1.82161748e-01 -7.34483063e-01 -2.63498873e-01 7.35406220e-01 1.42581714e-02 -7.01812983e-01 2.62050718e-01 -3.49172026e-01 1.81967676e-01 -1.68650839e-02 6.84522033e-01 2.47697517e-01 -1.04301713e-01 1.38519570e-01 -8.11171383e-02 -3.16509366e-01 -7.81245947e-01 -3.09246220e-02 1.05128860e+00 3.93517107e-01 -3.42056155e-01 1.10957432e+00 6.40211850e-02 -2.53840059e-01 -1.17980115e-01 -7.82085001e-01 -4.31393474e-01 -7.96777248e-01 4.01860550e-02 7.15083063e-01 -1.07112205e+00 -8.89181972e-01 7.73154497e-02 -1.21956229e+00 1.78440325e-02 3.62632513e-01 6.26661897e-01 1.28221989e-01 9.53575298e-02 -7.65495121e-01 -5.98074913e-01 -1.07751943e-01 -1.21939313e+00 7.49335408e-01 4.68944153e-03 -6.21748805e-01 -6.65559947e-01 -5.47403097e-01 3.92276198e-01 4.35222685e-01 2.24233732e-01 1.22350156e+00 -9.46994901e-01 -2.68012375e-01 -4.80754405e-01 -4.56070185e-01 4.45521474e-01 2.66885608e-01 2.14178368e-01 -5.47508538e-01 3.89368087e-03 -5.00003994e-01 4.56048064e-02 7.36188114e-01 3.30006629e-01 9.44609106e-01 -2.52041757e-01 -2.23050892e-01 5.38219810e-01 1.46013927e+00 8.56104374e-01 1.47884712e-01 5.23060858e-01 4.55506921e-01 5.94697952e-01 9.36930478e-01 5.26103079e-01 6.37386501e-01 6.07855380e-01 3.01436722e-01 -1.20958284e-01 3.74936163e-01 -2.01850161e-01 2.30629817e-01 6.61755264e-01 2.60305833e-02 2.69398279e-02 -1.18574464e+00 5.75267136e-01 -1.45353663e+00 -7.59367168e-01 -2.38728479e-01 1.62606037e+00 5.86947978e-01 2.75828242e-01 -1.09903440e-01 2.05458567e-01 7.70148396e-01 -1.74625382e-01 -6.74012005e-01 -9.32535231e-01 2.01082632e-01 -2.08191350e-01 1.12684250e+00 3.50241661e-01 -1.32886887e+00 7.98201799e-01 6.64778996e+00 7.07145989e-01 -8.99876416e-01 -1.82753816e-01 8.03065956e-01 3.02405357e-01 -1.75849020e-01 1.58261403e-01 -8.94825399e-01 6.82952166e-01 5.63134789e-01 -7.49465153e-02 6.55942678e-01 1.12245762e+00 -5.58314286e-02 -1.38225570e-01 -9.99767780e-01 7.86303520e-01 -1.06716037e-01 -1.45965540e+00 8.75772610e-02 3.22362185e-01 7.41567552e-01 -6.46681115e-02 1.90754309e-01 4.73693162e-01 8.30819130e-01 -7.32647955e-01 1.34702265e+00 1.02406606e-01 8.48990500e-01 -8.75565827e-01 1.12887347e+00 4.82025109e-02 -1.09318054e+00 -7.65878141e-01 -5.66330671e-01 -3.19673121e-01 1.88556999e-01 6.51807070e-01 -7.76071906e-01 4.92710710e-01 6.17185771e-01 6.31550908e-01 -4.56704229e-01 8.27215195e-01 -2.77137607e-01 5.67159295e-01 3.24437357e-02 1.54007539e-01 8.27959299e-01 -2.78300017e-01 -8.40767398e-02 1.23990965e+00 5.71193755e-01 2.34627813e-01 7.57742375e-02 6.97362125e-01 -2.24460483e-01 -8.55561495e-02 -4.54438597e-01 -2.58102685e-01 5.93130410e-01 1.04078960e+00 -9.97289956e-01 -1.94889665e-01 -5.24504602e-01 7.99010098e-01 -3.96218337e-02 2.21574664e-01 -9.07158732e-01 -8.53825450e-01 1.00337243e+00 -1.55517697e-01 5.06348431e-01 -4.35942829e-01 -4.44297880e-01 -8.52935851e-01 -1.19050695e-02 -8.63091171e-01 4.42652285e-01 -6.31366968e-01 -1.16386783e+00 2.79526025e-01 -4.12148803e-01 -1.52040648e+00 -3.42457891e-02 -1.01514232e+00 -1.79488793e-01 9.11332011e-01 -1.58418298e+00 -1.07105637e+00 3.57439727e-01 -2.58087404e-02 4.08382177e-01 -4.28267628e-01 6.54961228e-01 5.12268901e-01 -4.05513823e-01 5.44639409e-01 4.86137629e-01 9.68967795e-01 5.11693358e-01 -1.03961921e+00 6.43623710e-01 8.29581559e-01 1.15491850e-02 1.02519512e+00 7.37247884e-01 -7.58181870e-01 -1.45867360e+00 -8.90567839e-01 1.30967271e+00 -3.62443745e-01 9.48007107e-01 -2.91467428e-01 -4.39969838e-01 6.94101870e-01 -7.61500150e-02 -4.73027349e-01 1.04290581e+00 1.41370267e-01 -4.04484868e-01 -3.50070089e-01 -1.24548352e+00 8.25117975e-02 6.47352815e-01 -6.03158116e-01 -7.03145683e-01 4.38976139e-01 4.46677804e-01 -3.71365815e-01 -1.01931810e+00 -9.83131900e-02 9.29188073e-01 -3.99325073e-01 5.81512749e-01 -6.08066022e-01 5.82371652e-01 -4.33083475e-01 -3.61469507e-01 -1.26240373e+00 -8.20371985e-01 -2.56363302e-01 6.33443534e-01 1.44188058e+00 7.77396560e-01 -4.65574920e-01 3.72287571e-01 1.02234268e+00 -3.00998777e-01 -4.39379245e-01 -8.98433387e-01 -7.45065272e-01 5.26932515e-02 -4.52156574e-01 8.95613492e-01 1.33422887e+00 -1.57969326e-01 -5.40555567e-02 -3.97277296e-01 4.29606706e-01 2.01808304e-01 4.86710221e-01 3.44943315e-01 -1.29051864e+00 -4.28158529e-02 -4.60902035e-01 -2.29155719e-01 -9.07945633e-01 3.27229470e-01 -1.23324549e+00 -5.05376607e-02 -1.69885683e+00 -1.63622111e-01 -9.11586761e-01 -3.45959932e-01 7.69217074e-01 4.59152311e-01 1.79734394e-01 2.68906653e-01 1.81767672e-01 -2.44045049e-01 -3.10997695e-01 7.46812046e-01 -4.83399063e-01 1.34647608e-01 1.06108072e-03 -1.35343480e+00 8.16396117e-01 7.85137177e-01 -5.98239541e-01 2.08926782e-01 -9.80620861e-01 8.02821457e-01 -2.36287445e-01 4.77178507e-02 -5.01145601e-01 3.39919955e-01 -4.09107625e-01 6.92997813e-01 -6.73424900e-01 -1.89777747e-01 -1.15302169e+00 3.15714061e-01 7.32944548e-01 -4.34868097e-01 4.57363158e-01 1.21941775e-01 2.43725538e-01 -1.34560108e-01 -4.69967753e-01 4.21588421e-01 -2.76040316e-01 -9.29487348e-01 8.21679235e-02 -7.29576051e-01 -5.80248535e-01 1.01661563e+00 -1.05191126e-01 2.49321107e-02 -2.46582747e-01 -6.25536203e-01 2.68053740e-01 3.94386023e-01 6.44880712e-01 2.79744416e-01 -1.33745801e+00 -6.65318966e-01 3.28951389e-01 7.54025206e-02 -6.40165865e-01 -2.28799030e-01 4.25711006e-01 -8.82113278e-01 7.71772385e-01 -2.41337016e-01 1.80582236e-02 -7.13406920e-01 3.57356101e-01 -3.73931107e-04 -6.42161220e-02 -3.21548164e-01 8.21850777e-01 -3.16984743e-01 -3.73484641e-01 1.10185958e-01 -5.91652513e-01 -3.39998364e-01 5.06165385e-01 5.25096536e-01 2.68970370e-01 2.68329054e-01 -9.19386685e-01 -6.82175040e-01 6.63375854e-01 -3.30728203e-01 2.56967425e-01 1.56816351e+00 3.98912416e-05 -2.07723275e-01 -6.67887554e-03 1.00672150e+00 2.93908119e-02 -9.38378572e-01 6.04706928e-02 3.21932733e-01 -7.38682866e-01 -6.87379017e-02 -1.03815305e+00 -1.13990664e+00 6.86580598e-01 3.07597011e-01 3.67754102e-01 7.18278587e-01 -1.53835863e-01 9.81745064e-01 4.42199439e-01 7.35941470e-01 -1.35745823e+00 -7.93210208e-01 5.21940112e-01 6.38697386e-01 -1.24286866e+00 1.86330099e-02 -3.20592597e-02 -7.54800081e-01 1.19511044e+00 1.55538768e-01 -7.62670040e-02 5.18142819e-01 8.91079288e-03 1.19253188e-01 -2.54006428e-03 -3.61430377e-01 6.34682104e-02 1.56670451e-01 4.04212326e-01 5.17576635e-01 5.05365849e-01 -5.23937166e-01 8.15372109e-01 -2.77865738e-01 -2.44116902e-01 5.40274024e-01 6.80696487e-01 -3.38118494e-01 -8.25554848e-01 -3.33689809e-01 9.66776431e-01 -7.67862618e-01 -3.39110851e-01 -2.60960162e-01 6.63736165e-01 6.06046796e-01 1.13001800e+00 2.89771885e-01 -6.55471563e-01 2.65533954e-01 -1.13970764e-01 -1.85381144e-01 -6.25319719e-01 -8.69937181e-01 -3.88875037e-01 3.72494549e-01 -2.45160878e-01 -1.10873329e-02 -7.60335386e-01 -1.18208098e+00 -7.94634461e-01 -4.11196798e-01 4.02269274e-01 7.50554979e-01 9.26425993e-01 1.39312357e-01 -7.81171918e-02 8.31621170e-01 -3.97694111e-01 -6.61162853e-01 -7.21238971e-01 -1.05139577e+00 3.78056765e-01 4.19239789e-01 -5.66322148e-01 -3.02887738e-01 2.66209003e-02]	[9.647439002990723, 6.143615245819092]
a4deac45-5d35-4632-ad24-96468087ab28	fully-autonomous-programming-with-large	2304.10423	null	https://arxiv.org/abs/2304.10423v1	https://arxiv.org/pdf/2304.10423v1.pdf	Fully Autonomous Programming with Large Language Models	Current approaches to program synthesis with Large Language Models (LLMs) exhibit a "near miss syndrome": they tend to generate programs that semantically resemble the correct answer (as measured by text similarity metrics or human evaluation), but achieve a low or even zero accuracy as measured by unit tests due to small imperfections, such as the wrong input or output format. This calls for an approach known as Synthesize, Execute, Debug (SED), whereby a draft of the solution is generated first, followed by a program repair phase addressing the failed tests. To effectively apply this approach to instruction-driven LLMs, one needs to determine which prompts perform best as instructions for LLMs, as well as strike a balance between repairing unsuccessful programs and replacing them with newly generated ones. We explore these trade-offs empirically, comparing replace-focused, repair-focused, and hybrid debug strategies, as well as different template-based and model-based prompt-generation techniques. We use OpenAI Codex as the LLM and Program Synthesis Benchmark 2 as a database of problem descriptions and tests for evaluation. The resulting framework outperforms both conventional usage of Codex without the repair phase and traditional genetic programming approaches.	['Leon Moonen', 'Aki Härmä', 'Anastasiia Grishina', 'Vadim Liventsev']	2023-04-20	null	null	null	null	['program-repair', 'program-synthesis', 'program-repair']	['computer-code', 'computer-code', 'reasoning']	[ 1.80865616e-01 5.51621206e-02 -1.77276522e-01 -3.39247197e-01 -1.08969128e+00 -7.52846718e-01 4.88834500e-01 4.75984871e-01 -4.86961491e-02 6.07680082e-01 3.04525848e-02 -8.71869385e-01 9.63801518e-02 -8.44498158e-01 -7.09639251e-01 -1.61531687e-01 1.43289611e-01 5.32683611e-01 4.38830078e-01 -2.60299712e-01 8.41847599e-01 -5.05352654e-02 -1.85236633e+00 3.59444618e-01 1.28150868e+00 2.75565863e-01 2.07613438e-01 1.08103013e+00 -5.73454618e-01 8.48557770e-01 -9.75543380e-01 -4.09669667e-01 -6.98849037e-02 -8.09803724e-01 -8.82829249e-01 -1.24778368e-01 9.19417217e-02 -1.13219805e-01 4.57080662e-01 1.10644162e+00 3.40275675e-01 -1.00481585e-01 3.06101710e-01 -1.34294188e+00 -4.06214744e-01 8.99929762e-01 -5.19563913e-01 -1.78402647e-01 9.28148031e-01 4.30038989e-01 8.11670244e-01 -6.61814630e-01 4.42387879e-01 1.34516752e+00 5.29928148e-01 6.43050253e-01 -1.58272374e+00 -2.31141806e-01 -3.84597272e-01 -1.36940613e-01 -1.30454576e+00 -5.57930112e-01 1.89506978e-01 -5.49773514e-01 1.41671026e+00 4.50159013e-01 1.41653329e-01 7.79427767e-01 5.12358069e-01 3.95985484e-01 1.09102404e+00 -9.00765538e-01 4.73072201e-01 3.30742896e-01 3.32947910e-01 8.74110043e-01 4.12786245e-01 2.58350164e-01 -1.18569978e-01 -6.58601820e-01 9.36229080e-02 -5.20322263e-01 -3.06790173e-01 -1.44457102e-01 -1.06622362e+00 7.96611965e-01 -1.22882947e-01 3.05051059e-01 -1.44762799e-01 2.52020121e-01 4.69857424e-01 7.50842392e-01 -8.20389092e-02 9.82774973e-01 -4.47472364e-01 -4.95298296e-01 -1.11565483e+00 6.46395326e-01 1.20557249e+00 9.09529805e-01 9.24995124e-01 2.30432883e-01 -3.16035867e-01 5.63244045e-01 2.25439817e-01 7.48514524e-03 9.62916017e-01 -6.66076005e-01 5.16121387e-01 9.63749945e-01 -6.78106770e-02 -6.13694906e-01 -1.37287676e-01 -1.40992463e-01 -4.17393930e-02 6.39850736e-01 2.14102581e-01 -2.56044507e-01 -6.67335510e-01 1.73446310e+00 4.88551222e-02 -9.96012092e-02 1.32874057e-01 4.89063174e-01 3.30655634e-01 6.27927780e-01 -1.56645402e-01 -1.15226507e-01 9.40079987e-01 -1.03798163e+00 -3.11049163e-01 -4.70662951e-01 1.48371375e+00 -8.81263793e-01 1.33515930e+00 5.28815091e-01 -1.28924847e+00 -5.48257649e-01 -1.37080622e+00 4.26573664e-01 -2.31432050e-01 -8.94419774e-02 2.70815402e-01 9.64480817e-01 -1.35505688e+00 8.95122468e-01 -6.09672964e-01 -2.17251644e-01 -2.53407031e-01 2.52327114e-01 -5.92101440e-02 -2.03665361e-01 -4.07860845e-01 9.03436303e-01 5.72472692e-01 -3.30966324e-01 -7.36361027e-01 -7.63019502e-01 -9.05531049e-01 4.93245497e-02 3.85191858e-01 -4.82264310e-01 1.57637489e+00 -1.09772897e+00 -1.22839308e+00 6.20150149e-01 -2.02726737e-01 -3.37506354e-01 2.03306004e-01 -2.23845616e-02 -2.57264376e-01 -5.70594907e-01 2.92147100e-02 3.90493989e-01 5.46916604e-01 -1.32542706e+00 -5.85221112e-01 3.44795287e-02 9.06409994e-02 -3.32520217e-01 1.24301612e-01 3.70897532e-01 -5.25579937e-02 -3.45310003e-01 -2.26369366e-01 -1.03922987e+00 -2.95400143e-01 -6.21775806e-01 -4.79475915e-01 8.33284333e-02 5.60697079e-01 -6.60912454e-01 1.74298823e+00 -2.04182148e+00 3.44335318e-01 5.30882299e-01 2.34841425e-02 2.95876771e-01 -6.19211674e-01 5.76426864e-01 -3.92522961e-01 5.00612855e-01 -3.22023124e-01 -3.68324891e-02 2.72265762e-01 2.01249778e-01 -2.13272467e-01 2.18422711e-02 3.88150543e-01 8.79611552e-01 -9.30497050e-01 -4.02877927e-01 -2.77514815e-01 -2.62254864e-01 -9.70706224e-01 3.65351647e-01 -7.40458727e-01 -7.29207993e-02 -2.22867072e-01 7.62347519e-01 3.44139487e-01 -1.39855862e-01 2.57784814e-01 2.88795799e-01 -2.53670335e-01 3.71552378e-01 -1.24403083e+00 1.42859423e+00 -5.70019364e-01 5.50590336e-01 -2.12479740e-01 -6.86890900e-01 1.14464033e+00 2.55261749e-01 -2.52227366e-01 -7.77524233e-01 -1.17256477e-01 7.23205984e-01 4.16167647e-01 -5.66612482e-01 4.94680315e-01 2.49187931e-01 -2.77289778e-01 7.93644667e-01 -1.87472001e-01 -3.74444932e-01 4.62159604e-01 8.09810087e-02 1.71230745e+00 3.98008317e-01 4.61086333e-01 -2.47189417e-01 6.56365037e-01 3.19237053e-01 4.34269309e-01 8.36093068e-01 2.75160849e-01 6.37726843e-01 1.09201360e+00 -2.30865404e-02 -1.04255927e+00 -6.00749075e-01 3.41912061e-01 9.71481800e-01 -3.54807585e-01 -9.95726824e-01 -1.14263213e+00 -7.41587102e-01 -1.47921965e-01 1.34305823e+00 -4.35604841e-01 -5.37008226e-01 -7.66672552e-01 -5.79779148e-01 7.06272721e-01 3.23956966e-01 -5.43589629e-02 -1.17043805e+00 -1.03656471e+00 4.67645824e-01 1.45000651e-01 -2.77494550e-01 -3.43336582e-01 5.39806485e-01 -6.61048770e-01 -1.11352766e+00 -1.51446328e-01 -5.86309612e-01 6.84288263e-01 -1.15805604e-01 1.60591173e+00 8.53491724e-01 -2.50032842e-01 1.55055255e-01 -6.16852939e-01 1.10464096e-01 -1.33659136e+00 -9.32161808e-02 -3.75301719e-01 -4.72296357e-01 1.35081336e-01 -3.20206314e-01 3.55072096e-02 2.29448214e-01 -1.08990180e+00 -1.85729325e-01 7.32761025e-01 1.10214806e+00 2.41497517e-01 -1.85508057e-02 4.80814278e-01 -9.28910673e-01 1.19000435e+00 -5.72894990e-01 -9.74438071e-01 7.00786471e-01 -8.71010780e-01 5.81615508e-01 7.95731664e-01 -4.69545275e-01 -7.16036499e-01 -1.45057768e-01 -3.60840291e-01 -1.20209873e-01 -1.22271724e-01 8.40030491e-01 -1.42281443e-01 -3.27083260e-01 1.01490164e+00 2.00981647e-01 -6.63587227e-02 -3.41430664e-01 2.75882799e-02 3.99907023e-01 4.75302100e-01 -1.05178773e+00 7.07692444e-01 -7.22977281e-01 -3.79736960e-01 -3.49625319e-01 -7.99136385e-02 8.45566839e-02 -2.33609915e-01 5.35266995e-02 4.58104372e-01 -2.67585307e-01 -3.41275543e-01 1.11836739e-01 -1.31018138e+00 -6.42385662e-01 -2.16923669e-01 -5.87248877e-02 -4.73682255e-01 2.97829062e-01 -3.53815198e-01 -5.67368150e-01 -1.97620183e-01 -1.74989128e+00 1.05985498e+00 1.34222955e-01 -8.95630360e-01 -6.87968493e-01 3.71662170e-01 -7.81324729e-02 6.29810691e-01 2.77658552e-01 1.84502673e+00 -8.57232571e-01 -4.70022917e-01 -3.20334077e-01 8.57603997e-02 2.74197400e-01 -1.55187130e-01 3.57156396e-01 -4.92149740e-01 -2.92218328e-01 -1.17103748e-01 -3.61133426e-01 1.86908141e-01 -4.98323329e-02 7.41886377e-01 -3.32782596e-01 -4.23094839e-01 4.26126838e-01 1.60479486e+00 4.01089996e-01 8.00752759e-01 4.15459692e-01 3.54496777e-01 5.22417068e-01 5.71920335e-01 4.18391466e-01 8.59502554e-02 6.90809667e-01 2.76004821e-01 4.80940253e-01 7.38872401e-03 -2.53824800e-01 6.43451035e-01 7.72941589e-01 6.69593215e-01 -3.32632661e-01 -1.28071284e+00 4.39293832e-01 -1.74239755e+00 -5.67220271e-01 -3.26651424e-01 2.61577320e+00 9.42596853e-01 4.28986341e-01 3.35995224e-03 4.51643646e-01 2.92994857e-01 -3.06007266e-01 -1.19885087e-01 -1.04259157e+00 2.59325653e-01 4.70881194e-01 1.95636109e-01 6.53824031e-01 -3.74036223e-01 6.91257715e-01 6.50023699e+00 8.14624608e-01 -1.13657498e+00 -6.70328066e-02 4.89125937e-01 2.44908422e-01 -7.28235066e-01 5.73423862e-01 -7.18827426e-01 4.90117908e-01 1.57067716e+00 -5.31314790e-01 5.81592619e-01 8.65228534e-01 -6.52775285e-04 -4.84368354e-01 -1.42898548e+00 4.76662636e-01 6.04335256e-02 -1.29833484e+00 -2.22669274e-01 -1.76678255e-01 6.16074622e-01 -2.80512482e-01 -4.22799021e-01 6.06350780e-01 3.86281222e-01 -1.11982071e+00 9.35514867e-01 4.38061446e-01 4.93723303e-01 -7.44199574e-01 7.04085171e-01 5.84684610e-01 -1.11358821e+00 -1.77413091e-01 2.42406428e-02 1.95586476e-02 -9.24873054e-02 2.06093058e-01 -9.98852372e-01 3.72155875e-01 3.28251928e-01 -1.06907144e-01 -1.09584963e+00 1.27731049e+00 -6.09528832e-02 5.64243853e-01 1.71250347e-02 -3.08992743e-01 5.73449768e-02 -1.88393481e-02 4.17556942e-01 1.26222110e+00 6.16807401e-01 -2.73406088e-01 2.79400915e-01 1.15639198e+00 4.11977649e-01 1.18745109e-02 -5.79646349e-01 -2.37436190e-01 5.69586992e-01 9.63957548e-01 -8.24541569e-01 -4.36091661e-01 -4.42515165e-01 7.53191113e-01 2.71796346e-01 2.04322025e-01 -7.72118986e-01 -6.10591710e-01 2.82751203e-01 2.07070589e-01 3.86819169e-02 -2.02256128e-01 -3.78570884e-01 -8.29467356e-01 1.35050863e-01 -1.74130845e+00 -6.73225382e-03 -7.58636951e-01 -5.81335783e-01 9.01468933e-01 1.61174044e-01 -9.37043011e-01 -7.97378778e-01 -2.35150427e-01 -1.07043982e+00 1.06346226e+00 -1.02243960e+00 -4.51754659e-01 -1.47364095e-01 8.32675491e-03 5.36868632e-01 -9.22454596e-02 8.57826412e-01 3.47907484e-01 -4.69404936e-01 7.66299605e-01 -2.29054704e-01 -5.29776275e-01 5.17566979e-01 -1.38864946e+00 6.99950397e-01 1.03821528e+00 -2.68203229e-01 8.02801073e-01 9.74378109e-01 -8.52250040e-01 -1.79573095e+00 -9.72346842e-01 1.12301636e+00 -3.90696645e-01 5.80853164e-01 -2.31415346e-01 -1.31718087e+00 3.60836804e-01 1.49207532e-01 -4.73344892e-01 3.52345228e-01 -1.78991079e-01 -2.21782357e-01 2.24884853e-01 -1.09469128e+00 7.68364787e-01 5.28032839e-01 -3.18458229e-01 -5.10407627e-01 2.51878351e-01 8.25289130e-01 -4.55302089e-01 -6.66045249e-01 1.22589886e-01 2.73502320e-01 -1.18847096e+00 5.02624154e-01 -3.40562582e-01 6.05993450e-01 -3.97487432e-01 -3.01003814e-01 -1.41417551e+00 -1.81384888e-02 -7.73456156e-01 1.28223807e-01 1.49646676e+00 6.71027839e-01 -6.45291984e-01 4.66442287e-01 6.54218912e-01 -3.66250187e-01 -7.76554823e-01 -4.20234114e-01 -7.56229222e-01 -6.01111501e-02 -3.70397270e-01 8.92111659e-01 6.07070088e-01 1.65677950e-01 9.24534425e-02 -3.93259674e-02 -1.39966711e-01 2.25318149e-01 1.59275085e-01 9.96390104e-01 -8.57054532e-01 -8.72749150e-01 -8.18681717e-01 -9.99871194e-02 -5.18355548e-01 2.61494368e-01 -7.77802587e-01 3.46445143e-01 -9.84781682e-01 -1.59434024e-02 -4.23600644e-01 3.13110113e-01 5.81784666e-01 -1.67439044e-01 -2.27146983e-01 1.38504822e-02 -1.18267559e-01 -3.24232787e-01 4.95294482e-02 5.89012504e-01 9.60232597e-03 -3.84996384e-01 -4.39062789e-02 -7.62174845e-01 4.35671359e-01 5.24433076e-01 -6.79451168e-01 -3.96646231e-01 -3.60591024e-01 7.00394332e-01 8.28489244e-01 2.64742196e-01 -1.16382313e+00 2.76086599e-01 -2.20805854e-01 -2.26627529e-01 -2.91129053e-01 -3.87241960e-01 -4.20276284e-01 4.38351095e-01 8.25000107e-01 -5.61421275e-01 9.86146152e-01 3.65251303e-01 1.16546065e-01 -3.00854266e-01 -9.76554692e-01 5.49302399e-01 -2.08653420e-01 -6.79062784e-01 -3.60490203e-01 -4.84621465e-01 4.48651947e-02 1.02273810e+00 -3.36171925e-01 -4.86303091e-01 1.50208920e-02 -1.48325294e-01 2.49647915e-01 8.22315335e-01 5.21511555e-01 4.96678799e-01 -1.03859937e+00 -7.00914741e-01 5.38530350e-01 2.30044305e-01 -4.63142514e-01 -2.28371203e-01 6.49649501e-01 -7.71389663e-01 3.11486512e-01 -5.48746772e-02 -3.96179825e-01 -1.26128328e+00 5.85147142e-01 2.93391526e-01 -3.50701958e-01 -2.29461953e-01 5.33751190e-01 -1.18728504e-01 -5.90970874e-01 -6.01741625e-03 -4.57456857e-01 1.28806710e-01 -3.18310440e-01 4.15307939e-01 1.92285851e-01 5.01461923e-01 -1.17040224e-01 -4.03634071e-01 2.30550289e-01 1.45716891e-01 -9.24125388e-02 9.02504921e-01 3.37001383e-01 -3.65494400e-01 2.72487611e-01 1.06527340e+00 6.94663227e-02 -5.62383831e-01 6.35547657e-03 6.23413503e-01 -5.23886383e-01 -2.68050700e-01 -8.69388282e-01 -6.13542438e-01 6.08005702e-01 3.80570203e-01 4.06327963e-01 1.03276408e+00 -2.81758845e-01 4.88977224e-01 3.95540148e-01 4.64892238e-01 -7.33127773e-01 1.02121904e-01 5.48899472e-01 7.69245684e-01 -9.41939652e-01 -2.63506651e-01 2.76970547e-02 -3.25259089e-01 1.15088391e+00 9.54876959e-01 -1.69743389e-01 3.93906087e-02 7.80413806e-01 -2.49761462e-01 1.20950244e-01 -1.29509985e+00 1.98035732e-01 9.98758972e-02 4.35764462e-01 7.37461269e-01 -1.28116101e-01 -5.54123402e-01 4.92172927e-01 -2.90557474e-01 4.99116369e-02 7.54380763e-01 1.50885069e+00 -4.51216131e-01 -1.64652300e+00 -7.58011222e-01 6.00627124e-01 -1.74827993e-01 -1.90083146e-01 -4.02242243e-01 8.76459658e-01 -6.06151558e-02 9.40338433e-01 -1.93566695e-01 -6.83420241e-01 3.78989428e-01 3.05309772e-01 4.53312039e-01 -8.92246664e-01 -1.10214984e+00 -9.27703604e-02 3.74872506e-01 -5.78373730e-01 4.43504006e-01 -4.78700429e-01 -1.19889557e+00 -4.60891813e-01 -4.00845557e-01 2.78246254e-01 6.44601762e-01 1.01554430e+00 3.17810684e-01 7.63169050e-01 3.81976873e-01 -7.17719793e-01 -9.79546607e-01 -4.96010512e-01 6.36706948e-02 5.34655228e-02 2.94851422e-01 -3.16251785e-01 -1.45686880e-01 -5.54875955e-02]	[7.861971378326416, 7.576777458190918]
0d16ccf9-269b-47b3-b888-2cd4048dcf59	enhancing-the-robustness-of-qmix-against	2307.00907	null	https://arxiv.org/abs/2307.00907v1	https://arxiv.org/pdf/2307.00907v1.pdf	Enhancing the Robustness of QMIX against State-adversarial Attacks	Deep reinforcement learning (DRL) performance is generally impacted by state-adversarial attacks, a perturbation applied to an agent's observation. Most recent research has concentrated on robust single-agent reinforcement learning (SARL) algorithms against state-adversarial attacks. Still, there has yet to be much work on robust multi-agent reinforcement learning. Using QMIX, one of the popular cooperative multi-agent reinforcement algorithms, as an example, we discuss four techniques to improve the robustness of SARL algorithms and extend them to multi-agent scenarios. To increase the robustness of multi-agent reinforcement learning (MARL) algorithms, we train models using a variety of attacks in this research. We then test the models taught using the other attacks by subjecting them to the corresponding attacks throughout the training phase. In this way, we organize and summarize techniques for enhancing robustness when used with MARL.	['Jiacun Wang', 'Ling Wang', 'Ziyuan Zhou', 'Guanjun Liu', 'Weiran Guo']	2023-07-03	null	null	null	null	['multi-agent-reinforcement-learning']	['methodology']	[-1.49856955e-01 9.26576704e-02 -1.58338830e-01 1.85789570e-01 -9.48436975e-01 -8.03587854e-01 8.65449011e-01 2.02077776e-01 -5.87694287e-01 1.06155491e+00 -1.13889799e-01 -2.81056434e-01 -5.10678068e-02 -9.74067748e-01 -8.52979779e-01 -9.86156225e-01 -6.93379402e-01 4.63769913e-01 5.96294820e-01 -6.72227144e-01 2.49941334e-01 6.91986680e-01 -8.55485857e-01 -9.89167988e-02 5.11186063e-01 2.42587358e-01 -5.94565272e-01 1.16662252e+00 6.28402352e-01 1.44073117e+00 -1.33543062e+00 -2.78935581e-01 4.39343929e-01 -5.04109263e-01 -8.48155439e-01 -2.08533198e-01 1.16114460e-01 -8.09339464e-01 -7.83901989e-01 1.12527251e+00 7.36357272e-01 1.76154271e-01 4.05625224e-01 -1.88506675e+00 -3.95431548e-01 9.47834074e-01 -6.06647909e-01 2.20616341e-01 3.72969151e-01 6.02402985e-01 5.72219789e-01 1.83218315e-01 3.35426241e-01 1.68612027e+00 4.24141854e-01 1.01125014e+00 -1.07635224e+00 -8.33703279e-01 1.98780552e-01 2.45744199e-01 -7.34148562e-01 -4.03206311e-02 8.48689437e-01 2.67707873e-02 7.97147691e-01 -8.28684717e-02 3.11078995e-01 1.49172366e+00 5.97491682e-01 8.65725517e-01 1.48051488e+00 -2.74022788e-01 6.22233748e-01 -1.99071094e-01 -2.18376756e-01 7.59622872e-01 2.46529892e-01 1.09291184e+00 -1.21794194e-01 -6.50813818e-01 6.36294544e-01 -2.97301829e-01 2.29769066e-01 -3.37400258e-01 -9.34821248e-01 1.28960991e+00 3.70547324e-01 -1.64498612e-02 -2.67529726e-01 7.01415598e-01 8.57408285e-01 9.95233297e-01 -1.17114998e-01 7.26428032e-01 -3.71293187e-01 4.54907566e-02 -3.41281295e-01 7.17423320e-01 7.21320927e-01 4.37726855e-01 6.49505675e-01 7.32811630e-01 9.33630392e-02 3.61550301e-01 3.90425205e-01 7.05311894e-01 5.56029916e-01 -1.10476351e+00 2.18615428e-01 4.47560735e-02 3.89975449e-03 -5.90676129e-01 -4.97708887e-01 -1.39598384e-01 -5.98676085e-01 1.23812449e+00 2.87765324e-01 -8.59331608e-01 -6.23566270e-01 1.92930496e+00 4.41405475e-01 4.69563603e-01 9.22862172e-01 4.54119027e-01 3.54529023e-01 5.72407424e-01 9.79039297e-02 -1.79891974e-01 8.13758433e-01 -9.87170160e-01 -4.60403711e-01 -7.28920326e-02 5.38453400e-01 -5.15385866e-01 4.36995953e-01 4.12976414e-01 -9.15354073e-01 -4.75394696e-01 -1.39394903e+00 8.35358679e-01 -5.25882483e-01 -6.91315770e-01 5.07088542e-01 9.34750259e-01 -7.68163562e-01 7.59354711e-01 -1.18495131e+00 1.18078310e-02 2.71363556e-01 6.24861062e-01 -2.33681917e-01 2.81642318e-01 -1.71727395e+00 1.25937033e+00 4.61202830e-01 -5.19938350e-01 -1.98267722e+00 -3.44644964e-01 -8.98039103e-01 -2.89566189e-01 5.63704193e-01 -2.40333647e-01 1.58192492e+00 -8.33069444e-01 -1.93544245e+00 1.59954339e-01 8.12693119e-01 -1.04090846e+00 5.86372912e-01 -1.60013333e-01 -6.49499655e-01 4.37247515e-01 -1.05744064e-01 3.88167053e-01 1.21229923e+00 -1.47255218e+00 -5.10790527e-01 -1.24070346e-01 7.63086855e-01 4.56102267e-02 3.80134396e-02 1.79715469e-01 6.31127238e-01 -6.76309943e-01 -8.89193654e-01 -1.17555642e+00 -6.04025722e-01 -5.07827520e-01 -2.18633786e-01 -2.92186797e-01 1.29396009e+00 -1.12246662e-01 7.16889620e-01 -2.12483525e+00 5.36774620e-02 2.93327659e-01 -5.09511009e-02 7.45199263e-01 -7.62700915e-01 1.12440395e+00 -2.58552842e-02 -6.10502511e-02 -2.06715390e-02 -6.17051721e-02 3.57130557e-01 5.82480907e-01 -7.47158408e-01 6.88228846e-01 1.49368674e-01 8.29069793e-01 -9.80353951e-01 -2.30712846e-01 2.31121197e-01 8.64518210e-02 -5.94012797e-01 5.72503984e-01 -2.62578785e-01 4.62857902e-01 -6.31493568e-01 4.69644666e-01 4.90101129e-01 4.51498061e-01 9.01540145e-02 3.79122317e-01 3.01488101e-01 -2.29729593e-01 -1.18679500e+00 1.08376503e+00 -2.05921412e-01 2.96004236e-01 -2.33262759e-02 -9.98324811e-01 7.91305125e-01 6.56643510e-01 5.82889616e-01 -5.71495533e-01 7.43655711e-02 9.95595604e-02 4.64775413e-01 -2.85819083e-01 2.50014901e-01 -3.00572246e-01 -4.40841705e-01 7.80735850e-01 2.32396543e-01 -4.03996885e-01 3.05428833e-01 3.47075224e-01 1.42693818e+00 -9.40428302e-02 4.73945022e-01 1.68610409e-01 5.66633284e-01 -1.38917282e-01 4.60062861e-01 1.41099107e+00 -6.91971064e-01 -2.74027735e-01 5.16829431e-01 -4.52498168e-01 -1.01465714e+00 -1.20047164e+00 4.04706329e-01 1.09051704e+00 -1.21747516e-02 -3.31079006e-01 -8.12427938e-01 -1.19496870e+00 3.40791732e-01 6.54153705e-01 -8.16577852e-01 -7.43932962e-01 -6.75007522e-01 -7.12237477e-01 1.35724914e+00 4.73332852e-01 7.10338831e-01 -1.58197021e+00 -8.49607408e-01 5.37250578e-01 5.37275493e-01 -7.48463631e-01 -1.49808586e-01 2.74512261e-01 -4.47764575e-01 -1.39141381e+00 -3.20887715e-01 -6.78401053e-01 2.62180001e-01 -1.48965880e-01 8.24960172e-01 2.82991111e-01 3.99489291e-02 6.89156353e-01 -5.61733007e-01 -5.95457435e-01 -1.37336600e+00 -7.34077990e-02 6.45583272e-01 -4.72786427e-01 -2.86293346e-02 -5.35342872e-01 -7.02146739e-02 2.78179944e-01 -1.29785478e+00 -8.12807322e-01 5.08331597e-01 9.96491015e-01 -1.65508091e-01 3.67815405e-01 1.09338832e+00 -7.18839407e-01 1.05895782e+00 -2.34770760e-01 -9.59849536e-01 1.19462170e-01 -3.34018230e-01 1.28029570e-01 1.18443263e+00 -8.97372663e-01 -5.17264068e-01 -2.52575219e-01 -4.43280190e-01 -4.24058437e-01 -4.02304620e-01 2.61180997e-01 2.01528594e-01 -5.90751410e-01 8.50752115e-01 2.09130436e-01 3.65578204e-01 1.15462221e-01 3.87662202e-01 3.18647176e-01 2.44462058e-01 -7.18089104e-01 1.29576206e+00 2.06403688e-01 1.73732847e-01 -5.75053811e-01 -5.32870173e-01 3.51829112e-01 -1.28768831e-01 -1.61345482e-01 4.56818134e-01 -6.60914421e-01 -1.36162555e+00 8.19254041e-01 -7.50108242e-01 -9.69798028e-01 -3.33491981e-01 4.83050704e-01 -1.08080029e+00 5.09862363e-01 -9.39365745e-01 -6.72916770e-01 -1.80385232e-01 -1.42879248e+00 4.25036281e-01 2.87453234e-01 1.70675740e-01 -1.12288344e+00 9.26802218e-01 -8.10484812e-02 4.63816226e-01 5.10878861e-01 7.30872273e-01 -1.12674129e+00 -2.60452926e-01 -1.36431247e-01 6.23946249e-01 4.51842368e-01 -4.04448528e-03 1.38200164e-01 -6.18494987e-01 -1.08967853e+00 2.34291721e-02 -1.07079363e+00 4.04295146e-01 1.23040400e-01 7.27184534e-01 -6.50439084e-01 2.83434130e-02 2.23035902e-01 1.46703911e+00 5.23713231e-01 5.44687808e-01 7.86454022e-01 2.51149654e-01 1.10608503e-01 6.72731459e-01 4.96898562e-01 8.47947523e-02 3.89025062e-01 1.06589055e+00 5.33591919e-02 4.21833724e-01 -2.46677533e-01 1.05985129e+00 2.42023066e-01 1.93466336e-01 -1.06141284e-01 -5.95621526e-01 4.49307002e-02 -1.87928498e+00 -1.40920019e+00 5.47773719e-01 1.64759934e+00 8.07093203e-01 3.50718975e-01 5.46239018e-01 1.69829130e-01 4.43881720e-01 4.42687303e-01 -7.82048881e-01 -7.99652636e-01 -1.07713416e-01 2.28505343e-01 5.73227525e-01 7.03982770e-01 -1.30981457e+00 1.22363663e+00 7.00840044e+00 7.52208471e-01 -1.06154084e+00 -1.19811162e-01 8.84255320e-02 2.08461866e-01 2.54274338e-01 -2.72654872e-02 -5.28301299e-01 1.87352434e-01 1.06283450e+00 -3.45641315e-01 7.93194234e-01 9.57686841e-01 9.14485008e-02 -6.88224360e-02 -9.31373060e-01 3.28792900e-01 -1.09561771e-01 -1.08436108e+00 1.19962217e-02 8.90445802e-03 8.29522848e-01 7.73731321e-02 1.87983304e-01 1.02020490e+00 1.37427127e+00 -1.04457843e+00 2.34461889e-01 -1.02509648e-01 9.04740021e-02 -1.36175108e+00 6.74581110e-01 4.07473415e-01 -7.53868103e-01 -4.12532777e-01 -3.78475457e-01 -4.12637219e-02 -1.57658309e-01 -3.45525444e-01 -7.18683481e-01 5.56179404e-01 3.54903996e-01 3.01096886e-01 -6.35559738e-01 7.76027739e-01 -5.46495378e-01 8.57717395e-01 -6.84190467e-02 -1.53898925e-01 7.56736577e-01 1.56280607e-01 6.95794761e-01 8.54778469e-01 -3.31274867e-01 -1.51420906e-01 7.06099570e-01 2.30100825e-01 1.88425928e-02 -3.57904643e-01 -9.22964454e-01 -1.93022825e-02 4.47706521e-01 1.04144204e+00 -2.71694064e-01 -4.16453540e-01 -4.01635915e-01 7.74153292e-01 4.74152118e-01 3.08857471e-01 -1.07332575e+00 -4.22773033e-01 8.46322954e-01 -5.96658170e-01 2.17871606e-01 -3.71788919e-01 5.18969238e-01 -8.19155216e-01 -6.18122101e-01 -1.84033597e+00 6.46575451e-01 -3.99672091e-01 -1.51431680e+00 5.95587611e-01 -1.20687209e-01 -1.16788733e+00 -7.51573682e-01 -3.96842837e-01 -8.84265721e-01 2.18143627e-01 -1.51712310e+00 -1.05341768e+00 4.09585625e-01 1.15409136e+00 3.81563097e-01 -8.90677035e-01 1.16069472e+00 -1.92103997e-01 -5.93134880e-01 8.48770082e-01 2.38297150e-01 4.87560779e-01 7.79478312e-01 -1.43719840e+00 3.79712582e-01 1.00623965e+00 4.61344607e-02 3.40356797e-01 1.04823709e+00 -5.66463232e-01 -1.36281800e+00 -1.10201955e+00 -2.80392587e-01 -2.46125922e-01 1.18567467e+00 -1.35464333e-02 -8.09129477e-01 1.01880765e+00 8.04109335e-01 -4.76170443e-02 5.21746576e-01 -2.42770344e-01 -4.11448210e-01 -2.37974450e-02 -1.36790776e+00 1.01210856e+00 1.58875823e-01 -5.43631256e-01 -6.94958329e-01 2.01524198e-01 7.87687361e-01 -4.46353793e-01 -1.06303358e+00 2.42471442e-01 1.40081972e-01 -7.75446296e-01 9.10070479e-01 -1.14512348e+00 2.32129917e-01 -4.54611033e-01 -2.78394837e-02 -2.10028744e+00 -2.21107751e-01 -1.03664613e+00 -2.81012177e-01 7.81712532e-01 -2.24131029e-02 -8.33488226e-01 6.63985252e-01 -4.99703102e-02 4.65820730e-02 -4.33143109e-01 -1.00100875e+00 -1.02031529e+00 7.04334915e-01 3.56387086e-02 5.39097130e-01 9.26622093e-01 2.19599307e-01 -8.26514512e-03 -8.08750212e-01 6.93307281e-01 7.56235421e-01 -1.88351989e-01 1.21787548e+00 -5.72312236e-01 -6.83611095e-01 -3.15950364e-01 -4.25095707e-01 -3.85498971e-01 5.37283659e-01 -5.46799898e-01 -6.80646002e-02 -8.54058862e-01 -2.59013861e-01 -6.05001561e-02 -4.05260861e-01 7.20029712e-01 -2.27587938e-01 8.37124288e-02 6.18520498e-01 -1.27718478e-01 -7.99278796e-01 6.96396649e-01 1.19061089e+00 -2.84027904e-01 -5.80873489e-02 3.15998971e-01 -3.84987801e-01 6.28627956e-01 1.34541559e+00 -7.55468845e-01 -3.75467509e-01 9.91129056e-02 -1.62066698e-01 3.45610887e-01 4.30942744e-01 -1.18563449e+00 1.59943327e-01 -4.76900429e-01 1.05639689e-01 -1.78461865e-01 1.28334954e-01 -8.51867437e-01 -2.79332817e-01 1.21041071e+00 -5.86225629e-01 4.85157698e-01 3.69477153e-01 4.66728985e-01 -2.31576562e-01 -4.45735961e-01 1.36794484e+00 -1.58204392e-01 -5.34000278e-01 3.81743997e-01 -9.54716921e-01 1.66964203e-01 1.46596515e+00 3.93549740e-01 -6.51015818e-01 -5.18281043e-01 -7.80302942e-01 5.36621451e-01 3.58120292e-01 3.98533940e-01 6.84595048e-01 -1.22521782e+00 -8.85922909e-01 1.93993211e-01 -9.72996429e-02 -5.25516748e-01 1.59633994e-01 2.85901815e-01 -4.14512604e-01 -9.87376869e-02 -8.12849402e-01 -7.05972640e-03 -1.30706275e+00 1.17132330e+00 8.04215670e-01 -7.77039826e-01 -4.53625798e-01 1.28471941e-01 -3.42541844e-01 -5.85052013e-01 3.24466825e-01 3.82990718e-01 -2.54325032e-01 -3.75758290e-01 6.45506978e-01 3.76347691e-01 -3.13536793e-01 -2.07288027e-01 -2.12984428e-01 8.93112719e-02 -5.76111019e-01 -4.49195445e-01 1.24015021e+00 3.03082705e-01 1.63458407e-01 8.58997852e-02 9.05057609e-01 1.08202048e-01 -1.60733342e+00 -1.62801236e-01 -1.79172516e-01 -8.20191950e-02 -3.87586802e-01 -7.25204647e-01 -1.01931798e+00 6.16566718e-01 6.32916272e-01 6.69353485e-01 7.72741675e-01 -4.18963462e-01 3.47132236e-01 7.26665676e-01 5.96067488e-01 -1.16321635e+00 6.75728917e-01 8.65334570e-01 8.97075713e-01 -1.10962749e+00 -2.12077592e-02 4.06730652e-01 -8.37096393e-01 1.15879369e+00 8.76583874e-01 -1.02831328e+00 5.73227286e-01 5.61685503e-01 4.19545501e-01 2.33048722e-02 -8.88923466e-01 1.69959739e-02 -5.47594070e-01 1.06011140e+00 -3.84713858e-01 -6.05300143e-02 1.68435842e-01 -1.33900970e-01 -1.03013672e-01 -3.78245711e-01 1.08521235e+00 1.11026895e+00 -5.15723228e-01 -1.48141563e+00 -5.90036929e-01 1.19272433e-02 -5.47976077e-01 2.89092779e-01 -2.90694952e-01 1.12037575e+00 -3.46770316e-01 1.15844917e+00 -3.88826162e-01 -3.29820305e-01 3.20286483e-01 -3.59314650e-01 4.95252103e-01 -2.37846062e-01 -1.03058350e+00 -3.00292015e-01 -9.88266319e-02 -6.10601068e-01 -4.45630908e-01 -4.29702848e-01 -1.32147145e+00 -6.69556618e-01 7.85909593e-02 3.22388172e-01 2.09018499e-01 9.33368087e-01 -1.48484141e-01 6.96693420e-01 1.13330472e+00 -7.39653170e-01 -1.44024420e+00 -7.86277413e-01 -7.54561007e-01 4.36250806e-01 6.95772111e-01 -6.19067788e-01 -3.70893270e-01 -6.83435678e-01]	[3.848306655883789, 2.2773351669311523]
d5fef5ec-ef20-4a59-a326-ed246c1ec5ae	fast-interactive-search-with-a-scale-free	2306.01814	null	https://arxiv.org/abs/2306.01814v1	https://arxiv.org/pdf/2306.01814v1.pdf	Fast Interactive Search with a Scale-Free Comparison Oracle	A comparison-based search algorithm lets a user find a target item $t$ in a database by answering queries of the form, ``Which of items $i$ and $j$ is closer to $t$?'' Instead of formulating an explicit query (such as one or several keywords), the user navigates towards the target via a sequence of such (typically noisy) queries. We propose a scale-free probabilistic oracle model called $\gamma$-CKL for such similarity triplets $(i,j;t)$, which generalizes the CKL triplet model proposed in the literature. The generalization affords independent control over the discriminating power of the oracle and the dimension of the feature space containing the items. We develop a search algorithm with provably exponential rate of convergence under the $\gamma$-CKL oracle, thanks to a backtracking strategy that deals with the unavoidable errors in updating the belief region around the target. We evaluate the performance of the algorithm both over the posited oracle and over several real-world triplet datasets. We also report on a comprehensive user study, where human subjects navigate a database of face portraits.	['Matthias Grossglauser', 'Lucas Maystre', 'Lars Klein', 'Daniyar Chumbalov']	2023-06-02	null	null	null	null	['navigate']	['reasoning']	[ 1.14128321e-01 -1.37628198e-01 -2.88012564e-01 -5.07649541e-01 -1.47459221e+00 -9.52238619e-01 9.81680304e-02 3.66836250e-01 -6.32697523e-01 4.95390505e-01 -4.30120200e-01 -2.35611320e-01 -7.01385379e-01 -7.55015373e-01 -8.17232192e-01 -5.54315507e-01 -2.21425727e-01 1.05963969e+00 2.62869269e-01 -2.33111128e-01 3.30501288e-01 1.72818571e-01 -1.47400939e+00 -1.64608330e-01 4.48073953e-01 1.59014368e+00 -1.85729284e-02 4.62192118e-01 9.56270993e-02 3.24383646e-01 -5.63066006e-01 -9.50598001e-01 4.61882412e-01 -2.49037400e-01 -9.65719879e-01 -1.15089202e-02 4.20097709e-01 -6.74996302e-02 -1.52746132e-02 1.25610816e+00 3.81371945e-01 1.52318165e-01 5.14756024e-01 -1.27667427e+00 -5.69868267e-01 4.58998531e-01 -4.33925152e-01 3.02124530e-01 8.18464518e-01 -5.80513477e-02 1.65286529e+00 -9.84517097e-01 6.83760405e-01 1.03446937e+00 6.00789785e-01 1.34397984e-01 -1.35009718e+00 -4.04739797e-01 1.78496212e-01 3.96973908e-01 -1.84272265e+00 -3.31365913e-01 5.62668085e-01 -1.36818379e-01 8.05514157e-01 6.22969389e-01 4.57457453e-01 7.01447785e-01 -1.21416651e-01 7.41034269e-01 8.85157526e-01 -5.10211408e-01 4.93154585e-01 3.56865138e-01 1.54085532e-01 9.70321834e-01 1.41192645e-01 3.03585947e-01 -1.11303639e+00 -6.86098874e-01 2.84448534e-01 -2.63037950e-01 -1.38664216e-01 -7.30538368e-01 -9.51971591e-01 7.51694083e-01 3.89614880e-01 1.20823728e-02 -3.18268955e-01 2.14906722e-01 1.32052843e-02 5.49329460e-01 2.16993000e-02 4.22583997e-01 -5.26431620e-01 -1.36566356e-01 -7.17392385e-01 3.25547457e-01 9.66295660e-01 1.21654940e+00 9.28407669e-01 -6.97276711e-01 -3.27623077e-02 7.23672271e-01 3.92718941e-01 7.17700601e-01 2.05575079e-01 -1.12981570e+00 4.86021429e-01 4.73812044e-01 5.32840431e-01 -9.36598480e-01 2.99038365e-02 -3.03810418e-01 -2.71494359e-01 -1.19412050e-01 4.86827374e-01 2.48965561e-01 -4.73071396e-01 2.10212064e+00 5.18606484e-01 -2.96276987e-01 -2.31789902e-01 8.17148864e-01 1.36566371e-01 1.57315537e-01 -3.43068361e-01 -3.95849407e-01 1.31752563e+00 -4.95230317e-01 -4.17601466e-01 -9.59789157e-02 5.34287632e-01 -6.61996901e-01 1.21260786e+00 5.10637164e-01 -1.29696071e+00 -2.04582036e-01 -9.83617067e-01 -1.14494199e-02 -4.70077187e-01 -2.80936986e-01 4.78008300e-01 8.89942706e-01 -1.13764310e+00 3.54049593e-01 -5.25937438e-01 -3.23744535e-01 1.08851328e-01 5.89894235e-01 -3.21806222e-01 -4.09486324e-01 -1.28460920e+00 8.27177644e-01 1.74442753e-01 1.53713435e-01 -7.91650832e-01 -2.36962140e-01 -6.27721548e-01 -1.45153478e-01 9.80273187e-01 -7.73928821e-01 1.40951729e+00 -5.89391232e-01 -9.79317427e-01 9.98011470e-01 -3.49071085e-01 -2.70766139e-01 5.28014898e-01 4.42994684e-02 -9.50235948e-02 8.25079978e-02 4.83994901e-01 3.65821689e-01 8.28348994e-01 -9.40181613e-01 -7.15662658e-01 -7.58099616e-01 2.69841880e-01 5.45431316e-01 6.45043105e-02 -4.53572571e-02 -8.30002546e-01 -3.94431472e-01 4.48796660e-01 -9.58841264e-01 -1.77211300e-01 2.92561024e-01 -3.84991348e-01 -5.02830327e-01 3.49441767e-01 -1.60781354e-01 1.38056684e+00 -2.09420609e+00 -8.85467045e-03 9.00630653e-01 3.38324793e-02 -3.12922567e-01 7.86460862e-02 5.01165509e-01 1.60677493e-01 -9.04584583e-03 -3.59267294e-01 -3.85424465e-01 3.22103411e-01 4.35482740e-01 -2.33867377e-01 5.99067628e-01 -4.15997654e-01 6.54651821e-01 -8.67235959e-01 -6.05085313e-01 -3.36278975e-01 -9.53189880e-02 -6.52644873e-01 -1.88608782e-03 -3.93795431e-01 -3.41602564e-01 -5.24530828e-01 9.23260093e-01 3.49152148e-01 -4.59356457e-01 2.91453391e-01 1.69987772e-02 2.28499487e-01 1.30350202e-01 -1.49854958e+00 1.75290179e+00 -3.05037890e-02 -2.69172597e-04 1.27410695e-01 -7.55683362e-01 4.93661672e-01 5.49129210e-03 2.93247044e-01 -6.75762892e-01 -3.58639546e-02 3.26321602e-01 -3.33254516e-01 -1.97642788e-01 3.48890185e-01 -2.00737000e-01 -4.72594619e-01 6.29611194e-01 -1.39304712e-01 -2.12939054e-01 2.43650422e-01 3.55738968e-01 1.13305759e+00 -3.49505037e-01 1.27220303e-01 -2.00260997e-01 4.05530632e-01 -8.19782354e-03 3.63536865e-01 1.26948106e+00 -2.01139659e-01 2.29644865e-01 3.57833475e-01 -2.97844648e-01 -5.16589701e-01 -1.28818393e+00 -5.67817278e-02 1.39638448e+00 5.71194291e-01 -5.17673135e-01 -6.22318327e-01 -6.81277335e-01 2.29831234e-01 4.49104607e-01 -7.34759569e-01 -6.25025854e-02 -2.63190925e-01 -4.26963061e-01 4.60252255e-01 3.27943265e-01 4.65696275e-01 -7.06044376e-01 -7.17846334e-01 7.85878599e-02 -4.53707546e-01 -6.71604514e-01 -1.15822518e+00 3.16283286e-01 -6.23822093e-01 -1.36989284e+00 -1.81052238e-01 -6.64627731e-01 6.47780716e-01 2.40347404e-02 1.20060408e+00 2.19379831e-02 -4.89277542e-01 7.78804243e-01 -7.22896382e-02 -3.40323374e-02 1.09719940e-01 -1.91623792e-01 1.00895420e-01 5.07255010e-02 3.75018120e-01 -2.05833524e-01 -6.96306586e-01 6.11911476e-01 -7.72820354e-01 -6.11766517e-01 4.56262439e-01 9.13310051e-01 9.81967568e-01 2.92314082e-01 7.69490749e-02 -8.83260369e-01 8.00843179e-01 -4.39688325e-01 -8.44527602e-01 7.11212754e-01 -1.03175306e+00 4.84495670e-01 2.34359622e-01 -3.97706181e-01 -4.70392942e-01 1.51777357e-01 2.12802980e-02 -3.92702013e-01 2.83016086e-01 7.04780161e-01 -2.49607921e-01 -9.22384337e-02 8.34015012e-01 4.00681078e-01 -2.23534986e-01 -4.20787007e-01 4.38716412e-01 3.88909727e-01 7.98630416e-01 -9.57997978e-01 6.76570296e-01 3.05446327e-01 -3.57471267e-03 -3.02712291e-01 -5.61574042e-01 -5.73210776e-01 -1.81371450e-01 2.71507204e-02 2.76549786e-01 -5.28935075e-01 -1.23910105e+00 7.45413899e-02 -7.06756115e-01 3.23978886e-02 -4.59842026e-01 1.05555080e-01 -8.03874731e-01 5.11735141e-01 -2.05256119e-01 -8.92318428e-01 -2.10911229e-01 -1.15266895e+00 8.84485245e-01 -1.35950014e-01 -3.57235402e-01 -7.02416599e-01 1.12674840e-01 3.92608851e-01 2.83719063e-01 -1.23768196e-01 1.21320426e+00 -8.23573649e-01 -9.06036377e-01 -5.27149022e-01 -8.92987568e-03 -5.57055771e-02 -2.04049036e-01 -4.74725068e-01 -3.91158938e-01 -3.02385092e-01 1.65698707e-01 -4.32371467e-01 3.98991853e-01 -2.14094184e-02 1.03255796e+00 -6.41328514e-01 -5.07934332e-01 3.42682749e-01 1.39301729e+00 3.78564149e-01 8.22857320e-02 1.81047261e-01 -2.12927647e-02 3.41812879e-01 8.69469583e-01 5.30334771e-01 5.41530371e-01 7.50046194e-01 4.46311980e-01 5.16535223e-01 4.85496193e-01 -4.58345920e-01 -4.15809676e-02 1.06627353e-01 2.21830696e-01 -3.35015386e-01 -8.30317974e-01 5.65987468e-01 -1.75543916e+00 -7.13067949e-01 7.45150208e-01 2.66064596e+00 1.10705209e+00 3.03147852e-01 2.35373467e-01 1.88793719e-01 5.76743960e-01 -1.21542349e-01 -1.01020896e+00 -1.73745841e-01 -5.08526824e-02 2.94954002e-01 5.77192843e-01 7.47392595e-01 -6.72724843e-01 6.11654818e-01 6.99823999e+00 9.70332801e-01 -7.71024942e-01 -2.39817821e-03 6.43043280e-01 -2.42241532e-01 -4.74854767e-01 6.44223168e-02 -7.88870156e-01 4.79157567e-01 7.75751293e-01 -2.17060924e-01 8.82476032e-01 1.05133522e+00 -4.81707275e-01 -5.19585848e-01 -1.55299723e+00 1.36745429e+00 1.38658673e-01 -1.07667601e+00 -1.07999966e-01 1.08946264e-01 1.65426016e-01 -1.12402193e-01 5.05754113e-01 1.02777392e-01 4.85485613e-01 -8.17604721e-01 9.60905612e-01 3.13280553e-01 9.38373208e-01 -7.11691141e-01 2.74549305e-01 6.06942534e-01 -1.04213142e+00 -2.28858888e-01 2.19273381e-02 4.48675364e-01 -1.00055918e-01 1.51866138e-01 -8.04481328e-01 3.03675026e-01 1.01445627e+00 8.24465007e-02 -5.90387106e-01 9.55661416e-01 6.44470900e-02 1.32969990e-01 -8.31468523e-01 -1.70309499e-01 1.01936944e-01 -2.27739632e-01 5.08119047e-01 8.14415514e-01 3.02911967e-01 3.32551122e-01 1.52700439e-01 6.04554236e-01 -2.60976851e-01 1.33102924e-01 -5.22892118e-01 3.57086696e-02 8.87929916e-01 6.31955624e-01 -5.63122392e-01 -3.55470389e-01 2.80667059e-02 9.34257865e-01 4.38184232e-01 1.81987107e-01 -5.56312859e-01 -3.17894995e-01 4.63527948e-01 1.55834436e-01 4.24909443e-01 -1.13441311e-02 -8.32352042e-02 -8.17247808e-01 3.62106979e-01 -1.06735516e+00 1.07664323e+00 -6.74404263e-01 -1.36942804e+00 5.78488231e-01 2.16605932e-01 -7.98129022e-01 -4.28090394e-01 -3.94621670e-01 1.58552490e-02 8.47866416e-01 -1.01695061e+00 -4.75560814e-01 2.48019367e-01 9.88676131e-01 2.65017837e-01 1.98785499e-01 9.08793867e-01 -4.62242812e-02 -2.13521808e-01 9.51533794e-01 -2.18859073e-02 -1.78467408e-01 6.51463568e-01 -1.29520607e+00 2.28579789e-02 6.54190361e-01 3.77306938e-01 1.09626234e+00 8.26796353e-01 -4.01243657e-01 -1.70006561e+00 -4.68693912e-01 1.02635443e+00 -6.88062847e-01 5.19854426e-01 -2.97195107e-01 -4.52979743e-01 6.38398349e-01 -7.35316202e-02 8.39603916e-02 9.57085311e-01 2.92454064e-01 -7.36738741e-01 -4.16056156e-01 -1.53277600e+00 3.27933580e-01 1.17749822e+00 -8.92490923e-01 -4.86800939e-01 4.38294798e-01 4.33966905e-01 -4.76039082e-01 -8.09060633e-01 2.99890041e-01 7.67564058e-01 -1.01100492e+00 1.03707325e+00 -5.06273508e-01 -4.06349421e-01 -3.35746557e-01 -6.98554635e-01 -8.15307140e-01 -1.27727360e-01 -8.24471772e-01 -1.62349105e-01 4.22604591e-01 6.20348275e-01 -5.31953216e-01 8.97177279e-01 1.10261631e+00 3.74404460e-01 -9.31894302e-01 -1.49547052e+00 -6.09850764e-01 -3.35242897e-01 -5.41499734e-01 6.45034790e-01 5.04233599e-01 1.77964315e-01 2.37398610e-01 -1.24294721e-01 2.67091244e-01 8.65265310e-01 4.14248586e-01 3.15016627e-01 -9.40685689e-01 -6.70360267e-01 -2.82474905e-01 -1.15857191e-01 -1.42655528e+00 -2.50998586e-01 -6.70800030e-01 2.88989812e-01 -8.96218717e-01 1.89910486e-01 -8.08862448e-01 -4.00559098e-01 4.72400218e-01 2.86340192e-02 1.53496131e-01 -8.63033012e-02 2.28467762e-01 -1.15323138e+00 1.44613281e-01 7.91708529e-01 -1.08410656e-01 1.21472538e-01 3.21249932e-01 -8.23093593e-01 4.52323258e-01 2.04920501e-01 -5.47490299e-01 -5.45413435e-01 -3.92766476e-01 7.60767519e-01 6.10165417e-01 1.17425688e-01 -4.64713126e-01 8.11267495e-01 -7.69093707e-02 6.71656579e-02 -7.64502883e-01 6.46363199e-01 -8.81418943e-01 2.41316646e-01 4.39565450e-01 -6.82353556e-01 4.45161521e-01 -2.60716259e-01 9.22126710e-01 -4.04737890e-02 -1.41385451e-01 6.31514311e-01 -1.84398234e-01 -3.99082422e-01 5.55300951e-01 -1.26970306e-01 1.59929708e-01 8.14861953e-01 -2.19194651e-01 -8.48134607e-02 -4.47431803e-01 -9.39334393e-01 3.26549977e-01 4.95381713e-01 5.25852777e-02 7.17605412e-01 -1.09938145e+00 -3.33866477e-02 3.35877925e-01 4.53903317e-01 -4.74430695e-02 -1.54717416e-01 6.33650303e-01 -8.95294398e-02 4.73323584e-01 3.71968895e-01 -5.90138435e-01 -1.24222219e+00 8.58126819e-01 4.01037186e-01 -2.60681987e-01 1.41352057e-01 1.21915781e+00 -2.93037981e-01 -3.16469342e-01 6.18308425e-01 -2.40425169e-01 4.35658425e-01 -2.31190436e-02 2.91081399e-01 2.63945013e-01 1.74445346e-01 -5.38832426e-01 -7.02803135e-01 4.08060193e-01 -2.26684406e-01 -6.05651200e-01 7.79600203e-01 -4.17939484e-01 -1.48981184e-01 3.83182436e-01 1.18991256e+00 -3.38628329e-02 -7.62725711e-01 -7.38739014e-01 3.00617397e-01 -7.80699372e-01 -3.89683962e-01 -1.01674259e+00 -5.91777384e-01 2.22857848e-01 7.47144699e-01 2.20661744e-01 1.15441239e+00 3.70824516e-01 4.85725492e-01 8.69148612e-01 9.77467716e-01 -9.42086995e-01 1.24481283e-01 1.32284671e-01 6.16908014e-01 -1.16658902e+00 -3.21174711e-01 -1.71336696e-01 -3.87637824e-01 6.32392645e-01 2.24330485e-01 7.56873265e-02 7.69153893e-01 5.45792393e-02 -1.22329697e-01 -3.08224618e-01 -9.93527770e-01 -1.37703732e-01 1.65498301e-01 2.58166045e-01 -2.18695834e-01 -2.76392628e-03 -1.46898061e-01 4.85009909e-01 -3.08817565e-01 -8.09510425e-02 -1.96728855e-01 1.20763886e+00 -4.90724385e-01 -1.32428575e+00 -3.92798692e-01 4.04637367e-01 -3.30848545e-01 -2.05713168e-01 -4.16609019e-01 5.39058805e-01 1.57420695e-01 1.04015124e+00 -1.91849858e-01 -3.22445303e-01 3.54893506e-01 1.58095643e-01 6.01628363e-01 -4.71599340e-01 -4.01959062e-01 -7.09081963e-02 -1.37794435e-01 -7.75520682e-01 -1.89165458e-01 -9.25591290e-01 -9.69809413e-01 -1.88863009e-01 -4.04199630e-01 7.27225304e-01 5.64392865e-01 9.15643871e-01 2.54005164e-01 -5.14223576e-01 7.34820962e-01 4.61751013e-04 -9.92933035e-01 -5.53624272e-01 -6.21617377e-01 2.84192950e-01 3.15407276e-01 -5.92434824e-01 -3.36185545e-01 -9.91576687e-02]	[6.59250020980835, 4.523097991943359]
b4c12f78-b24b-4778-83d4-b2d2c59d3c02	selective-encoding-for-abstractive-sentence	1704.07073	null	http://arxiv.org/abs/1704.07073v1	http://arxiv.org/pdf/1704.07073v1.pdf	Selective Encoding for Abstractive Sentence Summarization	We propose a selective encoding model to extend the sequence-to-sequence framework for abstractive sentence summarization. It consists of a sentence encoder, a selective gate network, and an attention equipped decoder. The sentence encoder and decoder are built with recurrent neural networks. The selective gate network constructs a second level sentence representation by controlling the information flow from encoder to decoder. The second level representation is tailored for sentence summarization task, which leads to better performance. We evaluate our model on the English Gigaword, DUC 2004 and MSR abstractive sentence summarization datasets. The experimental results show that the proposed selective encoding model outperforms the state-of-the-art baseline models.	['Nan Yang', 'Furu Wei', 'Qingyu Zhou', 'Ming Zhou']	2017-04-24	selective-encoding-for-abstractive-sentence-1	https://aclanthology.org/P17-1101	https://aclanthology.org/P17-1101.pdf	acl-2017-7	['abstractive-sentence-summarization']	['natural-language-processing']	[ 5.81673622e-01 4.88968730e-01 -1.06973149e-01 -3.29451174e-01 -9.96989906e-01 -1.56736210e-01 4.49060827e-01 4.12228823e-01 -4.13293362e-01 9.09245372e-01 1.27322519e+00 -8.72599036e-02 5.62140465e-01 -5.96128762e-01 -6.86096430e-01 -3.14817607e-01 2.10028633e-01 1.32566333e-01 1.39580190e-01 -5.18792927e-01 7.11289465e-01 -1.48072824e-01 -1.03250325e+00 6.95563972e-01 1.20311069e+00 6.22766554e-01 4.87370610e-01 1.24740231e+00 -2.22611055e-01 1.43599212e+00 -1.08665359e+00 -4.62705523e-01 -3.87107909e-01 -9.70910072e-01 -9.93789554e-01 -7.32371733e-02 2.59807169e-01 -4.30098623e-01 -7.89936125e-01 9.78686810e-01 8.48777950e-01 3.21042359e-01 5.17426312e-01 -3.13692093e-01 -1.06622267e+00 1.11405492e+00 -3.88800800e-01 6.66466236e-01 6.33936286e-01 -5.47052780e-03 1.19202065e+00 -4.88332927e-01 5.98794639e-01 1.33378649e+00 3.18697065e-01 7.25308001e-01 -8.21936548e-01 -3.62250283e-02 2.91038185e-01 1.12702541e-01 -7.14589119e-01 -7.31195271e-01 7.39945054e-01 -6.49289489e-02 1.73405826e+00 4.35472965e-01 6.39216542e-01 1.09244108e+00 8.60636473e-01 1.23372436e+00 9.59732756e-02 -2.49481753e-01 1.34255230e-01 -3.81598145e-01 7.31189191e-01 4.81297284e-01 1.65473700e-01 -6.11848652e-01 -5.43422580e-01 4.74258251e-02 3.71826798e-01 -9.08627808e-02 -3.99429888e-01 3.97834003e-01 -7.85696447e-01 9.08817410e-01 4.93421078e-01 8.65337625e-02 -6.94891632e-01 1.68587387e-01 1.16156209e+00 4.76922989e-01 9.57241297e-01 5.43269575e-01 -2.22304136e-01 -2.78282464e-01 -1.09826195e+00 2.76205778e-01 9.41234529e-01 1.18694985e+00 1.41331330e-01 3.55427325e-01 -1.09571648e+00 9.77337480e-01 -3.55150513e-02 1.29793763e-01 8.67380202e-01 -6.52243078e-01 1.11789024e+00 7.07520962e-01 -2.76951700e-01 -8.00797045e-01 -1.74999297e-01 -6.52303517e-01 -1.27046847e+00 -8.44966412e-01 -7.98770010e-01 -2.68055201e-01 -6.56624317e-01 1.38240135e+00 -4.23000336e-01 -5.06002046e-02 5.05386829e-01 6.83348060e-01 1.58399129e+00 1.12163615e+00 -1.63170725e-01 -6.23715460e-01 1.20824492e+00 -1.57247353e+00 -1.08006716e+00 -4.36310887e-01 7.32692301e-01 -4.06989336e-01 9.24595416e-01 -5.62638156e-02 -1.65314817e+00 -5.85659921e-01 -1.24232626e+00 -6.32807732e-01 1.74248502e-01 3.74134481e-01 2.19910190e-01 2.47965176e-02 -1.12996030e+00 5.90259850e-01 -6.76773965e-01 -5.04461944e-01 4.07343745e-01 1.34844363e-01 -1.84214681e-01 2.72094935e-01 -1.15502286e+00 8.53924811e-01 8.76330853e-01 1.38898730e-01 -8.04453373e-01 -2.74264753e-01 -1.21687841e+00 6.27403438e-01 1.10958228e-02 -1.32834435e+00 1.59460402e+00 -6.67912960e-01 -1.93292928e+00 7.12738812e-01 -4.03012633e-01 -1.06383586e+00 1.22474223e-01 -4.32115257e-01 -1.84457153e-02 3.23392272e-01 1.65675208e-01 3.69335771e-01 5.97149849e-01 -6.65045381e-01 -3.41657817e-01 -1.64879203e-01 6.09635264e-02 5.73542476e-01 -1.53852612e-01 1.94683805e-01 -3.12438339e-01 -8.33811164e-01 -3.02853554e-01 -2.92898595e-01 -3.43762726e-01 -1.17587197e+00 -1.03467226e+00 -1.55039117e-01 5.80727041e-01 -1.05174887e+00 1.87807429e+00 -1.80804336e+00 6.86327934e-01 -8.61001551e-01 1.25849165e-03 4.33622539e-01 -3.89197856e-01 9.88692880e-01 -1.14194356e-01 -5.91040552e-02 -5.19819140e-01 -8.07744622e-01 -2.80123055e-01 1.02711720e-02 -6.12656176e-01 3.90757397e-02 2.68674046e-01 1.28935075e+00 -9.41588700e-01 -5.13437986e-01 -1.87764131e-02 -4.37636562e-02 -6.27465546e-01 6.18357122e-01 -2.78004766e-01 2.20808044e-01 -5.33168793e-01 1.46190464e-01 3.43227237e-01 -4.95190620e-02 -1.38238017e-02 1.50091708e-01 -1.17963232e-01 1.12752187e+00 -3.24825734e-01 2.17078662e+00 -2.28473052e-01 6.29418790e-01 -1.77254349e-01 -1.09060693e+00 1.12892747e+00 3.85256588e-01 -2.54168153e-01 -5.43260634e-01 4.38203603e-01 -1.37611151e-01 -1.66869983e-01 -6.74796104e-01 1.36782551e+00 6.14840984e-02 -4.46242094e-01 4.05429095e-01 3.62132281e-01 -3.69957387e-01 4.61034894e-01 7.81384349e-01 1.26638699e+00 -1.57150969e-01 6.83152020e-01 -3.11756402e-01 8.97265077e-01 -2.30859324e-01 5.49741685e-01 7.75725722e-01 1.19223937e-01 7.59228885e-01 9.97061849e-01 -2.14813054e-01 -1.05328262e+00 -7.11161792e-01 3.82816613e-01 1.08097315e+00 -2.21974477e-02 -9.10772383e-01 -1.07731593e+00 -6.03467703e-01 -4.27760392e-01 1.19346702e+00 -5.67785561e-01 -6.54268563e-01 -7.33194768e-01 -4.76991653e-01 7.13215888e-01 6.51417315e-01 8.55609298e-01 -1.44014883e+00 -3.94820958e-01 3.85865271e-01 -5.44875920e-01 -9.23426270e-01 -9.26760614e-01 -5.93894199e-02 -1.04847014e+00 -5.66012979e-01 -5.23541749e-01 -9.17022824e-01 3.61676455e-01 1.91852435e-01 1.18369901e+00 -5.57856262e-02 1.08702764e-01 -3.23191471e-02 -5.72138131e-01 -5.08937061e-01 -7.43383169e-01 7.72467971e-01 -3.92269939e-01 -1.80776715e-01 6.86588958e-02 -5.97029328e-01 -5.51476002e-01 -6.16821587e-01 -9.43853796e-01 3.78251553e-01 5.87895513e-01 8.38592231e-01 2.97384173e-01 -7.47718036e-01 1.04812932e+00 -1.10023642e+00 1.52490461e+00 -4.75600928e-01 1.60003677e-02 3.01694721e-01 1.25884399e-01 2.60576636e-01 9.79297042e-01 2.68955439e-01 -1.21009684e+00 -3.82050008e-01 -4.93964702e-01 1.91276222e-01 2.57833719e-01 7.85199881e-01 -2.01357782e-01 7.54860401e-01 4.81254190e-01 7.29627550e-01 -1.30242109e-01 -6.05394840e-01 3.70107710e-01 1.18002582e+00 7.84118831e-01 -1.28076315e-01 -3.60267423e-02 -1.58228338e-01 -4.25515771e-01 -1.03865242e+00 -1.23361397e+00 -2.88386226e-01 -6.08738244e-01 5.01573086e-02 9.58180428e-01 -9.81294930e-01 -3.87424141e-01 2.75907904e-01 -1.81521845e+00 -5.67390257e-03 -6.46915197e-01 1.89203471e-01 -8.47821116e-01 7.21658766e-01 -1.19792283e+00 -5.59758663e-01 -1.38871050e+00 -1.00260806e+00 1.28678250e+00 4.58247274e-01 -3.43975991e-01 -8.61255586e-01 2.96693355e-01 3.72417241e-01 3.70120764e-01 2.95573249e-02 8.44615519e-01 -8.52292836e-01 -1.65699884e-01 -3.00627649e-01 -8.61415341e-02 6.37557626e-01 -5.08245714e-02 -2.36677989e-01 -5.29028118e-01 -3.04175854e-01 2.47663319e-01 -4.56569195e-01 1.74878752e+00 5.84399462e-01 1.22326851e+00 -7.13498652e-01 1.28208119e-02 4.43766862e-01 1.00494266e+00 -3.69743668e-02 1.13818109e+00 1.11056745e-01 6.16788924e-01 1.91300333e-01 3.33154321e-01 5.33833861e-01 5.79936624e-01 2.74329573e-01 2.69181579e-01 2.54291952e-01 -1.96612388e-01 -5.22344530e-01 6.30153596e-01 1.74686050e+00 2.94982605e-02 -7.37960696e-01 -3.36798996e-01 5.10869026e-01 -2.31624365e+00 -1.23645830e+00 -1.93105131e-01 1.77747166e+00 9.46539581e-01 1.57113984e-01 -5.41358478e-02 -2.02761963e-01 6.37969017e-01 8.04596961e-01 -4.14045513e-01 -1.37310815e+00 -2.76461601e-01 1.05150938e-01 1.32246181e-01 6.10566318e-01 -9.43738043e-01 1.28790009e+00 6.42654085e+00 7.86710978e-01 -7.58877635e-01 -4.10865657e-02 4.26034212e-01 -1.70694470e-01 -2.63554662e-01 -1.99202165e-01 -7.17917562e-01 5.51332235e-01 1.14102650e+00 -6.52126729e-01 -3.38149071e-02 5.12098908e-01 4.06542838e-01 8.82321894e-02 -9.24049199e-01 6.92086160e-01 6.10005915e-01 -1.60093057e+00 6.83637559e-01 -4.82963175e-01 6.76757097e-01 7.17571080e-02 -4.78412211e-01 5.62444150e-01 2.34877378e-01 -9.05884027e-01 4.90630269e-01 8.72279584e-01 5.29961884e-01 -8.22368205e-01 9.39566731e-01 6.10885680e-01 -1.01275790e+00 -7.88822994e-02 -7.79882848e-01 -4.20881093e-01 5.56987107e-01 3.97641867e-01 -4.40472275e-01 1.13623559e+00 3.39846946e-02 1.36452520e+00 -5.20989120e-01 7.92126834e-01 -5.78924835e-01 7.31095016e-01 4.86868531e-01 -2.74159729e-01 2.99238473e-01 -3.27536404e-01 9.15253818e-01 1.73128581e+00 -4.76971548e-03 2.02448875e-01 2.49635894e-02 5.65723360e-01 -5.50941288e-01 1.67581365e-01 -6.70251787e-01 -1.74768403e-01 2.81836689e-01 9.39730227e-01 -1.32264689e-01 -7.05369294e-01 -1.90835521e-01 1.41306233e+00 6.88201368e-01 3.15533370e-01 -5.87350309e-01 -8.81798744e-01 2.72000819e-01 -4.38288957e-01 4.22911942e-01 -1.23786651e-01 -3.44378054e-01 -1.62128782e+00 5.03079547e-03 -7.43218780e-01 3.06972623e-01 -8.04095328e-01 -1.03960955e+00 7.98572183e-01 -2.56504685e-01 -9.33701038e-01 -3.90954077e-01 1.10903330e-01 -1.27947176e+00 7.76330531e-01 -1.33218813e+00 -9.81170297e-01 3.18791084e-02 3.42805684e-02 1.35578573e+00 -3.16832215e-01 9.35165048e-01 -5.31422906e-02 -1.03449929e+00 3.76524925e-01 1.63805798e-01 1.35920569e-01 2.94319391e-01 -1.16108024e+00 9.79187906e-01 9.24843967e-01 -4.38218653e-01 6.04740858e-01 6.94750547e-01 -7.36210108e-01 -1.21396422e+00 -1.40659392e+00 1.15301943e+00 -7.00550899e-02 3.29060435e-01 -3.81429076e-01 -8.89791429e-01 9.69654381e-01 9.89581704e-01 -7.85230279e-01 5.76723754e-01 -1.19521409e-01 5.05723506e-02 1.02097444e-01 -5.86868644e-01 6.27439976e-01 9.81273115e-01 -3.50483418e-01 -1.38063562e+00 4.26866859e-01 1.35250211e+00 -4.80504274e-01 -5.06992340e-01 2.00825676e-01 1.65131122e-01 -6.90319717e-01 5.91612101e-01 -9.39378142e-01 1.28632975e+00 1.66652739e-01 1.99093726e-02 -1.79042530e+00 -4.86224890e-01 -6.87366068e-01 -4.58815783e-01 1.39543319e+00 3.83756042e-01 -4.00619835e-01 4.50082242e-01 -3.72312940e-03 -9.41552401e-01 -8.91191244e-01 -7.80280650e-01 -5.85898757e-01 9.59652290e-02 2.38658369e-01 5.17004967e-01 3.26191336e-01 4.09210294e-01 1.49653482e+00 -4.74462092e-01 -4.32392657e-01 2.78385222e-01 1.26498058e-01 6.94471121e-01 -8.16245615e-01 -1.87477604e-01 -6.03986919e-01 -1.44743681e-01 -1.79066014e+00 5.67214727e-01 -1.04055679e+00 1.31132975e-01 -2.31026173e+00 7.24553168e-01 8.61754060e-01 9.78819001e-03 3.43630486e-03 -6.61811113e-01 -3.61749470e-01 2.54352778e-01 -3.94397043e-03 -1.11605549e+00 1.39235830e+00 1.22479868e+00 -4.20955211e-01 -2.94368118e-01 -9.24839377e-02 -9.99363601e-01 4.88756835e-01 9.28026080e-01 -3.20815712e-01 -2.85181135e-01 -8.45010519e-01 1.35700582e-02 4.99654055e-01 -1.54046804e-01 -7.07846403e-01 5.07076502e-01 2.48487428e-01 -1.20868348e-01 -1.01429009e+00 1.33789733e-01 1.61368027e-01 -5.78516185e-01 5.77977657e-01 -1.01347744e+00 2.06898853e-01 5.93193620e-02 5.84274292e-01 -5.18224180e-01 -5.34663856e-01 5.16549766e-01 -3.37309241e-01 -1.82630420e-01 8.10857937e-02 -5.63195407e-01 3.37423295e-01 5.95330954e-01 8.23503211e-02 -5.78441620e-01 -7.07014978e-01 -4.08402115e-01 6.85162306e-01 8.46742243e-02 5.14894605e-01 8.99118662e-01 -9.64914382e-01 -1.38858128e+00 -4.40673456e-02 -1.47730127e-01 1.27589598e-01 3.42993855e-01 6.02861583e-01 -6.84423566e-01 6.79715097e-01 -3.51224579e-02 -2.55926877e-01 -1.29350877e+00 2.98741728e-01 1.73264891e-01 -6.84454799e-01 -9.05243278e-01 7.66229272e-01 2.31256187e-01 -2.61321515e-01 1.97362527e-01 -3.62338960e-01 -6.40137494e-01 -4.12074476e-02 8.68037343e-01 5.86698651e-01 8.81180614e-02 -4.60355371e-01 -8.07162821e-02 1.43626079e-01 -6.25593424e-01 1.32516816e-01 1.46144831e+00 -3.54702026e-01 -6.44162178e-01 5.29572070e-01 1.20526505e+00 -2.90602267e-01 -6.88763499e-01 -8.29785466e-02 -7.57221133e-02 7.93187469e-02 -1.90615103e-01 -3.96077693e-01 -6.62359238e-01 1.00999796e+00 -5.55844963e-01 4.34306175e-01 1.21045756e+00 -1.16657227e-01 1.21071100e+00 5.47475100e-01 -1.84798107e-01 -1.04972029e+00 1.28961936e-01 1.26552427e+00 1.38277793e+00 -7.19596028e-01 1.14948153e-01 -1.93829611e-01 -9.99874473e-01 1.17919815e+00 4.91916686e-01 -6.95371628e-01 -1.16644487e-01 6.84968978e-02 -5.32840610e-01 -2.14515269e-01 -1.41441667e+00 5.72315603e-02 2.91750617e-02 1.04887486e-01 7.10630238e-01 7.43600400e-03 -9.54600751e-01 1.11968589e+00 -5.34335792e-01 6.27520382e-02 9.67673540e-01 7.36248016e-01 -8.14860582e-01 -7.98274159e-01 2.21727058e-01 7.34315038e-01 -5.80989003e-01 -4.45633501e-01 -8.14588845e-01 1.64717183e-01 -7.43247926e-01 1.05959022e+00 3.56998205e-01 -2.59813160e-01 7.37043858e-01 -7.15344027e-02 3.77561241e-01 -1.17876840e+00 -1.16948438e+00 -2.71188080e-01 5.65328419e-01 -1.87202394e-01 -5.58571592e-02 -4.22402650e-01 -1.19319987e+00 -2.64192581e-01 -2.57221937e-01 5.57555735e-01 2.50569493e-01 8.01028669e-01 7.81373620e-01 1.23237681e+00 7.60200739e-01 -8.24422598e-01 -7.92070031e-01 -1.58107901e+00 -4.64991897e-01 1.01552315e-01 5.72242260e-01 3.14955711e-01 -2.09446643e-02 1.83961857e-02]	[12.502543449401855, 9.530150413513184]
a38f4621-7f74-4227-ad75-f5095eb9ca43	neutron-induced-single-event-effects-on	2102.00112	null	https://arxiv.org/abs/2102.00112v1	https://arxiv.org/pdf/2102.00112v1.pdf	Neutron-Induced, Single-Event Effects on Neuromorphic Event-based Vision Sensor: A First Step Towards Space Applications	This paper studies the suitability of neuromorphic event-based vision cameras for spaceflight, and the effects of neutron radiation on their performance. Neuromorphic event-based vision cameras are novel sensors that implement asynchronous, clockless data acquisition, providing information about the change in illuminance greater than 120dB with sub-millisecond temporal precision. These sensors have huge potential for space applications as they provide an extremely sparse representation of visual dynamics while removing redundant information, thereby conforming to low-resource requirements. An event-based sensor was irradiated under wide-spectrum neutrons at Los Alamos Neutron Science Center and its effects were classified. We found that the sensor had very fast recovery during radiation, showing high correlation of noise event bursts with respect to source macro-pulses. No significant differences were observed between the number of events induced at different angles of incidence but significant differences were found in the spatial structure of noise events at different angles. The results show that event-based cameras are capable of functioning in a space-like, radiative environment with a signal-to-noise ratio of 3.355. They also show that radiation-induced noise does not affect event-level computation. We also introduce the Event-based Radiation-Induced Noise Simulation Environment (Event-RINSE), a simulation environment based on the noise-modelling we conducted and capable of injecting the effects of radiation-induced noise from the collected data to any stream of events in order to ensure that developed code can operate in a radiative environment. To the best of our knowledge, this is the first time such analysis of neutron-induced noise analysis has been performed on a neuromorphic vision sensor, and this study shows the advantage of using such sensors for space applications.	['Ryad Benosman', 'Bernabé Linares-Barranco', 'Alan D. George', 'Himanshu Akolkar', 'Seth Roffe']	2021-01-29	null	null	null	null	['event-based-vision']	['computer-vision']	[ 2.40106031e-01 -8.15933764e-01 7.37107754e-01 -1.86823651e-01 -8.16639513e-03 -4.02029216e-01 7.80912161e-01 1.76300243e-01 -1.11919570e+00 6.90766633e-01 4.18519266e-02 -4.72489558e-02 -1.59391806e-01 -7.62737036e-01 -6.98078930e-01 -8.04426253e-01 -1.16176829e-01 2.08730161e-01 7.77905643e-01 -9.81952772e-02 2.70383954e-01 7.81813979e-01 -2.10762000e+00 1.63101509e-01 9.35872793e-02 9.02272403e-01 5.13820112e-01 7.46001840e-01 2.21443459e-01 6.57254755e-01 -1.10643303e+00 4.46705341e-01 1.45630822e-01 -5.18063247e-01 8.98598731e-02 -6.12613916e-01 -1.04175955e-01 -2.53352553e-01 -6.78522706e-01 9.49968040e-01 7.06241071e-01 -2.81978045e-02 5.04321575e-01 -8.89498889e-01 2.52400011e-01 5.78197762e-02 1.41070575e-01 7.77832747e-01 6.25328004e-01 5.95264554e-01 -5.66910356e-02 -3.52946848e-01 8.23852003e-01 6.23775542e-01 5.44928193e-01 5.23426950e-01 -1.14401817e+00 -5.03019094e-01 -8.15103114e-01 2.00442553e-01 -9.34962094e-01 -4.16150570e-01 3.87126833e-01 -5.24640262e-01 1.52951932e+00 3.43452752e-01 1.02899778e+00 1.29585779e+00 1.04878128e+00 -4.49699849e-01 1.42849112e+00 -4.46668327e-01 1.17519844e+00 -2.87978649e-01 3.19177151e-01 1.92534551e-01 4.47769523e-01 7.37074852e-01 -8.57036591e-01 -1.49298489e-01 7.33949959e-01 -1.39531940e-01 -4.89912063e-01 2.88252413e-01 -1.22554231e+00 9.14565325e-02 1.96059316e-01 5.51939070e-01 -5.24048805e-01 6.40626132e-01 3.61357391e-01 3.14495832e-01 -1.66348994e-01 1.38901085e-01 -1.99186340e-01 -5.18750906e-01 -6.67158723e-01 1.87096313e-01 8.33916962e-01 5.85079789e-01 2.07245111e-01 3.45492035e-01 -4.67776917e-02 3.47035706e-01 2.06425130e-01 7.87132502e-01 7.81098187e-01 -5.20177007e-01 -3.95964861e-01 3.02790582e-01 -1.67845905e-01 -1.99744776e-01 -6.37319386e-01 -1.03286713e-01 -4.26064968e-01 9.32398975e-01 3.84916604e-01 -2.38211527e-01 -1.15716362e+00 1.38000607e+00 -1.46177277e-01 1.65815696e-01 3.01971138e-01 1.05228078e+00 6.85322583e-01 8.08381975e-01 -1.32576928e-01 -3.61377597e-01 1.50050509e+00 1.83329791e-01 -7.69883454e-01 -7.23593757e-02 -1.46711200e-01 -6.17821872e-01 4.26917762e-01 4.49597567e-01 -8.07766318e-01 -2.03911886e-01 -1.29796779e+00 5.50679982e-01 -2.76823968e-01 -2.95558959e-01 6.23660684e-01 9.49364841e-01 -9.86880004e-01 5.39924741e-01 -1.14794052e+00 -7.43055046e-01 -1.90828480e-02 3.53415698e-01 -1.59601048e-01 2.58488089e-01 -8.07082951e-01 9.07776773e-01 -1.13053277e-01 -2.05670998e-01 -1.25836515e+00 -7.14996874e-01 -3.44464600e-01 -1.82172120e-01 -1.71977818e-01 -5.93628645e-01 1.52544439e+00 -6.29016876e-01 -1.42832899e+00 5.16867757e-01 -1.07469358e-01 -8.11605573e-01 1.18550986e-01 4.42057014e-01 -7.23424673e-01 3.02157342e-01 -1.35106668e-01 3.25687863e-02 2.70048261e-01 -8.02840114e-01 -1.56382233e-01 -5.25216103e-01 -1.90288603e-01 -1.70611471e-01 2.97277812e-02 3.77826661e-01 7.98692480e-02 -1.86951444e-01 1.19810849e-01 -7.18105376e-01 -2.11761296e-01 -3.00078057e-02 4.02996242e-01 3.81760657e-01 6.49872482e-01 -1.39154512e-02 4.69691038e-01 -1.96490872e+00 -4.59869593e-01 3.80554586e-03 -3.65442753e-01 7.36772045e-02 3.22388798e-01 8.94537449e-01 -7.55315125e-02 -4.55468774e-01 -4.38087910e-01 2.41919413e-01 -5.27123094e-01 3.71484131e-01 -2.62393594e-01 5.92577279e-01 -2.10061625e-01 1.50299057e-01 -7.22494006e-01 3.39816153e-01 3.42780143e-01 5.74208558e-01 -3.42597924e-02 -3.88513319e-02 -1.66612431e-01 4.32850331e-01 -8.31526071e-02 6.28165305e-01 3.96896720e-01 5.23922741e-01 -1.39604986e-01 -3.66319358e-01 -8.02305698e-01 -2.57935692e-02 -1.08132064e+00 1.54381859e+00 -3.21870059e-01 9.86247957e-01 1.20932370e-01 -4.49345887e-01 1.13919389e+00 3.14284444e-01 2.86757886e-01 -1.46409976e+00 3.72850716e-01 3.34617317e-01 9.28530619e-02 -5.98793089e-01 4.29388523e-01 -6.59844756e-01 2.44094804e-01 2.20238268e-01 7.20377639e-02 -2.37005517e-01 1.34181947e-01 9.39449016e-03 1.78020227e+00 -3.78621444e-02 -1.28482252e-01 -4.37297076e-01 -3.40291820e-02 -4.76471893e-03 3.93175334e-01 6.37478292e-01 -7.41797835e-02 6.93662822e-01 9.94807258e-02 -1.93423033e-01 -1.04369533e+00 -1.10630918e+00 -4.76327598e-01 1.20956652e-01 3.38141829e-01 -2.58193135e-01 -7.64368832e-01 3.00332546e-01 -2.16381073e-01 8.99967194e-01 8.29233415e-03 -3.11408579e-01 -1.16068415e-01 -1.18397129e+00 6.87509239e-01 2.72564262e-01 2.69253522e-01 -1.17391312e+00 -1.64490700e+00 3.55659485e-01 3.47775877e-01 -1.02379143e+00 5.06843388e-01 7.89648652e-01 -1.06109035e+00 -1.14963353e+00 -2.54698813e-01 -2.70022482e-01 5.31369507e-01 1.54540632e-02 8.11151922e-01 -4.21694189e-01 -1.00366664e+00 1.06909537e+00 -2.05617145e-01 -7.27123260e-01 -2.95748562e-01 -1.09990084e+00 3.50976080e-01 -2.80437678e-01 4.91912484e-01 -8.64708185e-01 -5.54969490e-01 1.20636284e-01 -1.24574971e+00 -5.57771504e-01 2.73440391e-01 3.08818519e-01 5.54075718e-01 2.87003189e-01 1.93474621e-01 -1.85825467e-01 4.47903305e-01 -2.90840864e-01 -9.78676200e-01 -2.85504192e-01 -3.03758293e-01 7.83502087e-02 7.33715951e-01 -1.82418972e-01 -1.03050768e+00 1.32108286e-01 -1.17765134e-02 8.16728398e-02 -5.42009413e-01 2.42542952e-01 1.27760559e-01 -3.18100899e-01 1.18765593e+00 2.62903005e-01 -5.21331280e-02 -1.75588414e-01 -5.61285675e-01 6.29108489e-01 7.05912948e-01 -2.86618620e-01 3.24031323e-01 8.77930939e-01 3.46179754e-01 -1.46639156e+00 2.62673050e-01 -4.76309925e-01 1.24951765e-01 -7.47510135e-01 7.49974072e-01 -1.01123166e+00 -7.00380862e-01 1.01601994e+00 -1.11075675e+00 -2.09664449e-01 -4.67924953e-01 1.04658103e+00 -4.68353629e-01 2.01185688e-01 -4.65761393e-01 -1.22282326e+00 -1.68070838e-01 -8.92491877e-01 3.96388113e-01 7.46428907e-01 6.09922409e-03 -4.24202323e-01 5.40647745e-01 -2.48956401e-02 6.30244255e-01 3.21343452e-01 4.31526452e-01 -1.57819599e-01 -6.98587835e-01 -1.61709383e-01 2.38986880e-01 5.24889939e-02 -3.16508919e-01 5.42201214e-02 -1.23232961e+00 -2.99487263e-01 8.06871295e-01 2.76673734e-02 9.47698116e-01 5.42848766e-01 5.94808578e-01 5.03016174e-01 -2.32320964e-01 4.30172265e-01 2.13665366e+00 7.44057834e-01 1.14121461e+00 4.60188389e-01 -1.06409140e-01 7.40758255e-02 2.79877156e-01 7.16756344e-01 -4.34611022e-01 4.46316212e-01 7.68971205e-01 2.92099029e-01 -2.09563658e-01 3.16458791e-01 6.12215757e-01 3.05849046e-01 -1.63892612e-01 -2.79702008e-01 -8.45599174e-01 5.21611989e-01 -1.34526217e+00 -1.10454023e+00 -9.27938163e-01 2.69664097e+00 4.38940614e-01 2.17448384e-01 -2.17138022e-01 7.35925287e-02 6.97608769e-01 -1.52474388e-01 -4.48607504e-01 -5.40491939e-01 -2.88179904e-01 7.29713261e-01 1.03392684e+00 1.20864391e-01 -2.71005154e-01 -1.46880941e-02 6.28180647e+00 -3.63752595e-03 -1.07816613e+00 4.35544997e-02 -2.95172811e-01 -6.05726480e-01 -4.05057430e-01 4.72183436e-01 -5.29191673e-01 8.92011404e-01 1.66764832e+00 -9.60809365e-02 4.02332932e-01 3.17011684e-01 4.38380629e-01 -1.16475165e+00 -8.15003514e-01 1.10376370e+00 2.27422435e-02 -1.13867068e+00 -3.09544772e-01 7.64724314e-02 1.64461672e-01 4.05634671e-01 -6.83265448e-01 -2.85758615e-01 -1.82868585e-01 -6.50399685e-01 8.83901179e-01 1.16057360e+00 6.66847825e-01 -8.46291840e-01 6.70445263e-01 2.62906551e-01 -6.35344207e-01 -7.72794932e-02 -5.96714497e-01 -5.05999267e-01 4.14656788e-01 1.18199873e+00 -5.76246738e-01 3.16849232e-01 8.67174447e-01 -9.87071991e-02 -3.44772518e-01 1.52399957e+00 1.55439004e-01 7.07607090e-01 -7.52053022e-01 -4.16755766e-01 -3.14037710e-01 -2.03856364e-01 1.01512814e+00 1.04734254e+00 7.66623020e-01 3.23045343e-01 -8.19823325e-01 7.98159480e-01 4.40473586e-01 -6.65273070e-01 -7.56133795e-01 5.06721213e-02 2.61271685e-01 1.06925452e+00 -8.60914171e-01 2.50943452e-02 -3.60137552e-01 1.03899658e+00 -6.07011139e-01 2.64090151e-01 -7.57793665e-01 -5.80735028e-01 4.95967865e-01 3.46061081e-01 2.04942882e-01 -5.73500454e-01 -2.39028171e-01 -6.67909682e-01 1.84252396e-01 -2.56976098e-01 -4.88646738e-02 -1.24695575e+00 -1.02521122e+00 4.73472863e-01 -2.62804367e-02 -9.36442733e-01 9.70124975e-02 -7.70449817e-01 -8.41334939e-01 7.30826378e-01 -8.56629491e-01 -3.24258894e-01 -5.22887766e-01 4.81468916e-01 -4.48955223e-03 -2.37450615e-01 8.58751893e-01 4.67855223e-02 -2.07503214e-01 -1.65746555e-01 3.36939335e-01 -4.49217856e-01 5.62284708e-01 -9.75792825e-01 1.46865830e-01 1.16142094e+00 -6.44668415e-02 3.37081611e-01 1.09467757e+00 -9.66886282e-01 -1.87954783e+00 -7.16780484e-01 5.36730945e-01 -7.00563937e-02 5.39164186e-01 -4.77567762e-01 -5.88502944e-01 6.56490475e-02 2.62834102e-01 -4.75828275e-02 4.43679005e-01 -5.08922637e-01 9.99853909e-02 -3.56109589e-01 -1.42958021e+00 5.19497156e-01 9.04066205e-01 -5.82893550e-01 -6.27663910e-01 3.76654774e-01 2.30009899e-01 -1.84500918e-01 -5.54429173e-01 3.90703589e-01 3.90417367e-01 -1.53790569e+00 6.73770964e-01 2.19957113e-01 -1.60535630e-02 -6.30140781e-01 -1.32351741e-01 -1.01521742e+00 5.82988327e-03 -4.17917520e-01 3.47408026e-01 1.07368517e+00 -6.35821521e-02 -9.30387199e-01 4.37458575e-01 4.56914276e-01 -4.84773546e-01 8.65987986e-02 -1.35964644e+00 -1.01587689e+00 -6.90862536e-01 -5.68044186e-01 -2.19871420e-02 1.77013546e-01 -6.49364069e-02 -5.55776879e-02 3.02906930e-01 3.84884775e-01 7.57855594e-01 -4.41819936e-01 1.50939956e-01 -1.17811215e+00 -4.48439479e-01 1.27671182e-01 -1.04866171e+00 -1.57946497e-02 -5.50963819e-01 -6.31972313e-01 2.37123892e-01 -1.60379279e+00 1.62818909e-01 1.28253013e-01 -1.05922274e-01 1.06046021e-01 6.45482779e-01 2.04089224e-01 -2.41878465e-01 9.59452614e-02 1.23978108e-01 3.83424461e-01 3.11918736e-01 3.32149863e-01 1.28987625e-01 -2.31249228e-01 2.50337601e-01 4.97708619e-01 6.45146906e-01 -6.55727744e-01 -3.28492582e-01 -1.76303193e-01 3.89399469e-01 7.10732192e-02 1.03432882e+00 -2.05182624e+00 8.20454955e-01 2.52106965e-01 5.10152400e-01 -4.97080415e-01 4.47594136e-01 -9.80559945e-01 9.83689606e-01 7.51967311e-01 2.53096461e-01 1.01781167e-01 5.86032450e-01 8.67424905e-01 -7.68513903e-02 -7.93469608e-01 8.15072060e-01 -4.06075686e-01 -7.97407210e-01 -3.26141536e-01 -1.37325954e+00 -4.00779277e-01 1.29606521e+00 -5.36578953e-01 -6.22130454e-01 1.95800111e-01 -3.50354195e-01 -5.84899664e-01 1.00084567e+00 -1.56637967e-01 5.38107336e-01 -9.67434168e-01 -1.41068101e-01 4.88503844e-01 7.88527504e-02 -4.32510465e-01 4.88028854e-01 5.57842910e-01 -8.72225940e-01 2.35198200e-01 -7.50673890e-01 -6.30184591e-01 -1.23034751e+00 2.01548770e-01 5.65383315e-01 5.58559060e-01 -6.99392378e-01 6.46991968e-01 -4.26992506e-01 1.00373469e-01 7.00006783e-02 -5.14587581e-01 7.03197494e-02 -3.32700983e-02 6.36822164e-01 4.31424171e-01 6.70209408e-01 2.52357163e-02 -6.14798784e-01 4.82288569e-01 4.72715169e-01 -6.48754358e-01 1.39921486e+00 2.97001988e-01 -3.17322880e-01 7.97911346e-01 6.38585091e-01 -5.91222420e-02 -1.12314177e+00 3.83734375e-01 -2.16658354e-01 -6.77564442e-02 3.30730975e-01 -1.13806593e+00 -5.74308455e-01 4.29118752e-01 1.19132626e+00 3.61119248e-02 1.48568034e+00 -3.72179210e-01 4.99787986e-01 3.61219376e-01 8.56761694e-01 -1.30786395e+00 -2.67816097e-01 5.45866668e-01 5.45108378e-01 -2.77914584e-01 -1.51838427e-02 -5.19453026e-02 -4.75334190e-02 1.36570001e+00 4.20155644e-01 -2.53169298e-01 3.75277042e-01 1.03960526e+00 -1.16133094e-01 -4.58652586e-01 -8.94299328e-01 -1.01439834e-01 -5.91047883e-01 9.04061973e-01 1.00211762e-01 -7.05703497e-02 -6.67835951e-01 3.10261846e-01 1.09288216e-01 3.73915553e-01 1.15695012e+00 1.29257464e+00 -6.11349404e-01 -9.19360161e-01 -8.23275924e-01 4.02150810e-01 -1.60622135e-01 6.23457469e-02 -4.77927446e-01 3.89842898e-01 3.46133858e-02 8.50314677e-01 3.29555541e-01 -1.82748184e-01 7.71444976e-01 1.98207140e-01 8.15380871e-01 -3.69699985e-01 -8.23525190e-01 -3.07625979e-01 2.77932376e-01 -7.07388937e-01 -4.18543428e-01 -7.06369042e-01 -1.62164807e+00 -4.13475037e-02 8.15832466e-02 -7.73079172e-02 1.54901302e+00 6.56995714e-01 2.53453165e-01 9.86213326e-01 1.87281504e-01 -6.15984201e-01 -2.79642731e-01 -5.59370875e-01 -7.80292511e-01 -1.64039135e-02 4.41202559e-02 -3.83013219e-01 -7.63413548e-01 -5.51969297e-02]	[8.756023406982422, -1.2848315238952637]
acae04e4-2227-46ec-a3b4-ce239421970b	on-the-privacy-utility-trade-off-in	2103.02895	null	https://arxiv.org/abs/2103.02895v2	https://arxiv.org/pdf/2103.02895v2.pdf	On the privacy-utility trade-off in differentially private hierarchical text classification	Hierarchical text classification consists in classifying text documents into a hierarchy of classes and sub-classes. Although artificial neural networks have proved useful to perform this task, unfortunately they can leak training data information to adversaries due to training data memorization. Using differential privacy during model training can mitigate leakage attacks against trained models, enabling the models to be shared safely at the cost of reduced model accuracy. This work investigates the privacy-utility trade-off in hierarchical text classification with differential privacy guarantees, and identifies neural network architectures that offer superior trade-offs. To this end, we use a white-box membership inference attack to empirically assess the information leakage of three widely used neural network architectures. We show that large differential privacy parameters already suffice to completely mitigate membership inference attacks, thus resulting only in a moderate decrease in model utility. More specifically, for large datasets with long texts we observed Transformer-based models to achieve an overall favorable privacy-utility trade-off, while for smaller datasets with shorter texts convolutional neural networks are preferable.	['Thorsten Strufe', 'Javier Parra-Arnau', 'Francesco Aldà', 'Daniel Bernau', 'Dominik Wunderlich']	2021-03-04	null	null	null	null	['membership-inference-attack']	['computer-vision']	[ 4.65076715e-01 3.63756716e-01 6.82514831e-02 -5.80767512e-01 -9.71159875e-01 -1.12148035e+00 6.57337368e-01 4.94541436e-01 -8.40509951e-01 4.13123012e-01 -4.18117344e-02 -8.48448396e-01 9.61845964e-02 -7.76841760e-01 -6.32851958e-01 -6.77231371e-01 -1.86240867e-01 2.39148408e-01 -1.06056377e-01 1.14387840e-01 3.78538445e-02 6.38229609e-01 -1.19327044e+00 5.15675247e-01 5.98182797e-01 1.06613946e+00 -6.25692666e-01 7.31877327e-01 1.03625759e-01 7.13666499e-01 -6.74453139e-01 -1.12073457e+00 7.36812711e-01 -8.01320150e-02 -8.02104950e-01 -5.45306265e-01 6.00940168e-01 -6.13239944e-01 -6.74057007e-01 1.08027411e+00 4.27003354e-01 -5.69462664e-02 5.60463071e-01 -1.50513494e+00 -3.43653023e-01 8.55360091e-01 -2.25374058e-01 -1.34402916e-01 -6.62613586e-02 6.57879934e-02 1.26205564e+00 -1.02328457e-01 1.52333066e-01 1.04889071e+00 7.65354931e-01 7.00735092e-01 -1.75238657e+00 -1.12447870e+00 -1.32916749e-01 -1.71793371e-01 -1.30400753e+00 -5.32293499e-01 5.33300579e-01 -1.55336544e-01 7.13991165e-01 8.14381242e-01 2.33906552e-01 1.11620653e+00 3.14582109e-01 7.14490652e-01 1.09633851e+00 -1.14388146e-01 2.74875075e-01 6.85086906e-01 2.99092472e-01 5.54970860e-01 6.06291831e-01 -2.60971580e-03 -3.55417609e-01 -7.73624599e-01 6.82968050e-02 4.86171432e-02 -3.40649277e-01 -4.60973412e-01 -4.84864533e-01 8.78940880e-01 3.78706932e-01 1.54832944e-01 2.17789590e-01 1.51428744e-01 8.22887301e-01 8.24122787e-01 4.47805405e-01 6.02887273e-01 -7.40958750e-01 3.56399894e-01 -7.78325498e-01 4.58221793e-01 1.10620248e+00 7.99247921e-01 4.76188868e-01 -1.90361261e-01 -2.50666067e-02 4.54889059e-01 -1.00930408e-01 8.30226094e-02 5.33118427e-01 -6.51713669e-01 7.15879261e-01 3.54485452e-01 -7.47232288e-02 -1.08195233e+00 -1.98563457e-01 -4.55541104e-01 -1.00970721e+00 3.42849940e-01 7.89125919e-01 -3.51919770e-01 -4.89077687e-01 2.02734113e+00 8.38082284e-02 -5.09745479e-01 3.82876545e-01 3.35747659e-01 4.37890971e-03 4.26568240e-01 2.09495828e-01 6.68844208e-02 1.55801582e+00 -3.53102177e-01 -3.85721594e-01 -1.81722656e-01 1.19619024e+00 -4.26218629e-01 1.19312310e+00 2.30536744e-01 -1.04115486e+00 1.34452924e-01 -1.25933695e+00 -2.88081259e-01 -6.30687296e-01 -2.77800590e-01 5.33327103e-01 1.40670252e+00 -8.76765609e-01 7.21268535e-01 -8.30824375e-01 7.33292550e-02 9.85935092e-01 6.77845359e-01 -6.12953067e-01 1.62898436e-01 -1.32687402e+00 6.46640897e-01 3.69938850e-01 -1.87959582e-01 -6.41467035e-01 -9.23295259e-01 -7.17279971e-01 5.52636027e-01 -7.56923333e-02 -4.91388470e-01 1.20507216e+00 -8.33494186e-01 -1.12573612e+00 9.38119352e-01 2.14401186e-01 -1.26731551e+00 9.38338816e-01 1.04849748e-01 4.39530909e-02 -1.42613441e-01 -4.33207840e-01 6.03052080e-01 6.24850214e-01 -1.11910093e+00 -5.97462416e-01 -7.84299076e-01 2.92855829e-01 9.68857482e-02 -1.19509506e+00 -5.40086776e-02 2.56673813e-01 -6.63695693e-01 -1.66237950e-01 -1.09670329e+00 -2.12517083e-01 2.31437162e-01 -4.21143740e-01 9.50295627e-02 1.06217992e+00 -7.18809247e-01 9.57655668e-01 -2.07663226e+00 -4.21038568e-01 3.09576094e-01 3.80105734e-01 1.71326980e-01 1.98332742e-01 1.17069006e-01 1.45135790e-01 4.67373133e-01 -2.74204046e-01 -8.09872508e-01 1.96200594e-01 1.04567468e-01 -5.02226651e-01 6.80526435e-01 -2.18403175e-01 7.43047655e-01 -1.71838582e-01 -1.62582949e-01 -1.54536963e-01 5.10852039e-01 -7.60088861e-01 -8.31894949e-03 -2.40432367e-01 -6.02425225e-02 -2.21818641e-01 1.53181732e-01 8.69052052e-01 1.70037821e-02 3.57348561e-01 2.59429123e-02 5.12843966e-01 3.93985569e-01 -7.24730849e-01 1.15099299e+00 -5.47547579e-01 1.09553099e+00 3.19741160e-01 -7.82985806e-01 5.97772598e-01 4.17143315e-01 1.17184423e-01 -4.23920333e-01 4.48764563e-01 1.00230589e-03 5.68891205e-02 -2.42693373e-03 6.02885544e-01 -1.77851707e-01 -2.29844868e-01 7.60045946e-01 -2.40599319e-01 1.35157496e-01 -5.67474604e-01 2.26142734e-01 1.05939496e+00 -5.99054813e-01 -2.03641847e-01 -5.77951133e-01 3.51882875e-01 -3.34381759e-01 2.93455899e-01 9.43425477e-01 -2.20930532e-01 3.30398858e-01 7.51699090e-01 -4.88776535e-01 -1.03995931e+00 -5.97509682e-01 -1.96575776e-01 1.14874113e+00 -9.01440680e-02 -4.19213712e-01 -1.11701000e+00 -1.02039647e+00 1.46021172e-01 8.56879652e-01 -7.59756863e-01 -5.20135939e-01 -3.03134739e-01 -7.32753932e-01 1.38416386e+00 3.67451131e-01 6.02607429e-01 -3.80647391e-01 -7.05521941e-01 -3.18033814e-01 2.37247441e-02 -9.92457688e-01 -6.10720992e-01 5.67248166e-01 -8.67485404e-01 -8.62053037e-01 -4.71818805e-01 -5.03327966e-01 8.30272317e-01 1.31589072e-02 6.50873244e-01 -1.45487115e-02 -1.60780817e-01 1.31144866e-01 2.03765407e-02 -6.27104998e-01 -7.03649759e-01 4.21378821e-01 -1.36508048e-01 -2.22468078e-02 3.55857700e-01 -6.87406898e-01 -4.94759947e-01 2.88481623e-01 -1.30503678e+00 -2.02607423e-01 2.09552035e-01 8.82620931e-01 -3.17229591e-02 4.46870148e-01 5.33263236e-02 -1.28784943e+00 7.61297762e-01 -2.18257815e-01 -7.71085560e-01 2.55364090e-01 -8.90850544e-01 3.45576912e-01 1.11682618e+00 -6.32718205e-01 -9.91695285e-01 8.22386742e-02 -1.98244795e-01 -2.08938465e-01 -1.03871092e-01 9.20456797e-02 -3.95273209e-01 -4.37728584e-01 8.73120427e-01 2.13352934e-01 2.76967913e-01 -3.82962435e-01 2.63677001e-01 8.49313676e-01 1.97564766e-01 -5.65387130e-01 7.07976103e-01 5.74324608e-01 1.66367695e-01 -5.47917306e-01 -5.39272487e-01 5.54239228e-02 -4.00641978e-01 4.48751509e-01 5.67907631e-01 -7.10846007e-01 -1.03051758e+00 5.59521496e-01 -7.95759499e-01 -2.68866330e-01 -2.20186859e-01 5.65129295e-02 -3.36498946e-01 4.78298157e-01 -8.77395868e-01 -7.89471090e-01 -6.55840635e-01 -1.13863981e+00 6.14558101e-01 -2.28625476e-01 -2.89629668e-01 -1.08279037e+00 -4.96587604e-01 5.13753951e-01 5.16731203e-01 1.46437004e-01 1.08991110e+00 -1.27875113e+00 -2.08929017e-01 -6.74807191e-01 -4.71222447e-03 4.81379002e-01 -8.30442682e-02 -4.52824354e-01 -1.43375397e+00 -7.68670082e-01 2.73254603e-01 -4.24466729e-01 8.24039459e-01 -9.03709754e-02 1.51955342e+00 -1.04067969e+00 -1.71391070e-01 8.14674497e-01 1.18733001e+00 8.92363861e-02 7.24861801e-01 2.89631903e-01 6.22641385e-01 9.43260252e-01 -3.28131882e-03 3.08327764e-01 1.28168777e-01 4.76894915e-01 3.13413441e-01 6.74194545e-02 6.04513228e-01 -4.22198892e-01 2.35412732e-01 -4.27436829e-02 8.19204807e-01 -3.54753673e-01 -7.16849267e-01 2.09189102e-01 -1.39498174e+00 -6.96860433e-01 1.41255081e-01 2.37724233e+00 9.85117257e-01 4.63312596e-01 1.49287924e-01 3.50197583e-01 3.37519139e-01 4.57780994e-02 -4.29595202e-01 -5.99748552e-01 -1.30988270e-01 1.13511737e-02 1.17136824e+00 5.10039747e-01 -1.14569390e+00 7.08067954e-01 5.78162146e+00 9.13275003e-01 -1.07873952e+00 7.16421753e-02 1.22685909e+00 -3.24325740e-01 -4.30059671e-01 5.53680621e-02 -6.66797161e-01 4.46262479e-01 1.32475233e+00 -4.81779069e-01 1.78817242e-01 8.36069822e-01 -3.92334938e-01 3.75641137e-01 -1.25821447e+00 7.51168311e-01 -3.23556751e-01 -1.20738292e+00 1.24219194e-01 4.42813754e-01 3.43413055e-01 -1.92793146e-01 4.66728568e-01 1.49842739e-01 5.02918124e-01 -1.04495013e+00 5.89670241e-01 -1.34883597e-01 7.42564857e-01 -1.18987286e+00 6.67588234e-01 4.33720738e-01 -7.00451076e-01 -3.33210796e-01 -5.11150360e-01 9.72770974e-02 -3.74943167e-01 3.09175342e-01 -7.73578823e-01 2.63278782e-01 5.82385063e-01 -2.95484457e-02 -4.67312932e-01 4.90287602e-01 2.60124832e-01 5.01525581e-01 -5.75416744e-01 -1.77885070e-02 2.91316777e-01 1.96784679e-02 2.91672230e-01 1.20090127e+00 4.88819592e-02 -3.44305150e-02 -2.66901851e-01 7.67185867e-01 -6.33502722e-01 3.32523547e-02 -8.23187768e-01 -7.39442790e-03 6.04108512e-01 1.09450352e+00 -4.28820908e-01 2.33893190e-02 -4.49855775e-02 1.16965151e+00 2.53377974e-01 1.30021974e-01 -4.47636276e-01 -5.74958146e-01 9.83913660e-01 8.93885419e-02 -9.34222490e-02 -7.00517446e-02 -7.54143417e-01 -1.06505024e+00 9.34237018e-02 -9.31652546e-01 6.72239661e-01 -2.60827672e-02 -1.10869682e+00 3.88785124e-01 -8.15913901e-02 -8.19501102e-01 -1.08645134e-01 -4.76646096e-01 -3.42640579e-01 8.57071579e-01 -9.90232170e-01 -1.25730705e+00 2.81032532e-01 4.23561722e-01 3.02993739e-03 -2.19424799e-01 1.17360663e+00 1.99450299e-01 -4.75044429e-01 1.75337291e+00 4.49374914e-01 5.09922981e-01 3.66128832e-01 -1.11154795e+00 7.16274500e-01 7.45456815e-01 2.32916370e-01 8.66746128e-01 7.82089472e-01 -2.91901588e-01 -1.36011815e+00 -1.26079166e+00 7.78491855e-01 -6.53641403e-01 4.90862876e-01 -9.08632100e-01 -1.02622426e+00 8.28085005e-01 -1.14649031e-02 -9.94912088e-02 9.57166553e-01 -4.52171415e-02 -1.00433993e+00 -2.23745957e-01 -1.81369853e+00 9.21183228e-01 6.20549023e-01 -9.35057700e-01 -1.07942484e-02 1.14357255e-01 9.36326265e-01 -2.51253724e-01 -1.00564194e+00 9.82493535e-02 8.97873819e-01 -8.22927952e-01 8.28852594e-01 -6.53438985e-01 7.46901408e-02 2.58644521e-01 -3.10531139e-01 -9.42833960e-01 1.86252385e-01 -8.76247942e-01 -4.87639531e-02 1.28450239e+00 7.43952453e-01 -1.09767163e+00 1.47729194e+00 1.45631850e+00 6.65187359e-01 -6.59470797e-01 -1.10087442e+00 -6.52459145e-01 7.48449624e-01 -3.07257652e-01 6.96845055e-01 1.17177761e+00 8.00025165e-02 2.95391306e-02 -3.71612221e-01 1.79012924e-01 8.70110333e-01 -3.34502578e-01 6.19212329e-01 -1.07479620e+00 -2.97199875e-01 -4.91952807e-01 -4.74246323e-01 -8.97066116e-01 5.70555806e-01 -9.42776740e-01 -2.87494898e-01 -5.27355909e-01 5.56730032e-02 -7.21047699e-01 -2.08034381e-01 6.69900000e-01 1.47221657e-02 1.82271108e-01 2.57203698e-01 1.82525456e-01 1.55282551e-02 2.58484542e-01 4.90325958e-01 -4.17010278e-01 -1.68710053e-01 4.24149930e-01 -1.11909878e+00 2.98144639e-01 1.20348668e+00 -7.58177817e-01 -7.22528994e-01 -4.33671057e-01 1.39754742e-01 -1.22277588e-01 3.35833132e-01 -8.54821026e-01 3.54447812e-01 3.36428314e-01 3.23871195e-01 -1.91692486e-01 2.87492663e-01 -1.36290288e+00 1.58070385e-01 7.75148213e-01 -1.15089428e+00 -1.06326424e-01 4.94008362e-01 6.71238899e-01 1.63661420e-01 -1.31741181e-01 9.75539327e-01 1.35518923e-01 3.36989313e-01 3.00814658e-01 -5.22836149e-01 -1.65651113e-01 1.02308226e+00 -1.75811470e-01 -3.40168864e-01 -5.52761495e-01 -5.17572701e-01 2.11911663e-01 7.02880144e-01 2.42365867e-01 2.26752907e-01 -8.07142258e-01 -4.57085520e-01 4.92596835e-01 3.93431354e-03 -2.44216979e-01 1.49688587e-01 2.59310186e-01 -4.71930802e-01 5.13302028e-01 -1.27613455e-01 -2.24490985e-01 -1.84882927e+00 4.14929152e-01 5.94713807e-01 -4.06030804e-01 -5.39180875e-01 9.88334835e-01 3.18289936e-01 -5.37522376e-01 7.17370629e-01 -2.21052408e-01 2.63095587e-01 -9.51718241e-02 5.51653445e-01 2.48300508e-01 4.15576518e-01 -3.49152565e-01 -7.69339278e-02 -2.22185269e-01 -6.06964767e-01 -2.27373481e-01 8.38724375e-01 2.43962761e-02 1.08584594e-02 -1.55731350e-01 1.87268901e+00 -9.34471935e-02 -1.14506400e+00 -2.51797348e-01 -4.52062711e-02 -6.12640381e-01 1.97460607e-01 -6.05401576e-01 -1.02762914e+00 1.04190516e+00 7.18517423e-01 5.64972460e-01 1.07955623e+00 -4.81372505e-01 1.08514547e+00 6.87028110e-01 3.65675628e-01 -7.10234642e-01 -5.51923633e-01 2.24747971e-01 2.55209357e-01 -8.66008759e-01 -7.41995359e-03 -1.88699529e-01 -5.44197738e-01 8.62394750e-01 3.46624464e-01 2.32186809e-01 7.04520047e-01 5.85305989e-01 -8.58708918e-02 1.74096137e-01 -7.89331257e-01 6.44199073e-01 -5.17298728e-02 5.11118948e-01 1.23694286e-01 3.36401872e-02 1.11053698e-01 7.21361101e-01 -5.22309482e-01 -5.78991413e-01 4.19347256e-01 1.03034008e+00 -3.11089680e-02 -1.29413104e+00 -2.44988546e-01 7.35171854e-01 -1.21679378e+00 -1.11494690e-01 -7.07067788e-01 5.76127768e-01 -3.80134940e-01 7.79766977e-01 3.43815200e-02 -6.78974152e-01 1.87714621e-01 1.97276190e-01 2.81597406e-01 -9.89437774e-02 -1.17378914e+00 -5.15994430e-01 2.06624076e-01 -2.60798007e-01 3.02023351e-01 -5.31062424e-01 -8.73358905e-01 -9.55105543e-01 -5.17929137e-01 2.57371306e-01 6.59170687e-01 5.73493421e-01 4.86521661e-01 -5.57260867e-03 7.55187631e-01 -4.92410958e-01 -1.21447659e+00 -5.13691068e-01 -6.12616122e-01 4.86210823e-01 4.74598289e-01 1.44535854e-01 -6.06984973e-01 -1.52998239e-01]	[5.94254732131958, 6.951089859008789]
f69aca99-9984-4e64-85e0-2c7cc875fe82	dfnet-enhance-aboslute-pose-regression-with	2204.00559	null	https://arxiv.org/abs/2204.00559v4	https://arxiv.org/pdf/2204.00559v4.pdf	DFNet: Enhance Absolute Pose Regression with Direct Feature Matching	We introduce a camera relocalization pipeline that combines absolute pose regression (APR) and direct feature matching. By incorporating exposure-adaptive novel view synthesis, our method successfully addresses photometric distortions in outdoor environments that existing photometric-based methods fail to handle. With domain-invariant feature matching, our solution improves pose regression accuracy using semi-supervised learning on unlabeled data. In particular, the pipeline consists of two components: Novel View Synthesizer and DFNet. The former synthesizes novel views compensating for changes in exposure and the latter regresses camera poses and extracts robust features that close the domain gap between real images and synthetic ones. Furthermore, we introduce an online synthetic data generation scheme. We show that these approaches effectively enhance camera pose estimation both in indoor and outdoor scenes. Hence, our method achieves a state-of-the-art accuracy by outperforming existing single-image APR methods by as much as 56%, comparable to 3D structure-based methods.	['Victor Adrian Prisacariu', 'ZiRui Wang', 'Xinghui Li', 'Shuai Chen']	2022-04-01	null	null	null	null	['camera-relocalization']	['computer-vision']	[ 4.55697805e-01 -4.68446910e-02 1.37939200e-01 -3.83611888e-01 -1.20569956e+00 -1.15562892e+00 7.05894172e-01 -4.86906111e-01 -2.74824709e-01 5.36938906e-01 1.92844763e-01 1.33672714e-01 2.77322680e-01 -5.80223620e-01 -1.05734336e+00 -4.24591452e-01 6.99350953e-01 4.33251113e-01 4.18334812e-01 -1.13141052e-01 2.98014730e-01 7.14076161e-01 -1.63761723e+00 -1.25313327e-01 6.95004344e-01 9.46052730e-01 1.50714070e-02 7.16229081e-01 4.92418706e-01 4.61959034e-01 -3.00665975e-01 -4.04806674e-01 8.14321578e-01 -4.88388464e-02 -2.98364937e-01 5.78612328e-01 1.32828665e+00 -7.27338850e-01 -5.17837286e-01 8.35660219e-01 5.23088992e-01 2.30669919e-02 4.15855139e-01 -9.49582458e-01 -2.46481702e-01 -2.68482178e-01 -6.60536528e-01 -3.05121303e-01 1.01652265e+00 3.66022140e-01 6.39816761e-01 -1.22633958e+00 8.62436771e-01 1.04515636e+00 9.91650939e-01 3.28820080e-01 -1.51597822e+00 -4.14246917e-01 1.05300508e-01 -1.39520541e-01 -1.44346917e+00 -6.31252289e-01 1.01028967e+00 -5.97675920e-01 1.01615965e+00 2.18120277e-01 6.79260731e-01 1.25093126e+00 6.03823848e-02 3.40273172e-01 1.21808779e+00 -4.52871263e-01 1.33138329e-01 -1.92671008e-02 -5.65433979e-01 6.82245970e-01 1.73999891e-01 3.55997205e-01 -7.20240355e-01 1.57748014e-01 9.46072280e-01 -3.41322869e-02 -3.50835443e-01 -1.14164793e+00 -1.48846340e+00 3.56815904e-01 2.78881878e-01 -3.91909808e-01 -7.24013150e-02 9.74648818e-02 -1.15203403e-01 1.73082441e-01 5.07592618e-01 7.14139640e-01 -5.92383444e-01 4.12744284e-02 -1.03872478e+00 3.81180048e-01 5.47682106e-01 1.35339642e+00 1.08156264e+00 7.05974624e-02 2.78032333e-01 6.54485703e-01 1.67956620e-01 1.05578065e+00 8.20017532e-02 -1.63058615e+00 5.17964840e-01 5.64240277e-01 2.08475322e-01 -9.02052701e-01 -2.43831590e-01 -2.17388645e-01 -2.71332353e-01 3.42407256e-01 4.45089549e-01 6.74403389e-04 -7.34828651e-01 1.60930800e+00 5.06491363e-01 -4.29918356e-02 -7.94447139e-02 7.86546409e-01 6.29061997e-01 2.14036912e-01 -6.88438535e-01 -1.88109726e-01 1.05185950e+00 -9.88681436e-01 -3.55539799e-01 -4.77879852e-01 1.26035362e-01 -1.23729801e+00 9.69224453e-01 4.13280547e-01 -1.18205416e+00 -6.01591408e-01 -1.27256787e+00 -2.72213608e-01 8.74183315e-04 2.41314590e-01 2.70182699e-01 7.71747649e-01 -1.05112541e+00 4.35131013e-01 -6.91895306e-01 -5.62539041e-01 4.82098982e-02 4.58930969e-01 -6.01450205e-01 -2.23083898e-01 -5.08702874e-01 7.19512880e-01 -1.80777498e-02 -2.62274265e-01 -8.95708025e-01 -9.85512316e-01 -1.24025846e+00 -3.42212647e-01 7.04484403e-01 -1.09740138e+00 1.27714550e+00 -8.68393183e-01 -1.72019351e+00 1.18188536e+00 -2.19948292e-01 -1.97593138e-01 8.12016487e-01 -4.86865669e-01 -1.53501496e-01 2.37609074e-01 1.52255058e-01 8.08855891e-01 9.87999141e-01 -1.51583672e+00 -4.41803128e-01 -4.97851074e-01 1.22375071e-01 5.38525581e-01 -1.16717562e-01 -4.17934805e-01 -8.49650741e-01 -6.10230625e-01 3.62456411e-01 -1.18022597e+00 -1.59245446e-01 4.39001858e-01 -2.88258553e-01 5.73914349e-01 9.02230203e-01 -5.38885951e-01 6.71516538e-01 -1.94681156e+00 2.33744562e-01 1.45532206e-01 1.90181389e-01 4.94014844e-02 -5.13930842e-02 4.79411066e-01 -4.49378937e-02 -5.07681727e-01 -1.03805028e-01 -6.66820943e-01 -1.46534190e-01 1.00703621e-02 -3.26058239e-01 6.30689800e-01 1.57117531e-01 6.60377204e-01 -7.33005643e-01 -2.28688970e-01 9.09230530e-01 4.30315554e-01 -7.55342543e-01 3.81338656e-01 -2.33326569e-01 7.09819078e-01 2.88333576e-02 8.04154158e-01 1.06639087e+00 -1.11123566e-02 4.13188994e-01 -4.66936946e-01 -3.50311518e-01 9.47472155e-02 -1.41500282e+00 2.28800869e+00 -5.99488318e-01 6.41480625e-01 -7.69657269e-02 -2.94562548e-01 1.00274348e+00 -1.38999894e-01 6.06030285e-01 -5.73191345e-01 -2.96274051e-02 2.85126209e-01 -8.07640433e-01 -1.89096972e-01 7.02353656e-01 1.74119383e-01 -7.93255493e-02 9.21958089e-02 1.55728832e-01 -8.64181757e-01 -4.26025502e-02 6.25136271e-02 1.03808999e+00 9.47872341e-01 3.83113414e-01 -7.75959343e-02 7.90660560e-01 -1.33745849e-01 6.37202084e-01 3.18162918e-01 7.07897469e-02 1.17695594e+00 2.08465293e-01 -4.95698243e-01 -1.45068681e+00 -1.41972423e+00 1.46670699e-01 5.26303828e-01 4.82334524e-01 -4.44823354e-01 -9.48945820e-01 -6.09955907e-01 1.51831105e-01 3.74164492e-01 -4.60928172e-01 -6.55098855e-02 -6.16114080e-01 -1.53622761e-01 1.39160812e-01 6.01554811e-01 6.41453505e-01 -2.36604422e-01 -4.43050057e-01 -6.91756904e-02 -1.15471438e-01 -1.56546938e+00 -7.32775748e-01 -1.24980755e-01 -8.03554296e-01 -1.26283836e+00 -4.81366992e-01 -5.89638174e-01 6.86748743e-01 7.48503685e-01 1.23577631e+00 -3.17316800e-01 -2.02179074e-01 7.87442386e-01 -7.39861801e-02 -6.42056065e-03 -2.12373570e-01 3.39206830e-02 2.73006856e-01 -1.01949263e-03 5.29458076e-02 -7.22943187e-01 -9.24430728e-01 6.70946360e-01 -6.34134293e-01 2.13595301e-01 4.69349116e-01 6.30634427e-01 7.51094520e-01 -7.05882967e-01 -2.59208471e-01 -6.23731613e-01 -1.43848062e-01 2.17069760e-01 -1.25722897e+00 8.44751894e-02 -7.88220763e-01 1.00429423e-01 6.51664138e-01 -2.38214090e-01 -1.29224586e+00 7.90068388e-01 6.77532777e-02 -6.43421173e-01 -2.03705549e-01 -3.72550040e-01 -3.92056704e-01 -5.99350989e-01 8.90571713e-01 9.70024467e-02 6.00359477e-02 -2.57873416e-01 4.26170081e-01 2.74863958e-01 9.55918729e-01 -5.47542930e-01 1.45377958e+00 9.09685671e-01 2.71439403e-01 -8.05205643e-01 -8.26503277e-01 -5.43099821e-01 -1.14791334e+00 -4.27933484e-01 8.48199069e-01 -1.35145307e+00 -6.97110116e-01 5.28792322e-01 -9.95593846e-01 -2.14532718e-01 -3.70825499e-01 6.27553463e-01 -9.30124402e-01 5.22062242e-01 -2.84624100e-01 -3.94135147e-01 9.95130390e-02 -1.35762966e+00 1.80146706e+00 1.47228062e-01 -8.66995826e-02 -8.26124609e-01 2.81862736e-01 7.13028491e-01 1.14000589e-02 5.30460238e-01 1.50272086e-01 4.90834527e-02 -1.14797068e+00 -1.12220034e-01 -9.04497057e-02 3.16691965e-01 -5.60811199e-02 1.18939556e-01 -1.31033409e+00 -5.05232155e-01 -1.81715026e-01 -4.00668442e-01 6.39721274e-01 2.78268278e-01 7.42534101e-01 1.43726915e-01 -1.49162009e-01 1.19653189e+00 1.65152621e+00 -6.30855039e-02 7.30824113e-01 4.91834253e-01 8.79137993e-01 4.32755768e-01 7.49785602e-01 3.36705416e-01 4.96584982e-01 1.01705384e+00 5.81083238e-01 -2.94180095e-01 -3.82127047e-01 -4.93495017e-01 4.19356972e-01 6.14659846e-01 -1.97355166e-01 1.31572470e-01 -6.75319672e-01 2.94196635e-01 -1.64263344e+00 -8.36272538e-01 -2.09216580e-01 2.52327800e+00 5.15187860e-01 1.02766313e-01 -2.50790156e-02 -9.72289741e-02 3.96966815e-01 3.16997200e-01 -6.74540639e-01 -7.47961504e-03 -2.48781636e-01 9.87373739e-02 9.55167949e-01 7.02363551e-01 -1.15922070e+00 9.96978223e-01 6.33516645e+00 3.34908009e-01 -8.43972147e-01 -1.53842136e-01 2.74416506e-01 -3.17906328e-02 -4.04609650e-01 1.65915340e-01 -7.84256399e-01 3.43797617e-02 4.51320678e-01 1.92432091e-01 6.92801833e-01 1.06343055e+00 1.52408853e-01 -1.71171829e-01 -1.24280846e+00 1.30511928e+00 5.78228831e-01 -1.36963856e+00 -1.85359612e-01 1.76814422e-01 1.12271523e+00 1.07067171e-02 8.53453502e-02 -2.20112503e-01 1.63337544e-01 -4.37016070e-01 9.10080910e-01 6.99091613e-01 1.06197727e+00 -6.86577380e-01 3.77012193e-01 1.08745024e-01 -1.28633416e+00 -3.53301838e-02 -1.68747321e-01 8.20110068e-02 2.10234687e-01 4.54576880e-01 -7.77093887e-01 8.16017091e-01 6.28758788e-01 8.88060153e-01 -9.11541343e-01 8.03281724e-01 -4.96227831e-01 -5.50282672e-02 -2.61855870e-01 4.64461803e-01 -1.89422131e-01 -3.60357523e-01 6.18392050e-01 8.56431663e-01 3.67783844e-01 -2.11262956e-01 2.05900565e-01 6.58020675e-01 -2.29320601e-01 -2.96017855e-01 -8.69599819e-01 6.31597340e-01 4.27371562e-01 1.35157573e+00 -4.02833045e-01 -7.17989057e-02 -5.46273410e-01 1.28223956e+00 1.57341912e-01 2.74778366e-01 -8.00089955e-01 -2.02371612e-01 8.09772372e-01 3.02354723e-01 4.06739652e-01 -5.14074624e-01 -1.90498278e-01 -1.52114630e+00 3.05336952e-01 -8.68012965e-01 -9.80289951e-02 -1.16645706e+00 -9.48960066e-01 2.95657158e-01 4.81563434e-02 -1.72358727e+00 -5.67634165e-01 -7.32096076e-01 -1.33453354e-01 5.57939351e-01 -1.45697653e+00 -1.56602216e+00 -8.07641149e-01 5.35831749e-01 8.74657214e-01 -1.01811208e-01 7.58237898e-01 2.00812310e-01 -1.41482726e-01 5.18013537e-01 3.38667452e-01 -1.79509297e-01 1.18549919e+00 -1.32191837e+00 7.00446725e-01 1.12480152e+00 9.03337821e-02 5.00266135e-01 5.75159013e-01 -3.96231383e-01 -1.74767399e+00 -1.02974629e+00 4.41258729e-01 -9.26666558e-01 3.44603986e-01 -7.59250164e-01 -2.60466278e-01 7.09520400e-01 8.31708759e-02 1.19263493e-01 3.79771531e-01 -2.82878578e-01 -6.92874432e-01 -5.66472411e-01 -1.00655901e+00 6.39668047e-01 1.36752212e+00 -6.02785468e-01 -4.59178507e-01 3.82659703e-01 7.19345152e-01 -8.85277629e-01 -8.07718992e-01 3.72845769e-01 8.08674395e-01 -1.40131116e+00 1.36114252e+00 1.92604549e-02 3.66103560e-01 -6.54893041e-01 -5.17462134e-01 -1.21519649e+00 -2.48447418e-01 -8.95189047e-01 2.61710789e-02 1.12628293e+00 2.39360750e-01 -5.15541971e-01 8.36989224e-01 3.77048463e-01 -2.57740468e-01 -2.75314301e-01 -5.81557214e-01 -8.95495296e-01 -3.77044111e-01 -2.92083681e-01 5.20350397e-01 7.54737258e-01 -7.03371167e-01 2.96792984e-01 -4.41341966e-01 2.56776750e-01 1.00288522e+00 1.96776614e-01 1.51395106e+00 -1.06563556e+00 -3.89955282e-01 4.74334396e-02 -5.21298170e-01 -1.26938534e+00 1.79179579e-01 -3.70416284e-01 -6.17587194e-02 -1.19807589e+00 1.21315539e-01 6.02387302e-02 4.57533419e-01 3.39814872e-02 2.42557051e-03 6.93688571e-01 4.40384559e-02 1.52086839e-01 -5.12581050e-01 3.04423809e-01 1.08535242e+00 6.79006055e-02 -1.63057670e-01 -1.15811452e-01 -4.07143772e-01 9.65541244e-01 3.52082640e-01 -1.28217742e-01 -4.86561507e-01 -5.48526287e-01 2.96519101e-01 -9.61459726e-02 4.60657358e-01 -1.31180727e+00 1.59331292e-01 -1.03881024e-01 8.52085233e-01 -8.08374524e-01 7.29142487e-01 -8.68124783e-01 2.74925053e-01 2.68116891e-01 3.83281596e-02 1.99915946e-01 1.80295482e-02 6.77072406e-01 6.36306554e-02 1.96258709e-01 8.62485290e-01 -1.78903386e-01 -6.95216179e-01 3.38912338e-01 -8.53345990e-02 1.20494794e-02 1.02869415e+00 -5.59073746e-01 -2.38728955e-01 -5.38382113e-01 -3.82753879e-01 -6.08346751e-03 1.27544236e+00 3.24133396e-01 6.63726330e-01 -1.48551667e+00 -3.78667712e-01 4.96233523e-01 4.85104293e-01 1.74784452e-01 1.74291939e-01 7.16947854e-01 -9.36248004e-01 2.33495727e-01 -3.23005132e-02 -9.86350954e-01 -1.43276691e+00 5.69981635e-01 3.39155287e-01 -2.22098399e-02 -2.85583913e-01 6.03644729e-01 2.12534532e-01 -8.28665376e-01 5.36704101e-02 -2.06443742e-01 4.18334335e-01 -1.65925503e-01 2.41774961e-01 4.57206100e-01 2.76027411e-01 -7.53915966e-01 -3.50655675e-01 1.14465225e+00 1.20182827e-01 -3.19796085e-01 1.43277061e+00 -3.91114354e-01 2.13479519e-01 2.05513895e-01 1.28567052e+00 4.59087223e-01 -2.02128458e+00 -2.30719179e-01 -4.63101000e-01 -9.93146539e-01 -1.05751671e-01 -5.69755614e-01 -8.09255719e-01 5.81559598e-01 7.44448900e-01 -4.24048692e-01 1.19693244e+00 -8.92591849e-02 6.16692662e-01 5.01783907e-01 4.44052041e-01 -1.25055623e+00 3.61377388e-01 4.38757062e-01 7.63167918e-01 -1.36403847e+00 4.19805497e-01 -7.13874161e-01 -3.09458941e-01 1.24973035e+00 7.46316254e-01 -3.34355682e-01 1.80044144e-01 3.46351773e-01 1.35953426e-01 1.74392499e-02 -3.52101535e-01 -8.44882131e-02 5.05498528e-01 9.05880451e-01 9.18451548e-02 -3.06678206e-01 1.59671545e-01 -2.51385689e-01 -3.26875925e-01 -3.80720317e-01 4.77844119e-01 8.55976820e-01 -1.98121145e-01 -1.28668714e+00 -7.16100752e-01 -1.15418389e-01 1.26488358e-01 3.01924758e-02 -5.91576993e-01 9.47294414e-01 -3.12240366e-02 7.45790958e-01 2.74888817e-02 -5.23522556e-01 7.62203217e-01 -2.86854059e-01 9.86468077e-01 -4.91195440e-01 -3.76941472e-01 2.94004381e-01 8.50645676e-02 -1.09164345e+00 -5.74432075e-01 -9.71119165e-01 -5.81788480e-01 -1.73515990e-01 -3.18127334e-01 -6.19281650e-01 6.69748545e-01 6.98223233e-01 3.77770394e-01 2.47630820e-01 1.21933901e+00 -1.36934054e+00 -4.03464824e-01 -5.21439612e-01 -2.19649717e-01 4.69943583e-01 4.11162615e-01 -8.06830525e-01 -4.05120462e-01 3.66818219e-01]	[8.282886505126953, -2.4643445014953613]
71f94576-29f7-4f3a-b3c1-4ee0c6b6742a	biometric-face-presentation-attack-detection-1	1909.08848	null	https://arxiv.org/abs/1909.08848v1	https://arxiv.org/pdf/1909.08848v1.pdf	Biometric Face Presentation Attack Detection with Multi-Channel Convolutional Neural Network	Face recognition is a mainstream biometric authentication method. However, vulnerability to presentation attacks (a.k.a spoofing) limits its usability in unsupervised applications. Even though there are many methods available for tackling presentation attacks (PA), most of them fail to detect sophisticated attacks such as silicone masks. As the quality of presentation attack instruments improves over time, achieving reliable PA detection with visual spectra alone remains very challenging. We argue that analysis in multiple channels might help to address this issue. In this context, we propose a multi-channel Convolutional Neural Network based approach for presentation attack detection (PAD). We also introduce the new Wide Multi-Channel presentation Attack (WMCA) database for face PAD which contains a wide variety of 2D and 3D presentation attacks for both impersonation and obfuscation attacks. Data from different channels such as color, depth, near-infrared and thermal are available to advance the research in face PAD. The proposed method was compared with feature-based approaches and found to outperform the baselines achieving an ACER of 0.3% on the introduced dataset. The database and the software to reproduce the results are made available publicly.	['Andre Anjos', 'Olegs Nikisins', 'Anjith George', 'Zohreh Mostaani', 'David Geissenbuhler', 'Sebastien Marcel']	2019-09-19	biometric-face-presentation-attack-detection	http://publications.idiap.ch/downloads/papers/2019/George_TIFS_2019.pdf	http://publications.idiap.ch/downloads/papers/2019/George_TIFS_2019.pdf	ieee-transactions-on-information-forensics-1	['face-presentation-attack-detection']	['computer-vision']	[ 4.49684650e-01 -4.92668033e-01 2.38812447e-01 -7.10190386e-02 -7.35401809e-01 -1.02918541e+00 7.02323794e-01 -4.76908460e-02 -2.06115365e-01 3.19840521e-01 -1.18010469e-01 -5.11963308e-01 -9.12812278e-02 -5.60818613e-01 -2.94459850e-01 -9.73009884e-01 -3.44135225e-01 -3.73507351e-01 -8.55136812e-02 -2.85330057e-01 3.70688200e-01 1.13485444e+00 -1.45838940e+00 5.89781821e-01 2.37373725e-01 1.32908833e+00 -5.60452342e-01 9.91136789e-01 5.56044877e-02 -1.32693231e-01 -9.26522374e-01 -1.00003159e+00 5.29028237e-01 -2.20196307e-01 -3.60068202e-01 -2.58482546e-01 8.42838228e-01 -4.54014361e-01 -2.94404954e-01 9.77591276e-01 9.50417995e-01 -3.93591523e-01 6.33896649e-01 -1.50530183e+00 -6.48692727e-01 1.00444198e-01 -9.14815128e-01 2.53568590e-01 6.50812864e-01 1.35377303e-01 3.76514286e-01 -9.79606152e-01 2.27144465e-01 1.36885440e+00 6.89531028e-01 8.15004349e-01 -9.33311105e-01 -1.15430129e+00 -5.08352697e-01 5.72756752e-02 -1.41608000e+00 -4.96564388e-01 1.02845693e+00 -2.90988058e-01 6.10624611e-01 3.26738864e-01 1.98374912e-01 1.50189328e+00 1.98083743e-01 3.99368346e-01 1.50237048e+00 -4.96949464e-01 -9.73656327e-02 2.24820390e-01 -3.06105334e-02 6.17035389e-01 5.68025470e-01 2.75390625e-01 -6.22135401e-01 -6.74400508e-01 4.55657065e-01 -1.59130976e-01 -1.12891853e-01 7.76112899e-02 -4.48620349e-01 6.56846702e-01 1.73033401e-01 2.58100539e-01 -1.50837868e-01 -6.66132942e-02 5.00456393e-01 3.68507534e-01 1.46223560e-01 1.27464920e-01 -3.16940784e-01 -5.43121397e-02 -7.88605511e-01 3.13633651e-01 7.87779748e-01 2.66255647e-01 4.93348211e-01 1.26383007e-01 1.10172644e-01 8.29495788e-01 5.74533582e-01 9.44307089e-01 1.34355590e-01 -2.51873076e-01 5.60646176e-01 1.96740732e-01 -7.01605380e-02 -1.46069360e+00 -3.21775734e-01 1.05542339e-01 -8.49827111e-01 5.92882574e-01 6.59360111e-01 -3.32337081e-01 -1.01887870e+00 1.41633952e+00 2.08939046e-01 2.26134539e-01 1.60161953e-03 6.41560376e-01 8.69773149e-01 2.81107455e-01 1.05185963e-01 -1.30149141e-01 1.63892579e+00 -1.95481151e-01 -8.44189227e-01 2.58820742e-01 1.70717195e-01 -1.26526880e+00 6.23314321e-01 7.42075145e-01 -8.29394102e-01 -4.46595669e-01 -1.24351954e+00 5.00086427e-01 -7.05214024e-01 -8.38124007e-02 6.58703148e-01 1.90152466e+00 -1.08447778e+00 2.72668451e-01 -3.92739683e-01 -3.31010431e-01 9.21174884e-01 6.86877549e-01 -9.01651621e-01 -4.45993394e-02 -1.19781888e+00 4.72011983e-01 -1.13803454e-01 3.45698774e-01 -7.26430893e-01 -4.11271065e-01 -6.85607255e-01 -2.21274108e-01 -4.01398838e-02 2.51353949e-01 8.89829040e-01 -8.13299656e-01 -1.44150913e+00 8.68294239e-01 1.14654563e-01 -1.52009934e-01 3.72301519e-01 -2.55253255e-01 -8.83414924e-01 3.51580977e-01 -5.70978999e-01 2.18557432e-01 1.36088777e+00 -1.07642615e+00 -1.13882102e-01 -5.67984045e-01 -6.79138675e-02 -4.59749311e-01 -7.06606328e-01 5.71509659e-01 5.43654710e-02 -6.29151344e-01 -1.39647767e-01 -8.37189078e-01 3.60777110e-01 2.26526514e-01 -4.04111177e-01 9.75646749e-02 1.34004903e+00 -8.75418127e-01 9.47954535e-01 -2.11086941e+00 -4.79899317e-01 4.49441195e-01 -7.29482472e-02 1.03676522e+00 -2.37379372e-01 7.60031223e-01 -3.97670865e-01 4.09656674e-01 -1.73680529e-01 -3.71094137e-01 -2.23917495e-02 -1.78012699e-01 -2.42316067e-01 9.50369418e-01 3.68083119e-01 5.85926235e-01 -2.62132257e-01 -1.78608432e-01 2.24204123e-01 1.00469327e+00 -1.97140947e-01 8.12050104e-02 3.76404315e-01 2.41048947e-01 -2.73968667e-01 1.17714536e+00 1.60727119e+00 3.74330252e-01 -1.27743214e-01 -1.98934242e-01 2.05049157e-01 -2.13291556e-01 -1.31072450e+00 1.17771482e+00 -1.30384713e-01 7.68760204e-01 2.51588374e-01 -6.97015762e-01 1.07071495e+00 6.97347641e-01 2.98267812e-01 -5.50536335e-01 2.64829278e-01 3.08738530e-01 1.96049422e-01 -6.04512095e-01 1.13083608e-01 1.01703286e-01 2.75489658e-01 5.38838387e-01 5.88744506e-02 4.14599061e-01 -2.84881920e-01 -6.57497868e-02 1.01310706e+00 -1.47745863e-01 1.42798215e-01 -9.36799720e-02 9.18574512e-01 -7.90976942e-01 1.26879424e-01 6.53381467e-01 -6.92049682e-01 4.40289915e-01 3.37904721e-01 -4.95731592e-01 -8.49926829e-01 -1.01646173e+00 -3.19181114e-01 5.84333479e-01 -9.94598120e-02 -4.28756297e-01 -9.34644341e-01 -7.48340189e-01 2.86955565e-01 -3.98986429e-01 -6.27167881e-01 1.50138035e-01 -5.15920520e-01 -8.06250930e-01 1.24864328e+00 2.48770922e-01 9.38978851e-01 -8.63385797e-01 -4.02700961e-01 -2.50487238e-01 1.43033415e-01 -1.18484926e+00 -1.99059267e-02 -3.28492939e-01 -3.44267488e-01 -1.32426250e+00 -8.87115121e-01 -4.34700251e-01 4.58224356e-01 2.02373385e-01 3.92634392e-01 4.23450053e-01 -9.13949847e-01 5.36258936e-01 -3.40493113e-01 -6.96983516e-01 -3.72850657e-01 -2.64170617e-01 2.94473708e-01 4.35616404e-01 7.69041717e-01 -3.81183743e-01 -7.63173819e-01 1.74094126e-01 -9.72770214e-01 -7.93620646e-01 5.11236787e-01 6.16824746e-01 -2.96347827e-01 8.61028433e-02 3.74762177e-01 -7.36945808e-01 8.75625074e-01 -2.49976695e-01 -4.33562189e-01 7.87436292e-02 -3.02495629e-01 -3.75283539e-01 3.13639939e-01 -6.96024477e-01 -8.60146105e-01 -1.77270919e-02 -4.76703584e-01 -2.89553970e-01 -7.80073583e-01 1.07827276e-01 -4.24661368e-01 -8.46177220e-01 8.03161085e-01 1.14152648e-01 1.11046985e-01 -4.68683302e-01 2.31943484e-02 9.83974934e-01 2.48067215e-01 -4.19379115e-01 1.28249753e+00 4.83961284e-01 3.89295101e-01 -1.27137578e+00 4.04858328e-02 -3.59970421e-01 -5.11382282e-01 -4.07932520e-01 6.06599569e-01 -4.99735385e-01 -1.28712475e+00 1.18441939e+00 -1.01962042e+00 3.84896576e-01 8.80446613e-01 5.24769388e-02 7.24181309e-02 9.25600827e-01 -7.36170650e-01 -1.45610952e+00 -3.87477189e-01 -1.21584630e+00 1.06618047e+00 3.26663405e-01 6.86532781e-02 -7.96910286e-01 -1.52929276e-01 2.43296459e-01 8.71415079e-01 7.20811307e-01 4.18073773e-01 -6.59643352e-01 -2.42327407e-01 -7.02108800e-01 -2.79397964e-01 3.49584401e-01 5.61497390e-01 2.87988454e-01 -1.77642667e+00 -5.81078410e-01 -1.41540198e-02 -2.23818958e-01 6.35170162e-01 1.76949605e-01 1.11943126e+00 -3.59630674e-01 -2.36600861e-01 4.97830600e-01 1.35675943e+00 2.82555014e-01 1.06410575e+00 3.56761992e-01 5.46611011e-01 9.37489033e-01 7.75586739e-02 5.83505988e-01 -3.69559318e-01 7.92100251e-01 4.51688111e-01 -9.42261145e-02 -4.81369421e-02 5.87107204e-02 2.96144277e-01 -1.51408330e-01 -3.91804092e-02 -3.43924344e-01 -1.03128564e+00 2.05437206e-02 -1.16200650e+00 -1.00000906e+00 -5.03462739e-02 2.17626286e+00 2.90823907e-01 -8.12982246e-02 3.58277202e-01 8.05099308e-01 6.35456324e-01 1.56675160e-01 -2.40670927e-02 -8.18918109e-01 -3.38850319e-01 6.66859210e-01 6.12724781e-01 3.40537667e-01 -1.45414996e+00 6.10309064e-01 5.86820412e+00 7.06143975e-01 -1.43143332e+00 -1.12357281e-01 6.33673251e-01 4.30361599e-01 1.95502654e-01 -4.41470504e-01 -9.31184411e-01 5.09574413e-01 1.01949716e+00 4.90626127e-01 2.63608515e-01 5.15612125e-01 -7.18960464e-02 3.06810420e-02 -6.11549675e-01 1.57658577e+00 3.72131675e-01 -1.10984623e+00 6.09829426e-02 5.07572949e-01 2.95549482e-01 -5.98182976e-01 7.01329172e-01 -3.81365836e-01 -3.50689590e-01 -1.28821445e+00 8.99764672e-02 4.82573807e-02 8.34218264e-01 -9.43118632e-01 7.39129424e-01 -2.30236918e-01 -1.31685102e+00 -2.62013406e-01 -2.06067890e-01 1.39357790e-01 -1.72994941e-01 1.91378772e-01 -6.59023762e-01 6.28624678e-01 6.44076347e-01 2.66557217e-01 -8.07562590e-01 9.70571280e-01 5.36491610e-02 7.26133049e-01 -4.63833570e-01 1.14791773e-01 1.69404019e-02 2.93513179e-01 4.58074570e-01 1.36320746e+00 4.06309903e-01 -5.49743995e-02 -1.69426501e-01 3.59707624e-01 1.26157239e-01 1.10259980e-01 -9.17321622e-01 -3.40499401e-01 3.14726502e-01 1.43070674e+00 -6.58981264e-01 2.65799373e-01 -5.73657453e-01 1.09853518e+00 -2.94992179e-01 3.87138426e-01 -5.13351560e-01 -4.95823056e-01 9.37430918e-01 -9.24196690e-02 3.22683692e-01 -4.05779094e-01 -4.00544256e-02 -8.41945410e-01 4.57847640e-02 -1.14935434e+00 6.03680074e-01 -2.29209006e-01 -1.42064166e+00 7.51610994e-01 -1.66439265e-01 -1.08108878e+00 1.18470816e-02 -1.27442050e+00 -5.83652020e-01 1.27673352e+00 -1.59171271e+00 -1.40433395e+00 -3.14803183e-01 8.33984315e-01 1.55014127e-01 -5.84867954e-01 1.09814525e+00 5.79965115e-01 -4.35848117e-01 1.19743776e+00 -4.18221265e-01 4.49921072e-01 7.93581784e-01 -8.32732260e-01 5.76730371e-01 9.97869492e-01 1.48901001e-01 7.82367587e-01 5.30187011e-01 -6.19741440e-01 -1.78570330e+00 -4.71515119e-01 4.41944361e-01 -4.39768732e-01 3.28852981e-01 -4.92909312e-01 -9.21144128e-01 1.09883152e-01 3.23152483e-01 7.60531574e-02 1.02308488e+00 -2.01419294e-01 -9.52526748e-01 -1.16028741e-01 -1.62881804e+00 5.10072827e-01 5.11160612e-01 -7.89423704e-01 1.02647774e-01 7.75811356e-03 -1.40054122e-01 -1.64663941e-01 -5.87658346e-01 3.29281062e-01 9.92278934e-01 -9.74922061e-01 1.17295921e+00 -4.01250005e-01 -8.03732574e-02 -2.20605597e-01 -4.23558235e-01 -5.54980278e-01 1.28837913e-01 -1.01488590e+00 -2.45173588e-01 1.36278057e+00 2.02006087e-01 -7.96595871e-01 9.31308150e-01 4.54819292e-01 5.96322000e-01 -2.64648139e-01 -9.27625060e-01 -7.38698661e-01 -1.23842517e-02 -5.36707938e-01 6.94822848e-01 8.31118405e-01 -8.66262168e-02 -3.84258568e-01 -6.08239949e-01 6.88369036e-01 1.03411102e+00 -4.59681958e-01 7.88960397e-01 -1.29602516e+00 -4.16549705e-02 -2.95441568e-01 -9.40562546e-01 -3.74826372e-01 3.33887227e-02 -4.96771783e-01 -5.00018239e-01 -6.32741928e-01 -1.22712687e-01 -1.03759445e-01 -3.38442057e-01 4.25866872e-01 5.68594635e-02 9.22538161e-01 1.22837916e-01 -1.29107367e-02 3.41016144e-01 -8.47165063e-02 7.39649653e-01 -8.64886642e-02 1.50706932e-01 1.64446887e-02 -5.09159327e-01 4.44256246e-01 9.79027987e-01 -1.64847165e-01 -2.29829773e-01 1.40245169e-01 -1.27815589e-01 -1.47189945e-01 3.67029727e-01 -1.11134958e+00 5.00940196e-02 1.07680038e-01 8.72363806e-01 -4.98262793e-01 6.65239811e-01 -8.79934549e-01 -6.74422607e-02 6.03068531e-01 -1.87095046e-01 2.79666722e-01 6.49930000e-01 4.19242412e-01 1.22152241e-02 1.78283826e-02 9.04474854e-01 4.27759774e-02 -4.57086563e-01 5.42976260e-01 -4.02251542e-01 -5.72083056e-01 9.34033394e-01 -6.01524889e-01 -6.80511951e-01 -3.68540585e-01 -5.07233381e-01 -5.79847872e-01 2.43588388e-01 5.16195476e-01 9.52293456e-01 -1.18899727e+00 -6.95299029e-01 8.14027727e-01 1.15483440e-01 -9.99490499e-01 3.01228255e-01 4.78628993e-01 -7.07790673e-01 4.11227912e-01 -5.69701254e-01 -4.48258519e-01 -2.09111929e+00 5.06016791e-01 3.96202087e-01 4.18890655e-01 -3.62654358e-01 9.52873528e-01 -1.55587912e-01 6.21625111e-02 3.80817264e-01 3.74413788e-01 -3.56995463e-01 -5.80230430e-02 1.29262376e+00 2.62214959e-01 2.19842896e-01 -9.29929078e-01 -5.61145723e-01 5.76333225e-01 -2.73028463e-01 -1.90428570e-01 9.15343523e-01 4.64731045e-02 -1.61136895e-01 -1.50047317e-01 1.47054446e+00 1.50136515e-01 -8.21779966e-01 4.22965027e-02 6.61579594e-02 -1.12836409e+00 -3.03352624e-01 -7.68518984e-01 -1.03147745e+00 1.19193625e+00 1.30898011e+00 5.30267596e-01 1.16233695e+00 -4.63778079e-01 6.27246797e-01 3.10341120e-01 1.55684859e-01 -5.62656939e-01 2.44744331e-01 -3.07042003e-02 7.60108650e-01 -1.28655934e+00 -2.08461285e-01 -7.19070613e-01 -2.95660257e-01 1.48819876e+00 5.07108748e-01 8.29000175e-02 9.24731791e-01 5.07259369e-01 4.10433739e-01 -2.68530011e-01 -1.39323041e-01 1.24322899e-01 2.94275820e-01 1.24294007e+00 5.60954034e-01 -1.87124714e-01 -5.40804230e-02 5.34907952e-02 -3.94487195e-02 -3.30194652e-01 5.42697191e-01 1.01340866e+00 -5.41682839e-02 -1.57295656e+00 -8.00749600e-01 1.27867177e-01 -1.16197145e+00 1.28635570e-01 -7.10594058e-01 6.95715904e-01 6.09205253e-02 1.15148413e+00 -4.01421577e-01 -5.01364589e-01 6.22041784e-02 1.58138081e-01 5.66081166e-01 -3.51185948e-02 -6.88524663e-01 2.46591419e-02 -3.81562710e-02 -4.59093422e-01 -6.33982420e-01 -5.20073354e-01 -3.86070669e-01 -6.92132890e-01 -3.11827451e-01 -1.52140349e-01 1.08062518e+00 7.19585776e-01 1.21678136e-01 -6.35737106e-02 8.13588858e-01 -7.55220830e-01 -2.55211681e-01 -8.37854981e-01 -6.79451346e-01 4.87764299e-01 8.36017847e-01 -7.65753329e-01 -3.49154383e-01 -1.29766345e-01]	[13.074593544006348, 1.0952194929122925]
8d9c1ddb-324f-41fa-b456-e81ce629a659	alterfactual-explanations-the-relevance-of	2207.09374	null	https://arxiv.org/abs/2207.09374v1	https://arxiv.org/pdf/2207.09374v1.pdf	Alterfactual Explanations -- The Relevance of Irrelevance for Explaining AI Systems	Explanation mechanisms from the field of Counterfactual Thinking are a widely-used paradigm for Explainable Artificial Intelligence (XAI), as they follow a natural way of reasoning that humans are familiar with. However, all common approaches from this field are based on communicating information about features or characteristics that are especially important for an AI's decision. We argue that in order to fully understand a decision, not only knowledge about relevant features is needed, but that the awareness of irrelevant information also highly contributes to the creation of a user's mental model of an AI system. Therefore, we introduce a new way of explaining AI systems. Our approach, which we call Alterfactual Explanations, is based on showing an alternative reality where irrelevant features of an AI's input are altered. By doing so, the user directly sees which characteristics of the input data can change arbitrarily without influencing the AI's decision. We evaluate our approach in an extensive user study, revealing that it is able to significantly contribute to the participants' understanding of an AI. We show that alterfactual explanations are suited to convey an understanding of different aspects of the AI's reasoning than established counterfactual explanation methods.	['Elisabeth André', 'Ruben Schlagowski', 'Katharina Weitz', 'Tobias Huber', 'Christina Karle', 'Silvan Mertes']	2022-07-19	null	null	null	null	['counterfactual-explanation']	['miscellaneous']	[ 5.91827512e-01 8.86094868e-01 1.09263450e-01 -3.86196703e-01 1.86884090e-01 -6.06072664e-01 1.00364840e+00 4.29280907e-01 -2.46542349e-01 6.05026662e-01 4.07252550e-01 -7.68573165e-01 -1.54025748e-01 -9.83557642e-01 -7.31586814e-01 -1.18325315e-01 1.81323066e-01 3.95743221e-01 1.00428708e-01 -6.43609345e-01 6.28613353e-01 4.03839260e-01 -1.98178017e+00 5.58881164e-01 7.76166856e-01 4.09381360e-01 -1.36296511e-01 5.38247228e-01 -3.91125858e-01 8.98998082e-01 -5.98517954e-01 -2.04246908e-01 6.15478940e-02 -7.31165349e-01 -1.36633420e+00 -7.75120929e-02 -6.22920282e-02 -2.35042110e-01 3.45302373e-01 9.98684406e-01 -2.23321542e-01 5.73706403e-02 6.43080115e-01 -1.51675892e+00 -5.42024493e-01 9.83357728e-01 6.59171045e-02 -2.24993587e-01 9.24287438e-01 5.23564279e-01 8.07498515e-01 -2.75934964e-01 5.24687946e-01 1.53293908e+00 2.01154068e-01 8.19177330e-01 -1.55259323e+00 -3.23452562e-01 6.01042449e-01 4.03416902e-01 -1.01607871e+00 -3.82777572e-01 9.25314486e-01 -5.51101387e-01 7.26582527e-01 8.34936142e-01 9.68850136e-01 8.17903697e-01 2.65722752e-01 6.20232224e-01 1.33095765e+00 -9.57276821e-01 7.34540045e-01 6.71431184e-01 3.56176555e-01 2.48330608e-01 5.00512123e-01 4.46223438e-01 -4.55706924e-01 -2.90139407e-01 5.15614390e-01 -3.78030576e-02 -3.84955347e-01 -8.14994156e-01 -1.35723114e+00 7.45449305e-01 3.17212462e-01 8.40198636e-01 -5.98746657e-01 2.58861154e-01 2.91000865e-02 3.77838701e-01 1.46070486e-02 1.09454525e+00 -3.87206465e-01 -2.29869917e-01 -2.09661916e-01 5.93564928e-01 1.08652842e+00 4.54773486e-01 5.83277881e-01 -3.10019255e-01 -1.25149861e-01 -1.83349073e-01 5.49840808e-01 3.84102404e-01 5.92637181e-01 -9.38114583e-01 -1.05593756e-01 1.22424114e+00 5.84791601e-01 -8.84664774e-01 -1.86104447e-01 -1.04386091e-01 -1.97989836e-01 8.05763543e-01 5.43488801e-01 -2.71255150e-02 -7.62441337e-01 1.87230909e+00 1.88990861e-01 -2.77243257e-01 3.78532499e-01 9.45719063e-01 3.19384247e-01 2.62205243e-01 1.78136408e-01 -4.04218853e-01 1.43254828e+00 -2.85928726e-01 -9.25475538e-01 -3.55321109e-01 6.79836750e-01 -4.51611042e-01 1.35996008e+00 3.70744437e-01 -1.05026400e+00 -5.25651038e-01 -1.25307453e+00 2.78710842e-01 -5.21275163e-01 -6.02001905e-01 1.00633609e+00 7.87422538e-01 -7.91877925e-01 6.97662532e-01 -3.60845506e-01 -5.22494256e-01 1.31563917e-01 3.46721292e-01 -1.08623505e-01 6.74910471e-02 -1.29538381e+00 1.07609463e+00 4.81234998e-01 -6.04585409e-02 -6.07637048e-01 -4.29778099e-01 -7.63638735e-01 4.54049885e-01 7.51869559e-01 -8.56995404e-01 1.63159049e+00 -1.30053604e+00 -1.50204825e+00 5.83716094e-01 -2.59314895e-01 -5.64578772e-01 6.62957966e-01 -1.38489798e-01 -3.52707475e-01 -1.62367627e-01 -8.20365734e-03 5.46494961e-01 7.07281649e-01 -1.57393909e+00 -2.83122689e-01 -4.89254206e-01 6.62847519e-01 2.03534499e-01 2.72955060e-01 -3.58443558e-01 4.58937645e-01 -1.41307801e-01 2.18099549e-01 -8.59354913e-01 -4.87880617e-01 -4.63328622e-02 -5.13172090e-01 -5.19120134e-02 6.31664395e-01 5.18564768e-02 1.09162998e+00 -1.91019177e+00 -1.74266711e-01 4.66466963e-01 4.13331330e-01 9.87343714e-02 1.43343359e-01 5.93610287e-01 -4.04777437e-01 5.59966147e-01 -2.04385117e-01 3.33208531e-01 3.33873153e-01 3.01024504e-02 -3.92123133e-01 -1.03453167e-01 5.33764958e-02 7.97128141e-01 -7.16617465e-01 -8.23469739e-03 5.04050672e-01 1.35716408e-01 -6.28752530e-01 3.79753917e-01 -5.02697408e-01 4.15497214e-01 -4.39767867e-01 -1.14971362e-01 4.77463186e-01 -6.41931742e-02 4.06821460e-01 1.23537436e-01 -2.87000358e-01 4.55542773e-01 -1.22195959e+00 1.19973719e+00 -3.90936524e-01 3.92060876e-01 -7.08403289e-01 -6.90514982e-01 5.40385604e-01 5.69440126e-01 -2.54760504e-01 -4.20697510e-01 2.87155420e-01 7.01567382e-02 6.27886713e-01 -5.94882727e-01 -6.02007098e-02 -5.29663861e-01 7.82925561e-02 1.05652952e+00 -7.04276383e-01 -5.02488792e-01 7.52642453e-02 3.65099400e-01 7.78278887e-01 3.55713367e-02 1.14165211e+00 -3.94858003e-01 6.19500756e-01 2.20214620e-01 2.31750965e-01 1.00720048e+00 -1.18216025e-02 1.71936497e-01 6.63060427e-01 -1.05100238e+00 -5.44610798e-01 -7.62236714e-01 3.06513488e-01 4.78917897e-01 3.28918397e-01 -2.46570542e-01 -9.70937908e-01 -7.62627006e-01 -1.51438698e-01 1.73651636e+00 -1.03812015e+00 -5.39060056e-01 -2.05631211e-01 -1.44171461e-01 -7.69770741e-02 2.49948815e-01 4.66129810e-01 -1.41033649e+00 -1.67462051e+00 9.33576152e-02 -3.97051841e-01 -4.38677460e-01 -1.02600925e-01 -1.96921863e-02 -1.03331435e+00 -1.24738777e+00 8.58664587e-02 1.64224088e-01 8.12869906e-01 4.63948965e-01 1.18386006e+00 6.36681080e-01 -8.03057924e-02 4.26551670e-01 -2.67862797e-01 -9.29789782e-01 -8.54960680e-01 -5.18162668e-01 -3.78916673e-02 -8.26878771e-02 7.13854969e-01 -6.26611471e-01 -4.23322916e-01 1.39395401e-01 -1.00768256e+00 5.50499082e-01 5.19599319e-01 3.93550634e-01 3.44387516e-02 1.03525758e-01 9.51719433e-02 -1.37478411e+00 9.78626728e-01 -3.02756637e-01 -2.16122285e-01 4.41820085e-01 -1.00315881e+00 6.42489612e-01 6.20414078e-01 -4.52231288e-01 -1.41848278e+00 -1.46059290e-01 3.63109320e-01 3.74087900e-01 -7.46944308e-01 2.98335880e-01 -5.52706897e-01 2.35122621e-01 8.75250280e-01 6.19960800e-02 -9.03468728e-02 -1.09806776e-01 6.77076757e-01 2.52998561e-01 1.33553356e-01 -6.78727150e-01 8.00041914e-01 4.41404074e-01 -6.20909557e-02 -2.36251131e-01 -4.68041986e-01 2.79779360e-02 -5.06768465e-01 -1.88386813e-01 7.29457855e-01 -8.02749246e-02 -1.28842139e+00 -1.46890441e-02 -1.30046356e+00 -1.45134777e-01 -8.13910246e-01 3.91364038e-01 -7.51095653e-01 -3.15780528e-02 1.93833739e-01 -1.24897826e+00 2.16817394e-01 -1.17548573e+00 6.13996327e-01 1.73419446e-01 -1.01572549e+00 -7.64897823e-01 -9.26406123e-03 1.66690797e-01 5.30649960e-01 7.57252201e-02 1.27884340e+00 -1.04895866e+00 -5.31066835e-01 -1.48778766e-01 1.45322429e-02 -1.59658298e-01 3.79721373e-01 -2.75842726e-01 -1.14548278e+00 1.63084984e-01 4.32096660e-01 2.29662940e-01 1.17203198e-01 1.43804058e-01 7.95803607e-01 -7.39308476e-01 -4.59868610e-01 -3.77320617e-01 1.29276061e+00 7.22188711e-01 7.83033431e-01 2.70182192e-01 2.11486757e-01 9.85787988e-01 6.36860073e-01 2.83093512e-01 2.80432433e-01 7.84051955e-01 5.77173591e-01 -1.46762550e-01 2.99163550e-01 -4.07712877e-01 5.92768565e-02 -3.19899797e-01 -4.05722231e-01 -5.61104529e-02 -9.19921517e-01 3.30031276e-01 -1.90052783e+00 -1.07888758e+00 -3.04465145e-01 2.42848063e+00 4.60837573e-01 2.40814313e-01 -1.35797977e-01 4.43978339e-01 5.22986531e-01 -2.63703167e-01 -5.77491224e-01 -8.67729664e-01 4.29634571e-01 -1.39683068e-01 -1.56250179e-01 8.01277041e-01 -4.31353897e-01 6.34384692e-01 6.32583952e+00 -2.22444311e-02 -6.99597418e-01 -2.53038406e-01 4.57682729e-01 2.85030156e-01 -9.48401928e-01 5.17670095e-01 -3.35425735e-02 1.67408869e-01 8.73330355e-01 -6.42030060e-01 3.50928217e-01 7.27223516e-01 5.07752240e-01 -5.80633104e-01 -1.82109654e+00 4.00997818e-01 -1.43981084e-01 -1.20196092e+00 5.06519735e-01 2.78615654e-01 1.58804148e-01 -1.16500986e+00 -1.21650934e-01 5.94702177e-02 4.69903588e-01 -1.18357944e+00 9.62209046e-01 4.84436989e-01 9.31591541e-02 -7.89803863e-01 1.07248652e+00 6.83084130e-01 -6.12826824e-01 -4.37170900e-02 1.77784190e-01 -9.50850189e-01 -2.11312510e-02 3.46863508e-01 -1.04767954e+00 2.32108295e-01 3.25176775e-01 -1.41183928e-01 -3.91001552e-01 4.90035504e-01 -7.58607984e-01 3.62166345e-01 2.71449075e-03 -1.50927424e-01 -1.78938121e-01 3.22670713e-02 6.34142399e-01 5.92797041e-01 8.63535255e-02 5.63084066e-01 -1.86426044e-01 1.33763146e+00 4.60977465e-01 7.86153525e-02 -1.05857146e+00 3.88723724e-02 4.03814405e-01 4.95479107e-01 -6.46280587e-01 -5.36259890e-01 -5.45829237e-02 8.84231627e-01 -1.00185573e-01 3.00446421e-01 -5.55431843e-01 -4.88170609e-02 6.30934656e-01 1.72659189e-01 -2.04791337e-01 2.64920026e-01 -4.63606775e-01 -1.09490228e+00 -9.75046232e-02 -1.21071708e+00 1.28532231e-01 -1.09407103e+00 -6.94931269e-01 4.98352885e-01 2.81079203e-01 -1.00304842e+00 -5.53883910e-01 -1.86806560e-01 -8.41174662e-01 9.68143642e-01 -1.05025721e+00 -1.09701347e+00 -2.77183473e-01 2.36577868e-01 4.19486612e-01 2.55743295e-01 1.15448642e+00 -5.82420647e-01 -4.66965735e-02 -9.21700746e-02 -8.36462319e-01 -5.22855759e-01 2.04566747e-01 -1.35681391e+00 6.70494378e-01 1.00995886e+00 3.81682843e-01 1.20821846e+00 1.42214191e+00 -4.82296914e-01 -1.20690489e+00 -2.25868925e-01 1.13607109e+00 -6.77242994e-01 2.93793738e-01 -8.90653133e-02 -1.07543910e+00 9.27057564e-01 3.46096516e-01 -5.97485304e-01 7.70816326e-01 3.28369021e-01 -2.83127785e-01 2.98863232e-01 -1.39434171e+00 1.07229555e+00 1.05864251e+00 -1.99578673e-01 -1.41507936e+00 -1.18864931e-01 7.98823893e-01 8.32725763e-02 -1.55833825e-01 1.94884822e-01 6.63578808e-01 -1.48618531e+00 6.95767879e-01 -9.94721174e-01 3.29582065e-01 -6.79909706e-01 -2.79912427e-02 -1.54012060e+00 -2.19512776e-01 -6.65832639e-01 1.56537890e-01 9.56828237e-01 4.81616408e-01 -1.04053116e+00 4.49376523e-01 1.47643995e+00 2.99822092e-01 -1.63994938e-01 -5.66804707e-01 -2.77619570e-01 -2.48830780e-01 -6.94209933e-01 1.25181115e+00 8.47225785e-01 7.38848567e-01 2.81880468e-01 8.20084289e-02 1.74448177e-01 3.50800276e-01 3.81747633e-01 9.32523787e-01 -1.53924203e+00 -3.67745280e-01 -3.10377598e-01 -3.53345454e-01 -7.08792686e-01 -1.04203716e-01 -4.99084353e-01 1.00754537e-01 -1.32682061e+00 3.48310262e-01 -9.37370211e-03 -4.86323386e-02 5.37204504e-01 -2.52584159e-01 -2.82037705e-01 4.75419194e-01 2.42858037e-01 -1.71016261e-01 1.92435071e-01 1.12013841e+00 4.39088307e-02 -3.85177433e-01 8.19563270e-02 -1.21153212e+00 1.11357033e+00 6.11556411e-01 -4.40955311e-01 -7.92761564e-01 -6.94103763e-02 3.49029720e-01 -5.95762245e-02 6.05385959e-01 -8.68651807e-01 3.24895456e-02 -6.57841146e-01 1.50110185e-01 8.13552514e-02 9.42758694e-02 -1.32228720e+00 5.34702420e-01 1.07992208e+00 -7.39359140e-01 -5.49654104e-02 3.75149727e-01 3.34229469e-01 -7.53237829e-02 -2.67481655e-01 3.73940885e-01 -3.04407090e-01 -7.50198543e-01 -5.34637272e-01 -9.36330795e-01 -4.19511288e-01 1.14061272e+00 -4.31612074e-01 -2.07105517e-01 -6.99476063e-01 -7.80649006e-01 6.18694816e-03 5.95346093e-01 2.70907938e-01 6.28794253e-01 -9.21387970e-01 -3.17545623e-01 3.54911268e-01 3.95225555e-01 -4.19676125e-01 1.35634705e-01 4.56229359e-01 -2.16297343e-01 6.20848835e-01 -3.18069935e-01 -1.94253266e-01 -1.24163282e+00 8.10248494e-01 4.39569056e-01 -9.38990191e-02 -4.36908036e-01 4.30957615e-01 8.02797973e-01 -1.72449917e-01 -2.13573918e-01 -6.02484286e-01 -4.94582206e-01 -2.66384482e-01 8.28952193e-01 3.41602676e-02 -1.93713844e-01 -1.41671866e-01 -3.92182380e-01 1.93351805e-01 -1.75031722e-01 -4.73527372e-01 1.06982625e+00 -2.84347624e-01 -5.46811111e-02 8.38987231e-01 1.48189873e-01 3.24728005e-02 -8.66209507e-01 8.98382440e-03 8.41377825e-02 -7.36432493e-01 -3.43362719e-01 -1.22723854e+00 -1.61689863e-01 9.04751360e-01 5.41011274e-01 6.57597005e-01 9.95776951e-01 7.23600015e-02 5.74475946e-03 5.39813459e-01 6.84162557e-01 -5.97900748e-01 -1.88807949e-01 -8.36843997e-02 1.28420937e+00 -1.23841953e+00 3.63683514e-02 -6.65683031e-01 -8.08453619e-01 1.23765779e+00 6.22220337e-01 4.14144218e-01 1.05467774e-01 4.75302935e-02 3.22206050e-01 -5.01503825e-01 -9.75367844e-01 -2.17216596e-01 1.57888293e-01 5.97906530e-01 6.21283472e-01 2.90536225e-01 -5.71178973e-01 7.17778921e-01 -4.26570863e-01 2.22354740e-01 8.62563491e-01 9.29662108e-01 -4.17992234e-01 -1.15510309e+00 -6.00388229e-01 8.52364600e-02 8.74358937e-02 -9.76676238e-04 -1.01741338e+00 1.18759716e+00 1.58453226e-01 1.22635484e+00 -1.84601456e-01 -7.48856068e-02 5.54410815e-01 3.40563506e-01 4.62317437e-01 -9.09699738e-01 -4.72345591e-01 -5.73332846e-01 2.26522401e-01 -8.08670104e-01 -4.82780367e-01 -6.59296930e-01 -1.37967801e+00 -3.35945338e-01 -1.07673317e-01 5.98495901e-01 6.88019931e-01 1.40141666e+00 1.68757737e-01 5.38485587e-01 2.75984198e-01 -4.27300423e-01 -3.42878908e-01 -7.74300456e-01 -1.73291579e-01 7.47378349e-01 2.24238381e-01 -6.14039838e-01 -5.59553325e-01 1.69335216e-01]	[8.885815620422363, 6.092493534088135]
c4a21854-0b92-4a67-8c88-922e94a50a24	evaluation-of-noise-reduction-methods-for	2303.17829	null	https://arxiv.org/abs/2303.17829v3	https://arxiv.org/pdf/2303.17829v3.pdf	Evaluation of Noise Reduction Methods for Sentence Recognition by Sinhala Speaking Listeners	Noise reduction is a crucial aspect of hearing aids, which researchers have been striving to address over the years. However, most existing noise reduction algorithms have primarily been evaluated using English. Considering the linguistic differences between English and Sinhala languages, including variation in syllable structures and vowel duration, it is very important to assess the performance of noise reduction tailored to the Sinhala language. This paper presents a comprehensive analysis between wavelet transformation and adaptive filters for noise reduction in Sinhala languages. We investigate the performance of ten wavelet families with soft and hard thresholding methods against adaptive filters with Normalized Least Mean Square, Least Mean Square Average Normalized Least Mean Square, Recursive Least Square, and Adaptive Filtering Averaging optimization algorithms along with cepstral and energy-based voice activity detection algorithms. The performance evaluation is done using objective metrics; Signal to Noise Ratio (SNR) and Perceptual Evaluation of Speech Quality (PESQ) and a subjective metric; Mean Opinion Score (MOS). A newly recorded Sinhala language audio dataset and the NOIZEUS database by the University of Texas, Dallas were used for the evaluation. Our code is available at https://github.com/ChathukiKet/Evaluation-of-Noise-Reduction-Methods	['Anjula De Silva', 'Nipuna Upeksha', 'Dinithi Fernando', 'Chathuki Navanjana', 'Malitha Gunawardhana']	2023-03-31	null	null	null	null	['activity-detection']	['computer-vision']	[-3.70551869e-02 -5.71626127e-01 3.61118585e-01 -5.53031676e-02 -1.09549582e+00 -4.05615151e-01 -2.87321638e-02 2.50433296e-01 -6.94095790e-01 6.44598663e-01 5.72980940e-01 -3.51353347e-01 -4.23001766e-01 -6.06913626e-01 1.90250814e-01 -8.33347499e-01 6.08608779e-03 -2.54952222e-01 3.38955700e-01 -4.80073214e-01 1.45739272e-01 4.60840374e-01 -1.70434391e+00 -5.23165800e-02 7.21100271e-01 8.26192439e-01 3.83180767e-01 1.13167107e+00 2.94242114e-01 3.12009215e-01 -7.90320277e-01 -1.44184977e-01 1.10003404e-01 -5.98853111e-01 -2.90911973e-01 -2.74683684e-01 -4.08115461e-02 -2.77927462e-02 -5.19444384e-02 1.24382234e+00 1.51605821e+00 3.81506890e-01 3.30794394e-01 -7.03976750e-01 -3.81797701e-01 6.39238834e-01 8.60222280e-02 8.03519905e-01 4.62877601e-01 1.35096133e-01 7.11215615e-01 -9.06893730e-01 1.69461131e-01 1.07008708e+00 8.33855510e-01 4.04188395e-01 -1.19007170e+00 -6.06597006e-01 -5.36557972e-01 6.04857504e-01 -1.71833909e+00 -8.72512937e-01 9.42072153e-01 -1.76157638e-01 1.22221243e+00 5.19973576e-01 5.60680687e-01 4.92117614e-01 1.67643517e-01 1.92862064e-01 1.18193483e+00 -9.75439548e-01 3.82253081e-01 5.58369048e-02 1.10127412e-01 2.54111588e-01 -8.54843557e-02 1.50990799e-01 -3.74507248e-01 -2.22913161e-01 3.59371036e-01 -6.05058491e-01 -5.41231036e-01 3.30185801e-01 -6.26156449e-01 7.18436539e-01 -2.61607915e-01 6.16307139e-01 -6.15893126e-01 -2.27194712e-01 6.13000274e-01 6.01441503e-01 4.95789438e-01 2.26627961e-02 -5.17573118e-01 -5.59474528e-01 -9.02491510e-01 1.62094221e-01 7.66740620e-01 4.35105383e-01 1.73729375e-01 6.32444441e-01 7.18678609e-02 1.58778667e+00 4.98509794e-01 8.39045823e-01 6.44279718e-01 -9.78927493e-01 -1.02554895e-02 -2.19230533e-01 -1.77481472e-01 -8.29030991e-01 -3.37734640e-01 -2.40583062e-01 -5.48206210e-01 4.30048883e-01 2.16824785e-01 -2.90075094e-01 -6.80496395e-01 1.55827820e+00 1.35019794e-01 -2.69273728e-01 8.50504264e-02 6.85959458e-01 6.99872196e-01 7.20249712e-01 -2.78208517e-02 -9.70613420e-01 1.29025364e+00 -5.30990481e-01 -1.36174250e+00 1.94887578e-01 5.47525026e-02 -1.39027143e+00 1.31942952e+00 6.80487156e-01 -1.41247857e+00 -6.39916480e-01 -8.99870396e-01 8.81524682e-02 -1.86112627e-01 -5.42465188e-02 -1.27233177e-01 1.46102059e+00 -1.09113586e+00 2.88313866e-01 -8.25419068e-01 -3.40014488e-01 -1.24137767e-01 2.73012429e-01 -2.04435326e-02 3.20345163e-01 -1.08890498e+00 7.64362872e-01 -1.46010876e-01 -1.68038458e-02 -3.53634924e-01 -3.67826372e-01 -7.19927132e-01 -6.24012128e-02 1.87730849e-01 -3.62597369e-02 1.36741853e+00 -4.01313215e-01 -1.76520991e+00 3.89333963e-01 -4.17011380e-01 -2.95467049e-01 1.48799032e-01 -2.46044442e-01 -1.12935245e+00 5.86113147e-02 -1.58510320e-02 -1.55860260e-01 6.51742041e-01 -8.40261757e-01 -5.64778268e-01 -2.95617580e-01 -6.14900589e-01 1.70108795e-01 -2.04686850e-01 5.71855366e-01 -1.26041293e-01 -1.06721437e+00 2.90154904e-01 -4.09246653e-01 -4.58314903e-02 -4.39324826e-01 -1.19843371e-01 -6.89817593e-02 5.90767860e-01 -1.38319802e+00 1.87703860e+00 -2.43517804e+00 -3.18009406e-01 3.09980214e-01 -5.07409275e-01 4.56940264e-01 -4.54332866e-02 4.21941727e-01 5.41711450e-02 -3.19916941e-02 -2.71063834e-01 -6.44556582e-02 -7.58133084e-02 7.69550726e-02 9.45428535e-02 3.97864312e-01 -2.23531827e-01 4.42781299e-02 -6.30366445e-01 -4.42939878e-01 4.46206331e-01 8.08145046e-01 -4.83453274e-01 -7.83281028e-02 5.16777515e-01 7.21645430e-02 1.60702676e-01 8.05594087e-01 6.51132107e-01 1.04810965e+00 -2.55804420e-01 -3.98047477e-01 -5.23809552e-01 6.36178076e-01 -1.62719524e+00 1.03160989e+00 -4.94019866e-01 5.67269325e-01 4.73529458e-01 -7.77758479e-01 1.00673914e+00 8.29988301e-01 2.68314570e-01 -8.04866433e-01 2.45494395e-01 7.04940081e-01 3.23443085e-01 -7.13656187e-01 4.56429183e-01 -2.94827163e-01 3.98764014e-01 3.33176851e-02 2.81418785e-02 -5.60350657e-01 5.09449244e-01 -3.44432294e-01 8.73281837e-01 -4.56180841e-01 5.52872062e-01 -3.67866158e-01 7.61754274e-01 -5.17183125e-01 5.87982416e-01 3.76460612e-01 -7.35028088e-01 7.26840615e-01 1.07194958e-02 2.43267387e-01 -1.07157254e+00 -1.30861437e+00 -4.73420888e-01 1.16113102e+00 -5.25139213e-01 -3.38368863e-01 -8.98645937e-01 3.32362980e-01 -3.41114551e-01 8.70361805e-01 1.30241513e-01 -7.12713674e-02 -5.48920870e-01 -5.16070247e-01 6.60460413e-01 1.92602858e-01 4.57323164e-01 -1.24930525e+00 -5.04985511e-01 5.00574589e-01 -1.54190093e-01 -8.97474647e-01 -6.07698381e-01 4.01987135e-01 -6.21876001e-01 -6.41062498e-01 -5.08328974e-01 -1.04254162e+00 -6.39274418e-02 -5.72041944e-02 6.32703483e-01 -3.60828489e-01 -3.63290548e-01 5.82244277e-01 -5.04550219e-01 -7.45402992e-01 -5.22592127e-01 -2.91898817e-01 4.47218001e-01 -2.79030770e-01 6.48281038e-01 -8.47434044e-01 -5.69923162e-01 3.67301494e-01 -6.38738394e-01 -8.46735775e-01 3.51971447e-01 5.19977212e-01 7.37178981e-01 7.84201860e-01 6.39720619e-01 -9.58033279e-02 1.26166570e+00 -5.88755608e-02 -3.34041268e-01 -1.82787955e-01 -4.73086983e-01 -6.41665697e-01 4.54277605e-01 -4.07344967e-01 -9.82390225e-01 -3.08696449e-01 -8.76268148e-01 2.48435050e-01 -2.45919362e-01 4.80195284e-01 -4.13951606e-01 1.02433816e-01 7.71501243e-01 3.74147654e-01 1.76585789e-04 -7.55633116e-01 -2.01302506e-02 1.29843414e+00 7.17665017e-01 -1.99432239e-01 4.69541192e-01 2.03943700e-01 -5.25436878e-01 -1.61983776e+00 7.24315643e-03 -6.54997528e-01 -4.08447027e-01 -3.25883627e-01 7.51129270e-01 -6.33506119e-01 -4.27325726e-01 8.90679777e-01 -8.25029314e-01 -2.33332321e-01 -3.73420894e-01 8.01712692e-01 -5.06999731e-01 3.96576941e-01 -6.26085758e-01 -1.42389691e+00 -5.63873649e-01 -9.83818233e-01 3.31501603e-01 2.49663934e-01 -4.36850131e-01 -8.97817492e-01 5.88652901e-02 2.40257904e-01 7.23586857e-01 -1.22182466e-01 8.20155144e-01 -2.37429142e-01 4.11333799e-01 -1.44278556e-01 4.18772548e-01 1.07524061e+00 5.60566187e-01 5.70532568e-02 -1.07521296e+00 -2.40511626e-01 5.68137646e-01 2.20123231e-01 5.08999765e-01 1.06403494e+00 5.79157591e-01 -2.79271603e-01 4.04457331e-01 1.41885132e-01 1.22593284e+00 8.81984413e-01 6.65956318e-01 2.02167615e-01 -2.83034686e-02 3.28168631e-01 5.12163281e-01 5.05028605e-01 1.32507250e-01 6.43190384e-01 -5.91196828e-02 9.93912518e-02 -5.36803067e-01 1.23316109e-01 4.83484596e-01 1.53514993e+00 -2.97514200e-01 -1.09594904e-01 -7.33212292e-01 7.34124601e-01 -1.05729735e+00 -1.02624357e+00 -7.88215846e-02 2.35936713e+00 1.06915104e+00 2.83587605e-01 2.82724261e-01 1.20819366e+00 7.90918946e-01 -5.12136146e-02 -8.43843594e-02 -8.97228301e-01 -3.37488919e-01 7.64131725e-01 3.49989057e-01 1.06694543e+00 -9.62184072e-01 6.20691538e-01 5.96407652e+00 1.09828174e+00 -1.13346636e+00 3.47658753e-01 1.78600639e-01 -6.76205158e-02 7.95814097e-02 -4.61492062e-01 -5.13346493e-01 4.33640003e-01 1.16097426e+00 -4.11941767e-01 6.51245475e-01 5.65888941e-01 1.01441479e+00 -2.61398882e-01 -3.46195191e-01 9.61596727e-01 -1.76811039e-01 -6.77824259e-01 -3.64002496e-01 -1.04844145e-01 4.12736237e-01 -3.29632051e-02 9.41558853e-02 1.38820753e-01 -2.74788886e-01 -6.33568227e-01 8.53991866e-01 4.93888408e-01 5.67817628e-01 -8.69025111e-01 8.01135063e-01 -5.22291847e-02 -1.31715488e+00 -1.61543682e-01 -3.84792984e-01 -3.04484248e-01 3.14931303e-01 7.22904503e-01 -5.51968455e-01 6.33327290e-02 1.04700947e+00 -8.71524587e-02 -7.81938210e-02 1.36836183e+00 -2.24118680e-01 1.06621027e+00 -5.59711993e-01 -1.52664647e-01 -3.32993239e-01 -1.61002368e-01 9.35979068e-01 1.25687957e+00 5.01408160e-01 4.68472749e-01 -2.92056918e-01 2.23891556e-01 4.86265481e-01 5.84519327e-01 -1.76572353e-01 9.70222205e-02 8.73055160e-01 7.74213672e-01 -5.31207919e-01 1.04844041e-01 -4.92783874e-01 4.80230868e-01 -6.26407743e-01 3.68215799e-01 -4.42451775e-01 -8.83376479e-01 8.01207960e-01 2.05533192e-01 1.63086757e-01 -4.50988948e-01 -4.86341178e-01 -3.16495031e-01 2.20163763e-02 -1.04871523e+00 2.39754021e-01 -6.49443388e-01 -9.91855741e-01 6.14547014e-01 -4.64920737e-02 -9.67182934e-01 -5.26646897e-02 -5.76962769e-01 -6.82920277e-01 1.15895665e+00 -1.08388269e+00 -3.79724056e-01 1.72947899e-01 5.06880760e-01 8.63508999e-01 -3.39751124e-01 9.36046541e-01 7.88089633e-01 -3.55284065e-01 7.53735065e-01 2.58440673e-01 -6.87443912e-02 4.23231572e-01 -1.10920000e+00 1.08944371e-01 9.26025629e-01 -2.30849952e-01 4.65487778e-01 1.13252044e+00 -4.90698665e-01 -8.83644581e-01 -5.80582857e-01 1.12073672e+00 2.59601802e-01 6.71379626e-01 -9.30871740e-02 -9.31611061e-01 7.60344369e-03 2.34733820e-01 -4.40447181e-01 9.05463636e-01 -1.99381262e-01 3.28572899e-01 -3.19307625e-01 -1.36868584e+00 6.00070238e-01 7.32496798e-01 -7.64225602e-01 -4.80486393e-01 -3.62033434e-02 5.76921940e-01 -1.27450839e-01 -9.53443825e-01 3.97159606e-01 5.53794146e-01 -9.89273310e-01 8.32479239e-01 1.15383968e-01 -1.78117171e-01 -4.99329835e-01 -6.27346814e-01 -1.38642919e+00 -3.52974087e-01 -7.39536703e-01 2.55093098e-01 1.59629095e+00 5.54129362e-01 -7.03659534e-01 2.93689251e-01 1.28512993e-01 -3.36646736e-01 -3.47733855e-01 -1.27141762e+00 -8.94795954e-01 -2.01015443e-01 -1.06046116e+00 9.08374041e-02 4.18647021e-01 6.03804924e-02 2.21023709e-01 -2.74277538e-01 3.25996876e-01 5.38045764e-01 -9.27345812e-01 2.26912484e-01 -7.69178569e-01 -1.99269861e-01 -6.27820134e-01 -6.05677485e-01 -3.03701580e-01 -5.28245389e-01 -2.82810360e-01 3.85766453e-03 -1.48634362e+00 -3.61241639e-01 1.40964240e-01 -5.02708614e-01 3.76777470e-01 -4.12374362e-02 4.24621224e-01 6.40622899e-02 -2.32365087e-01 2.59419143e-01 4.43985581e-01 8.26439977e-01 -5.83307445e-02 -5.80904186e-01 5.69596708e-01 -3.54471475e-01 7.67767787e-01 9.59987164e-01 -3.86010200e-01 -4.53218162e-01 -1.88262194e-01 -1.07349023e-01 -4.16146889e-02 3.50717679e-02 -1.35197926e+00 2.24183708e-01 9.86347198e-02 -2.66580015e-01 -5.19056559e-01 4.33587581e-01 -6.87118590e-01 5.61283343e-02 6.65252566e-01 -5.79329878e-02 6.00445867e-02 4.31027353e-01 9.78009924e-02 -4.74631697e-01 -4.60551709e-01 1.10715556e+00 2.47284457e-01 -7.33348787e-01 -3.46859008e-01 -9.66843605e-01 -2.18972906e-01 5.51267564e-01 -5.59489012e-01 -6.09338656e-02 -6.68666899e-01 -1.20454597e+00 -2.77469397e-01 -2.21284200e-02 1.11903280e-01 7.73731291e-01 -1.22786140e+00 -9.10777867e-01 3.88347000e-01 -2.27911130e-01 -6.40048742e-01 3.89293194e-01 9.41346407e-01 -7.36880541e-01 1.51630774e-01 -2.81587303e-01 -2.14227140e-01 -1.79211950e+00 9.21279490e-02 4.19847161e-01 3.07919830e-01 -1.53102562e-01 8.85200202e-01 -5.89217126e-01 -2.18101710e-01 4.30169165e-01 -2.92435795e-01 -3.51269931e-01 3.15063410e-02 6.35434151e-01 1.07955873e+00 4.08080339e-01 -6.97738111e-01 -3.80183905e-01 6.73649907e-01 3.49268973e-01 -5.91522932e-01 1.16462028e+00 -4.88695681e-01 -1.90030605e-01 6.48770094e-01 1.06051683e+00 5.84397256e-01 -4.62231964e-01 -8.16943347e-02 4.53192592e-02 -3.54424834e-01 3.73651415e-01 -9.49511230e-01 -7.03211129e-01 7.92597711e-01 1.38157523e+00 6.23650730e-01 1.70052683e+00 -3.52650106e-01 7.36000717e-01 9.76777151e-02 6.61264211e-02 -1.57073653e+00 -2.38446549e-01 5.64067960e-01 9.41430151e-01 -8.69289637e-01 -2.39672482e-01 -5.35383046e-01 -3.23326737e-01 9.54726338e-01 8.74440148e-02 1.92430511e-01 1.16412449e+00 6.10031724e-01 7.02836812e-01 4.24701154e-01 -3.14263552e-01 -5.25482178e-01 7.13931620e-02 9.95264292e-01 8.25850606e-01 1.60768926e-01 -1.04346347e+00 6.68935239e-01 -8.34523916e-01 -1.42597660e-01 5.09799361e-01 9.04104829e-01 -9.54269588e-01 -1.06128323e+00 -8.28725696e-01 3.29178274e-01 -8.80937099e-01 -3.55899602e-01 3.35366756e-04 3.83600473e-01 2.04009950e-01 1.63018703e+00 -1.24907687e-01 -3.79616827e-01 8.75124931e-01 1.93450838e-01 2.62346387e-01 -3.83549720e-01 -4.27486748e-01 7.46458948e-01 2.19247490e-01 -1.05166305e-02 -3.25256258e-01 -9.06566978e-01 -1.24056649e+00 -4.06551689e-01 -3.27592283e-01 4.32615876e-01 9.05475914e-01 6.56034172e-01 -1.90006688e-01 7.38788009e-01 6.54703856e-01 -6.04701757e-01 -3.94532591e-01 -1.32523406e+00 -9.58800554e-01 2.89658811e-02 3.67549837e-01 -2.62957215e-01 -5.42809010e-01 7.38949999e-02]	[15.057908058166504, 5.74766731262207]
a8522f73-213c-4fc7-b49b-454c38d4a078	generalisation-and-sharing-in-triplet	1611.05301	null	http://arxiv.org/abs/1611.05301v1	http://arxiv.org/pdf/1611.05301v1.pdf	Generalisation and Sharing in Triplet Convnets for Sketch based Visual Search	We propose and evaluate several triplet CNN architectures for measuring the similarity between sketches and photographs, within the context of the sketch based image retrieval (SBIR) task. In contrast to recent fine-grained SBIR work, we study the ability of our networks to generalise across diverse object categories from limited training data, and explore in detail strategies for weight sharing, pre-processing, data augmentation and dimensionality reduction. We exceed the performance of pre-existing techniques on both the Flickr15k category level SBIR benchmark by $18\%$, and the TU-Berlin SBIR benchmark by $\sim10 \mathcal{T}_b$, when trained on the 250 category TU-Berlin classification dataset augmented with 25k corresponding photographs harvested from the Internet.	['Leonardo Ribeiro', 'John Collomosse', 'Moacir Ponti', 'Tu Bui']	2016-11-16	null	null	null	null	['sketch-based-image-retrieval']	['computer-vision']	[ 1.97847828e-01 -5.63043237e-01 6.38961568e-02 -5.10942876e-01 -7.05887020e-01 -8.65089715e-01 8.36666822e-01 -5.10494858e-02 -5.69587708e-01 4.17991549e-01 3.35710645e-01 -1.44249186e-01 -8.64785194e-01 -7.46569216e-01 -4.03988123e-01 -1.83915198e-01 -2.80000623e-02 3.36471349e-01 -1.53650627e-01 -2.97508329e-01 3.27112228e-01 9.41337705e-01 -1.67149448e+00 4.52908963e-01 1.36563122e-01 1.82740009e+00 -3.14881146e-01 7.65278578e-01 -1.74645960e-01 4.30378586e-01 -5.22382677e-01 -8.25920701e-01 6.69872999e-01 2.31783226e-01 -7.68532693e-01 -1.34650066e-01 1.65351748e+00 -6.63056731e-01 -6.77282512e-01 7.21229970e-01 5.10448337e-01 5.70570171e-01 8.48250449e-01 -1.45454907e+00 -1.35892868e+00 2.21125469e-01 -1.01998247e-01 3.00925165e-01 8.64023268e-02 -5.07638529e-02 1.21081638e+00 -1.19981992e+00 8.93009067e-01 1.32952416e+00 7.79710591e-01 4.67855513e-01 -1.10760367e+00 -9.27200139e-01 1.67115971e-01 1.29527718e-01 -1.57104146e+00 -4.36138093e-01 8.58119309e-01 -2.50762224e-01 9.90875542e-01 3.10326487e-01 5.70304751e-01 1.08290315e+00 -4.89701182e-01 8.21524024e-01 7.29573965e-01 -1.68932751e-01 -4.14659455e-02 -1.97192863e-01 3.13747168e-01 5.31713486e-01 5.42931445e-03 1.37318328e-01 -6.55549765e-01 -1.47981614e-01 1.16126549e+00 2.52447933e-01 4.46211547e-02 -4.72669274e-01 -1.40430546e+00 6.73896551e-01 7.93595254e-01 5.61585009e-01 -3.08324602e-02 7.76660502e-01 5.77868283e-01 7.65899003e-01 6.05666935e-01 7.22888291e-01 -5.65327346e-01 8.11371282e-02 -9.41077828e-01 5.05501747e-01 4.53147411e-01 1.39815485e+00 6.29246771e-01 1.11044198e-01 -4.15270358e-01 1.17472899e+00 -1.52798474e-01 4.63878393e-01 1.22242846e-01 -1.22252321e+00 4.58875149e-01 5.23305476e-01 -3.18303821e-03 -1.11391985e+00 2.02196836e-03 -1.65118977e-01 -1.04243314e+00 1.99756935e-01 4.01271105e-01 3.53091210e-01 -1.14944732e+00 1.42528915e+00 -2.08626956e-01 2.38127280e-02 -3.57744873e-01 8.96210492e-01 1.25095952e+00 2.72379786e-01 1.95570052e-01 5.71345747e-01 9.94724035e-01 -9.72284317e-01 -2.47537762e-01 1.22612625e-01 1.84066221e-01 -7.98313856e-01 1.28267193e+00 2.76885003e-01 -1.04702473e+00 -8.90110433e-01 -9.64034677e-01 -5.65383792e-01 -9.69109476e-01 3.69941592e-01 7.92779267e-01 3.64221752e-01 -1.51797783e+00 1.04722559e+00 1.01150107e-02 -6.30976498e-01 8.77461851e-01 3.19841504e-01 -6.32650137e-01 -5.28876364e-01 -9.82835412e-01 7.10869968e-01 2.73983572e-02 2.89113801e-02 -7.97761858e-01 -1.10846543e+00 -6.09201849e-01 2.24500179e-01 9.35534835e-02 -4.48172420e-01 8.18988681e-01 -9.40424204e-01 -1.21551001e+00 9.41547811e-01 4.06301677e-01 -1.18792675e-01 4.26935881e-01 -6.24702685e-02 -5.33333182e-01 7.45758861e-02 -3.25428396e-01 1.35445988e+00 1.07987142e+00 -9.21657443e-01 -4.96344984e-01 -2.10088506e-01 5.92201352e-01 1.60021111e-04 -6.84171736e-01 -2.19829995e-02 -5.15887022e-01 -1.05075824e+00 -1.29239947e-01 -8.77453208e-01 6.42842501e-02 9.16860640e-01 9.40907300e-02 -4.86173302e-01 9.55186307e-01 -2.93523192e-01 7.58083582e-01 -2.19340467e+00 1.06267832e-01 2.24432498e-01 3.35524559e-01 4.43066120e-01 -9.71447825e-01 4.90265757e-01 -8.05180371e-02 3.63202304e-01 8.30533206e-02 -3.64182740e-01 4.13516164e-01 6.93553463e-02 -4.17065531e-01 1.61994681e-01 3.01072955e-01 1.27662718e+00 -6.27662778e-01 -2.86237895e-01 4.42017436e-01 6.25062525e-01 -2.92085379e-01 -1.14257364e-02 -9.68865380e-02 -3.85166794e-01 -1.45417795e-01 9.63334680e-01 6.20552540e-01 -1.90783128e-01 -6.63564876e-02 -4.89745706e-01 2.32848495e-01 -9.25902054e-02 -1.05364287e+00 2.18749237e+00 -4.21795189e-01 1.05772734e+00 1.62202135e-01 -8.48327756e-01 1.04244924e+00 1.37066290e-01 3.17529678e-01 -9.05593753e-01 -4.90776636e-02 -7.94805288e-02 -5.69402456e-01 1.84570536e-01 7.61164904e-01 3.92509364e-02 -1.82717741e-01 5.39732993e-01 3.64319921e-01 -3.22215110e-01 1.71815395e-01 2.48527572e-01 9.87886190e-01 1.27492547e-01 -2.49240562e-01 -4.47246879e-01 2.58963734e-01 -7.02067539e-02 -2.31300980e-01 9.60966468e-01 -2.24178538e-01 7.67491460e-01 5.99016175e-02 -1.08043313e+00 -1.30419636e+00 -1.32634592e+00 -1.15709789e-01 1.62112808e+00 1.25834703e-01 -2.45479673e-01 -2.74203181e-01 -7.34500527e-01 4.71000105e-01 2.43779674e-01 -9.09923494e-01 -1.42636181e-05 -3.43236983e-01 -1.37346134e-01 6.56142890e-01 7.01876462e-01 6.73205316e-01 -1.37323916e+00 7.02774292e-03 -1.12335244e-02 2.73084223e-01 -9.54255104e-01 -6.40243351e-01 -3.57360840e-01 -6.37153506e-01 -9.84551728e-01 -8.74209821e-01 -6.82093680e-01 3.16699415e-01 5.95805705e-01 1.63693178e+00 4.85868335e-01 -7.20190883e-01 8.57208431e-01 -3.43225509e-01 -3.18755209e-01 1.79779887e-01 1.29151046e-01 -6.90388605e-02 -2.90609807e-01 4.80853736e-01 -4.99282837e-01 -8.43620598e-01 3.89916629e-01 -1.13492155e+00 -2.21628547e-01 5.78352809e-01 8.28008354e-01 3.98645222e-01 -3.07607919e-01 3.84155065e-01 -2.18561962e-01 7.81389952e-01 -3.66480425e-02 -4.79073584e-01 5.30840039e-01 -4.57033604e-01 -1.03248417e-01 3.13355625e-01 -7.91705072e-01 -5.40732980e-01 -1.64061412e-01 7.05776066e-02 -1.04197514e+00 -2.85945058e-01 1.78529337e-01 2.47386813e-01 -5.31507432e-01 5.75921774e-01 1.53340816e-01 -2.16808274e-01 -4.67379212e-01 8.10976624e-01 5.31303108e-01 4.19379503e-01 -7.20920682e-01 9.45707321e-01 5.25578856e-01 5.98531812e-02 -8.36005807e-01 -5.88436007e-01 -4.25476551e-01 -8.81562054e-01 -1.27861694e-01 7.13241220e-01 -8.56901705e-01 -8.51415873e-01 1.81585491e-01 -9.85133290e-01 -3.23211730e-01 -5.02922058e-01 -3.49501222e-02 -5.41705191e-01 2.21547097e-01 -4.96482939e-01 -4.99980003e-01 -6.99165165e-01 -8.37403715e-01 1.31816256e+00 -2.66819745e-01 -6.03195503e-02 -7.89004266e-01 -1.36032924e-01 1.44927442e-01 9.16987956e-01 7.86823928e-02 9.35729504e-01 -4.16381747e-01 -8.06837678e-01 -5.07287264e-01 -1.01492083e+00 4.37240422e-01 -1.64909307e-02 1.05597405e-02 -9.06033695e-01 -3.40867579e-01 -8.48491430e-01 -7.11961508e-01 1.10900462e+00 -3.82561721e-02 1.36653662e+00 -2.50727445e-01 -3.50893475e-02 6.63658142e-01 1.55237782e+00 -1.33676097e-01 6.27386332e-01 1.42216027e-01 7.17112660e-01 2.37755999e-01 3.55819613e-01 1.47030309e-01 5.81803322e-02 7.88677275e-01 4.54668492e-01 -1.27977595e-01 -5.08073211e-01 -1.80066884e-01 -2.03017071e-01 3.45591813e-01 -5.06866798e-02 -2.48386919e-01 -6.42308235e-01 7.81976700e-01 -1.55669332e+00 -9.56203997e-01 5.13009131e-01 1.86281753e+00 3.13826293e-01 -2.66399443e-01 6.18287213e-02 7.54028326e-03 3.63564700e-01 4.78424847e-01 -3.57795656e-01 -2.69565284e-01 -8.23183209e-02 6.85242295e-01 4.95716482e-01 1.33622259e-01 -1.22124910e+00 1.12278366e+00 7.15819836e+00 1.09099567e+00 -1.02327931e+00 -3.57316248e-02 5.54029346e-01 -5.55048227e-01 -3.06819558e-01 -4.47879225e-01 -3.79775673e-01 1.15088716e-01 5.42401493e-01 2.43918821e-01 9.20963168e-01 8.61346662e-01 -5.56605160e-01 4.00930762e-01 -1.32761168e+00 1.26728642e+00 2.63411462e-01 -1.80883014e+00 4.26765740e-01 -8.80067572e-02 7.11888194e-01 1.10207699e-01 4.53427583e-01 4.56553191e-01 4.46103543e-01 -1.25400805e+00 6.92577899e-01 5.74033737e-01 1.34804213e+00 -5.23005962e-01 3.63701642e-01 -3.17172378e-01 -1.45421207e+00 -3.05937916e-01 -6.75346255e-01 6.76961839e-02 -3.72795939e-01 5.62588423e-02 -5.78257620e-01 4.11300212e-01 1.12292898e+00 9.26210642e-01 -8.54839206e-01 7.22016513e-01 2.87048310e-01 -5.00345230e-02 -2.64296502e-01 -2.21976891e-01 4.37099516e-01 -1.07732058e-01 2.87711740e-01 1.15819764e+00 3.57652724e-01 1.63499758e-01 -1.05937682e-01 7.34548807e-01 -5.05721867e-01 -1.44862473e-01 -6.48336887e-01 -3.25768769e-01 3.11618656e-01 1.43192685e+00 -4.12971109e-01 -5.59018314e-01 -1.22412376e-01 8.96070063e-01 5.12119651e-01 4.84423280e-01 -4.26732659e-01 -4.23101455e-01 1.10987306e+00 -1.49856880e-01 6.75376415e-01 -4.48841333e-01 -1.51533931e-01 -1.04239666e+00 -4.71885614e-02 -6.31647170e-01 4.16548222e-01 -1.27352071e+00 -1.77822495e+00 6.95672750e-01 -2.11215541e-02 -1.15218151e+00 1.49995655e-01 -1.02097559e+00 -3.65377069e-01 7.80413330e-01 -1.56516647e+00 -1.63092852e+00 -5.27106762e-01 7.34328091e-01 5.96344650e-01 -4.07716393e-01 9.05594468e-01 4.04472023e-01 -5.65435439e-02 9.25474346e-01 2.68403500e-01 4.27175701e-01 7.57541597e-01 -1.05979013e+00 8.33285153e-01 1.86102077e-01 4.89965796e-01 5.87029636e-01 9.04781595e-02 -1.86886072e-01 -1.45516193e+00 -1.17192757e+00 5.92396498e-01 -6.87721372e-01 8.02702427e-01 -6.39672160e-01 -5.42884469e-01 3.58034462e-01 1.79578848e-02 4.42259669e-01 4.09044862e-01 2.75380462e-01 -1.07454741e+00 -4.88912642e-01 -1.42109811e+00 5.46187043e-01 1.62894177e+00 -1.01903403e+00 -2.96634376e-01 3.28059435e-01 6.32621944e-01 -3.65061089e-02 -1.44940519e+00 1.50317520e-01 1.22647095e+00 -6.95437551e-01 1.57339275e+00 -1.04228342e+00 6.42047048e-01 1.70726001e-01 -5.86437464e-01 -1.06221795e+00 -7.88035989e-01 -6.41832650e-02 3.93922597e-01 9.99632061e-01 6.85738251e-02 -1.58411302e-02 9.10351038e-01 6.54656947e-01 1.48178846e-01 -5.09345531e-01 -1.00070679e+00 -8.05872977e-01 2.02943772e-01 -4.99805450e-01 9.71342623e-01 8.71834278e-01 -5.92943430e-01 -1.75544638e-02 -2.53681004e-01 -4.17646915e-01 6.42115057e-01 2.44058296e-01 7.47610211e-01 -1.39645827e+00 1.83034599e-01 -8.79220307e-01 -6.24997675e-01 -6.81735933e-01 1.80486709e-01 -7.18723416e-01 -2.70860136e-01 -1.51325905e+00 -4.66007143e-02 -7.54157484e-01 -7.60828555e-01 5.84591448e-01 1.76654309e-01 9.58287299e-01 6.96585357e-01 1.76999807e-01 -8.08076859e-01 3.94570500e-01 1.11403382e+00 -6.61037564e-01 3.29737544e-01 -4.44274276e-01 -5.21822751e-01 1.37682185e-01 2.91986138e-01 -4.26084325e-02 -4.51484919e-01 -8.06982875e-01 2.48529360e-01 -1.95531160e-01 7.73167670e-01 -9.60774124e-01 -8.15425720e-03 -1.43185303e-01 7.24996269e-01 -6.10529542e-01 8.60075533e-01 -1.11600721e+00 1.33405775e-01 -9.78308544e-02 -7.79601693e-01 9.65702236e-02 3.93203676e-01 6.13767326e-01 -4.30522822e-02 1.49889693e-01 5.61506331e-01 -3.79849970e-01 -1.02581394e+00 8.72254193e-01 4.87959981e-02 -3.02921593e-01 5.43149054e-01 -3.10639679e-01 -5.95074832e-01 -4.12760466e-01 -6.20829165e-01 -8.20850357e-02 4.52998281e-01 8.93369555e-01 7.85996079e-01 -1.84037220e+00 -4.94596273e-01 2.36139685e-01 4.81355846e-01 -4.50619578e-01 4.33623493e-01 1.44084617e-01 -3.30257654e-01 6.95995808e-01 -6.45967245e-01 -2.93582827e-01 -1.16213059e+00 7.78014719e-01 2.06313699e-01 -1.03099560e-02 -2.77979493e-01 1.10446656e+00 1.80525593e-02 -4.99715000e-01 3.41172189e-01 -2.52980739e-01 -9.97605454e-03 1.07907608e-01 4.77388322e-01 5.12319028e-01 2.43399799e-01 -4.37078655e-01 -3.67690057e-01 6.52361214e-01 -3.26707773e-02 -8.45817477e-02 1.56277013e+00 1.56459138e-01 -9.18439776e-02 7.30333105e-02 1.61941266e+00 -7.28832841e-01 -1.12048125e+00 -3.60173941e-01 -1.18917152e-01 -7.04718888e-01 7.34363571e-02 -1.10223913e+00 -1.22800207e+00 8.73132050e-01 9.06645477e-01 2.40260020e-01 1.03850102e+00 -1.14047982e-01 6.04851186e-01 8.56512725e-01 3.90919447e-01 -1.07782471e+00 2.09552720e-01 4.75320160e-01 1.43323684e+00 -1.22858894e+00 2.26104394e-01 1.13242194e-01 -3.84809136e-01 1.09681165e+00 4.39287990e-01 -7.06746399e-01 8.52637947e-01 -2.16888905e-01 8.16053525e-02 -3.33788335e-01 -6.78578079e-01 -6.60969317e-02 8.33433151e-01 5.75034261e-01 1.41676366e-01 -6.36113435e-02 1.98043704e-01 3.20766233e-02 9.52236801e-02 2.91326106e-01 2.53939186e-04 8.50212693e-01 -1.09639473e-01 -9.77410436e-01 -3.72724570e-02 8.21585059e-01 -1.15041696e-01 -3.68487597e-01 -8.50221276e-01 9.47470486e-01 5.47219217e-02 5.26140988e-01 5.19910991e-01 -3.43885988e-01 4.50148970e-01 6.88632429e-02 5.80416143e-01 -1.36297002e-01 -8.52832556e-01 -4.29692537e-01 1.62854612e-01 -6.67322934e-01 -4.75300521e-01 -3.24756503e-01 -1.53430119e-01 -4.90979642e-01 -2.94977687e-02 -3.82785439e-01 8.18278253e-01 6.92619205e-01 5.56717873e-01 3.65159400e-02 3.35833490e-01 -1.21711099e+00 -3.91051471e-01 -9.95359600e-01 -5.83659291e-01 6.76020443e-01 3.23080748e-01 -7.47440875e-01 -2.61934519e-01 -3.24910939e-01]	[11.62424087524414, 0.5299427509307861]
612a8cad-c197-4698-9802-29fd9e6aa6e5	synthesize-extremely-high-dimensional	2304.02169	null	https://arxiv.org/abs/2304.02169v1	https://arxiv.org/pdf/2304.02169v1.pdf	Synthesize Extremely High-dimensional Longitudinal Electronic Health Records via Hierarchical Autoregressive Language Model	Synthetic electronic health records (EHRs) that are both realistic and preserve privacy can serve as an alternative to real EHRs for machine learning (ML) modeling and statistical analysis. However, generating high-fidelity and granular electronic health record (EHR) data in its original, highly-dimensional form poses challenges for existing methods due to the complexities inherent in high-dimensional data. In this paper, we propose Hierarchical Autoregressive Language mOdel (HALO) for generating longitudinal high-dimensional EHR, which preserve the statistical properties of real EHR and can be used to train accurate ML models without privacy concerns. Our HALO method, designed as a hierarchical autoregressive model, generates a probability density function of medical codes, clinical visits, and patient records, allowing for the generation of realistic EHR data in its original, unaggregated form without the need for variable selection or aggregation. Additionally, our model also produces high-quality continuous variables in a longitudinal and probabilistic manner. We conducted extensive experiments and demonstrate that HALO can generate high-fidelity EHR data with high-dimensional disease code probabilities (d > 10,000), disease co-occurrence probabilities within visits (d > 1,000,000), and conditional probabilities across consecutive visits (d > 5,000,000) and achieve above 0.9 R2 correlation in comparison to real EHR data. This performance then enables downstream ML models trained on its synthetic data to achieve comparable accuracy to models trained on real data (0.938 AUROC with HALO data vs. 0.943 with real data). Finally, using a combination of real and synthetic data enhances the accuracy of ML models beyond that achieved by using only real EHR data.	['Jimeng Sun', 'Cao Xiao', 'Brandon Theodorou']	2023-04-04	null	null	null	null	['variable-selection']	['methodology']	[-8.98079425e-02 4.43535268e-01 7.93018192e-02 -5.68742394e-01 -1.34570014e+00 -3.58634621e-01 1.12008214e-01 6.56905055e-01 -1.80552423e-01 1.02724755e+00 4.87947732e-01 -5.11214674e-01 -9.41149816e-02 -8.69010270e-01 -8.80313873e-01 -3.41999143e-01 -4.37208354e-01 5.71154952e-01 -6.43154442e-01 4.34231877e-01 -7.51711607e-01 2.86193728e-01 -9.94083285e-01 3.05652320e-01 8.89497757e-01 6.41992807e-01 -2.86268413e-01 7.34929800e-01 2.61586815e-01 6.13253355e-01 -5.12344837e-01 -4.63283211e-01 2.94254422e-01 -1.74758479e-01 -2.01697990e-01 -2.49279320e-01 3.03064715e-02 -4.70084548e-01 -3.94496590e-01 5.78373730e-01 5.63375354e-01 -3.98874462e-01 6.45789623e-01 -1.01375604e+00 -7.25290716e-01 5.56088150e-01 -3.12391758e-01 -6.37375534e-01 6.27418995e-01 2.64346600e-01 6.99977636e-01 -4.68487531e-01 5.63309968e-01 1.02052176e+00 1.22467184e+00 4.83568013e-01 -1.79431534e+00 -5.47547162e-01 -3.07135373e-01 -7.94951141e-01 -1.66650641e+00 -2.40806609e-01 1.63275957e-01 -6.68741465e-01 5.78458071e-01 5.02630174e-01 7.82501400e-01 1.28613043e+00 5.55561483e-01 6.28237724e-01 9.40023661e-01 -2.53589284e-02 3.76180559e-01 3.16932172e-01 2.04294547e-01 4.13146734e-01 5.49305141e-01 3.44236016e-01 1.15353577e-02 -1.12637627e+00 7.17657804e-01 5.71706891e-01 -1.80525333e-01 -2.43800685e-01 -1.27883089e+00 6.75821424e-01 1.58541694e-01 -1.77058905e-01 -9.50556397e-01 4.27475572e-02 3.09120595e-01 2.02217400e-02 1.12624355e-01 3.61423552e-01 -5.55554271e-01 -2.18180656e-01 -8.63828838e-01 4.64321017e-01 7.44187057e-01 1.34913337e+00 2.09073767e-01 -1.03622582e-02 -6.76800787e-01 5.61386466e-01 2.89451450e-01 8.95490110e-01 2.28594229e-01 -6.42480254e-01 3.70814860e-01 6.64577663e-01 3.61145318e-01 -8.95731211e-01 -4.77924913e-01 -2.75098085e-01 -1.39659631e+00 -5.13662875e-01 1.39203399e-01 -5.00340819e-01 -8.16457629e-01 1.86276245e+00 3.04515988e-01 8.36803243e-02 4.28363144e-01 1.60145983e-01 7.07041383e-01 6.03817999e-01 4.75167722e-01 -2.90938348e-01 1.47153628e+00 8.10209662e-02 -8.72862697e-01 4.51482832e-01 9.62247491e-01 -3.75669181e-01 1.16732991e+00 1.00976862e-01 -1.09286833e+00 -2.65263259e-01 -3.99989277e-01 8.53440538e-02 -5.07789664e-02 9.92099345e-02 6.03367805e-01 6.58492327e-01 -8.94821286e-01 1.66047245e-01 -1.04622972e+00 -1.33748919e-01 6.90604448e-01 3.00076962e-01 -4.85798120e-01 -2.93453068e-01 -1.27382505e+00 1.09075718e-01 1.14398502e-01 -2.43069276e-01 -6.85444415e-01 -1.24440479e+00 -1.15798402e+00 2.40237787e-01 -2.59905607e-01 -1.06203139e+00 9.95332181e-01 -2.06487975e-03 -7.55550683e-01 4.74526852e-01 -2.75994897e-01 -6.63936436e-01 6.13023162e-01 -9.59741473e-02 -5.64062178e-01 -2.48750716e-01 4.27199341e-03 2.64963061e-01 2.24765733e-01 -1.24489629e+00 -3.04569930e-01 -4.76906270e-01 -8.12386394e-01 -4.24197555e-01 -1.68559074e-01 -1.84783593e-01 -2.32780054e-01 -6.89611554e-01 -1.53046221e-01 -9.59635019e-01 -6.32385254e-01 -1.25935093e-01 -5.67211092e-01 2.43375316e-01 2.59101808e-01 -1.04280519e+00 1.59164500e+00 -2.22760630e+00 -5.36358595e-01 3.66430670e-01 4.57745194e-01 1.68927386e-01 6.57971054e-02 4.89529669e-01 3.87437642e-02 3.66447777e-01 -5.00356317e-01 -2.96514690e-01 -1.55843660e-01 3.09051536e-02 -1.38086691e-01 2.12263882e-01 2.81307667e-01 1.20967197e+00 -8.37417722e-01 -4.70474660e-01 -2.92833429e-02 8.01049411e-01 -8.34499359e-01 5.96569538e-01 6.00494742e-02 3.80509585e-01 -5.25388777e-01 6.36685371e-01 7.21681058e-01 -5.93618631e-01 4.27471399e-01 -1.97833050e-02 3.71269166e-01 -1.15736715e-01 -9.69515622e-01 1.09552121e+00 -3.60921204e-01 8.49986151e-02 -1.63767070e-01 -3.07128549e-01 1.15288770e+00 6.37054741e-01 9.27825212e-01 -3.04161608e-01 -2.09154323e-01 -1.81564987e-01 -2.54409105e-01 -4.64405209e-01 2.67834187e-01 -1.52319625e-01 -5.07037103e-01 2.82599300e-01 -5.09715319e-01 1.62378326e-01 -3.63382429e-01 1.38860524e-01 1.37878323e+00 -3.43859732e-01 3.86799514e-01 9.65700671e-02 -1.40178069e-01 -7.97713399e-02 8.11760187e-01 8.80101323e-01 1.13835536e-01 7.85408437e-01 5.98628283e-01 -4.92977530e-01 -1.25000131e+00 -1.39683533e+00 -6.39915705e-01 2.09131911e-02 -6.21278644e-01 -6.23585880e-01 -5.28636754e-01 -5.44964373e-01 4.45292413e-01 8.73113930e-01 -5.57169080e-01 -2.05217332e-01 -2.57186830e-01 -1.17441630e+00 9.21966910e-01 5.02303600e-01 1.73788145e-01 -6.57558978e-01 -5.59274912e-01 4.93980467e-01 -2.73637265e-01 -1.06176996e+00 -3.74824435e-01 -2.04007298e-01 -9.03668642e-01 -1.03878641e+00 -5.52626193e-01 -2.77131468e-01 7.57189333e-01 -5.10301948e-01 1.19226468e+00 -3.25758040e-01 -4.43472087e-01 3.30074936e-01 -4.40098420e-02 -5.83968937e-01 -6.85219228e-01 -2.57426649e-01 5.06187640e-02 7.35745654e-02 5.96289754e-01 -3.31062168e-01 -6.20105267e-01 -6.93528280e-02 -1.14429593e+00 -8.75296816e-02 5.03098011e-01 1.05874455e+00 9.19138014e-01 -2.88255423e-01 8.68057787e-01 -1.31916475e+00 7.28295207e-01 -8.36784601e-01 -6.28601551e-01 3.27926695e-01 -8.39513183e-01 1.31910682e-01 7.73922920e-01 -3.64671350e-01 -8.57137978e-01 3.39155108e-01 -2.12560982e-01 -2.97508776e-01 -2.11429641e-01 6.20554447e-01 -5.74480109e-02 6.46549642e-01 6.92908347e-01 2.40765065e-01 2.29682833e-01 -5.92015922e-01 1.19548269e-01 1.13606656e+00 2.85144836e-01 -3.88261050e-01 3.18908840e-01 4.23777729e-01 2.86900578e-03 -5.22246301e-01 -2.91613936e-01 -1.40905306e-01 -3.70573014e-01 6.17129326e-01 7.50363708e-01 -1.33661938e+00 -1.07794917e+00 3.00365210e-01 -6.51296318e-01 -1.52784854e-01 -5.45039892e-01 7.89060354e-01 -3.97986948e-01 1.36334643e-01 -7.77368963e-01 -1.23559833e+00 -5.83334446e-01 -7.93795049e-01 1.35977590e+00 -2.74294972e-01 -7.03498602e-01 -9.50835824e-01 7.78247342e-02 8.79034400e-02 3.88935506e-01 9.70709980e-01 1.22789633e+00 -7.36638367e-01 -4.27614570e-01 -7.89464831e-01 -1.56646654e-01 6.78833574e-02 4.41989481e-01 -2.93413317e-03 -6.20253265e-01 -4.28272665e-01 -6.05035685e-02 -1.58920318e-01 1.84136897e-01 7.41193891e-01 1.42266643e+00 -7.43083298e-01 -4.83984321e-01 7.71880746e-01 1.32448018e+00 2.02598438e-01 6.63171411e-01 -3.80376428e-01 5.82107425e-01 4.17489409e-01 5.12583733e-01 9.80531931e-01 7.37374604e-01 5.04627049e-01 -4.14370969e-02 -6.27926171e-01 4.62068409e-01 -8.33565593e-01 -6.10603113e-03 5.55855095e-01 5.13864219e-01 5.89903705e-02 -1.19987535e+00 6.67510152e-01 -1.73711026e+00 -7.21153319e-01 -4.35125411e-01 2.79932380e+00 1.12464809e+00 -2.96582878e-01 2.31538177e-01 -3.55725378e-01 4.83903736e-01 -4.36976641e-01 -6.91892564e-01 -2.54790187e-01 -1.89695686e-01 3.67613472e-02 6.37426436e-01 2.70273358e-01 -8.87130618e-01 3.17308605e-01 6.98757935e+00 1.61255181e-01 -7.00213552e-01 -1.56098455e-01 1.09729218e+00 -2.14620009e-01 -6.41520917e-01 -2.76797175e-01 -7.75961637e-01 6.89689636e-01 1.52576256e+00 -2.92102426e-01 2.31958572e-02 9.10657287e-01 6.13846004e-01 3.93510073e-01 -1.19360900e+00 1.18654895e+00 -3.90425563e-01 -1.37112558e+00 1.41318738e-01 5.04598260e-01 8.13798189e-01 -1.37542665e-01 2.60973908e-02 3.95249486e-01 9.29967403e-01 -1.37699389e+00 -1.73345506e-02 9.12406623e-01 1.29197752e+00 -6.26117229e-01 9.93026853e-01 4.15343106e-01 -8.93231988e-01 -7.26877823e-02 -1.58518985e-01 4.48992074e-01 4.75779921e-01 9.89800632e-01 -1.29691505e+00 6.49824202e-01 6.71340227e-01 5.53211570e-01 -3.69429559e-01 8.40404809e-01 5.88137090e-01 7.99323618e-01 -4.58904862e-01 2.09357038e-01 -2.19805166e-01 -2.89232493e-03 5.74979112e-02 1.16733432e+00 6.03372037e-01 2.44704798e-01 1.13073960e-01 9.14145410e-01 -1.20170176e-01 1.77606001e-01 -9.26228821e-01 -1.27162978e-01 6.74360871e-01 7.84824193e-01 3.86353612e-01 -4.92836148e-01 -2.94537365e-01 4.29312617e-01 1.58435434e-01 5.46843052e-01 -8.36663306e-01 -1.49179846e-01 7.34674811e-01 5.67484498e-01 -2.35087380e-01 4.07990851e-02 -4.00518298e-01 -1.12037301e+00 9.72569287e-02 -1.15879762e+00 7.27346003e-01 -4.53629524e-01 -1.54496014e+00 7.67875195e-01 -9.72160101e-02 -1.29513025e+00 -6.95427597e-01 4.14660126e-02 5.59934564e-02 1.17226183e+00 -1.11350107e+00 -1.11486757e+00 -2.30982810e-01 5.80578864e-01 -1.79012299e-01 -2.04365812e-02 1.33933520e+00 3.45783114e-01 -4.00466591e-01 1.06143010e+00 5.12000501e-01 3.09191078e-01 5.18010020e-01 -1.06550658e+00 5.64942241e-01 3.05212796e-01 -3.03623050e-01 1.04594862e+00 3.26321185e-01 -1.00498056e+00 -1.41800988e+00 -1.70256341e+00 1.07552707e+00 -7.47089922e-01 3.70087735e-02 -4.39154446e-01 -1.20079589e+00 9.99885142e-01 -6.31348014e-01 2.49983184e-02 1.22717273e+00 9.91889462e-02 -1.18069008e-01 -1.93844154e-01 -1.78922844e+00 5.42565405e-01 6.02271438e-01 -3.54549766e-01 -1.79223061e-01 3.75323236e-01 9.51219618e-01 -2.97528416e-01 -1.75095224e+00 7.05334604e-01 6.69810176e-01 -4.12313372e-01 9.40157175e-01 -1.07004118e+00 2.02454790e-01 -1.23929784e-01 -1.85613394e-01 -9.46925879e-01 -2.60816991e-01 -8.42478216e-01 -2.60258198e-01 1.11640954e+00 6.41158044e-01 -8.15884173e-01 5.39754987e-01 1.45146120e+00 4.05716747e-01 -8.90761137e-01 -6.95291460e-01 -5.41279554e-01 -1.37921810e-01 -3.32951456e-01 1.16941953e+00 1.00853491e+00 -1.35006264e-01 -1.71647817e-01 -6.44564211e-01 3.87604207e-01 7.31252253e-01 -2.40827762e-02 9.34879661e-01 -1.23475814e+00 -4.10243481e-01 3.87769967e-01 -3.06129128e-01 -7.28249609e-01 -2.79775023e-01 -7.81893969e-01 -3.80504906e-01 -1.50837409e+00 2.83855855e-01 -9.24553335e-01 -3.00331116e-01 5.22662997e-01 -3.90037209e-01 -2.59843737e-01 -1.05893612e-01 2.11053684e-01 -3.56422807e-03 6.20284855e-01 8.29775453e-01 2.26313770e-01 -4.92681831e-01 2.52105445e-01 -8.33764970e-01 2.10934222e-01 6.32932067e-01 -6.62121296e-01 -3.02826732e-01 -5.95832355e-02 -1.31181404e-01 6.43092692e-01 2.86351323e-01 -6.44701242e-01 -1.75053284e-01 -6.76504597e-02 6.53191924e-01 -4.76425797e-01 2.46726587e-01 -9.69406605e-01 1.06720448e+00 6.72297716e-01 -5.74739456e-01 3.15080762e-01 2.53062516e-01 6.76816046e-01 -8.40148255e-02 4.47784334e-01 4.40484345e-01 -1.17391698e-01 4.27612841e-01 5.20922065e-01 -2.00571314e-01 -8.65653306e-02 1.07680500e+00 -2.54590977e-02 1.29485339e-01 -4.74365324e-01 -8.73709500e-01 3.92384827e-01 4.74783182e-01 2.04356015e-01 7.92599559e-01 -1.31304753e+00 -1.06185150e+00 6.40882432e-01 3.66005629e-01 2.86560297e-01 4.05046999e-01 6.08483553e-01 -3.65597546e-01 4.08621252e-01 1.69106632e-01 -7.11157739e-01 -1.20181584e+00 7.56255805e-01 2.80377595e-03 -5.05609393e-01 -8.12360048e-01 7.70466030e-02 2.01057121e-01 -8.04920256e-01 1.15800105e-01 -5.90901613e-01 2.96878219e-01 -3.86413306e-01 6.25272334e-01 5.39572574e-02 -9.20245126e-02 -2.39659414e-01 -1.48613691e-01 1.38233110e-01 2.34733876e-02 1.47756293e-01 1.48135662e+00 -6.45758137e-02 1.78988695e-01 4.69373703e-01 1.26318383e+00 1.41653568e-01 -1.21221352e+00 -6.42333180e-02 -1.11479864e-01 -5.73226035e-01 -4.50579762e-01 -7.13240206e-01 -6.36677802e-01 5.44278324e-01 6.32866025e-01 1.92018270e-01 9.60495114e-01 -1.02605186e-01 8.23787868e-01 4.53449553e-03 3.37901115e-01 -4.47604030e-01 -6.47369087e-01 -1.79518566e-01 5.79682231e-01 -1.05842197e+00 -2.19574332e-01 -2.59420753e-01 -9.73955870e-01 4.27787364e-01 1.46403953e-01 2.58859456e-01 8.00850213e-01 4.72291857e-01 1.45117968e-01 -9.94621962e-02 -1.01380944e+00 4.96557653e-01 3.47711109e-02 8.71997178e-01 6.23237908e-01 6.10041857e-01 -1.59353644e-01 8.97149980e-01 -2.31699839e-01 4.47037458e-01 4.65975255e-01 7.30390191e-01 3.87224376e-01 -1.07577538e+00 -3.47395599e-01 1.09266043e+00 -6.01524591e-01 -2.89098710e-01 3.21568921e-02 5.87195337e-01 -3.14204603e-01 7.69452274e-01 5.94109632e-02 -1.94651857e-01 6.07291043e-01 2.06255078e-01 -3.15638512e-01 -4.80408669e-01 -7.60948777e-01 -1.66407321e-02 9.36891511e-02 -4.78567451e-01 4.28253680e-01 -8.62774551e-01 -1.33072436e+00 -3.41115028e-01 1.15716100e-01 2.35809058e-01 5.59105694e-01 2.28941381e-01 1.24873626e+00 4.84802604e-01 6.89472616e-01 1.34552419e-01 -9.61942434e-01 -7.05733597e-01 -6.30652606e-01 9.46427584e-01 4.89594728e-01 7.29185790e-02 1.30018117e-02 2.40104899e-01]	[6.415262699127197, 6.680747985839844]
01f6abe7-d6d1-4bd0-8851-1cbfbe48454a	the-runner-up-solution-for-youtube-vis-long	2211.09973	null	https://arxiv.org/abs/2211.09973v1	https://arxiv.org/pdf/2211.09973v1.pdf	The Runner-up Solution for YouTube-VIS Long Video Challenge 2022	This technical report describes our 2nd-place solution for the ECCV 2022 YouTube-VIS Long Video Challenge. We adopt the previously proposed online video instance segmentation method IDOL for this challenge. In addition, we use pseudo labels to further help contrastive learning, so as to obtain more temporally consistent instance embedding to improve tracking performance between frames. The proposed method obtains 40.2 AP on the YouTube-VIS 2022 long video dataset and was ranked second place in this challenge. We hope our simple and effective method could benefit further research.	['Song Bai', 'Xiang Bai', 'Qihao Liu', 'Yi Jiang', 'Junfeng Wu']	2022-11-18	null	null	null	null	['video-instance-segmentation']	['computer-vision']	[-4.01785374e-01 -2.55342185e-01 -7.26709068e-01 -1.39101878e-01 -9.04145896e-01 -7.66570628e-01 1.85079336e-01 -3.43620539e-01 -6.35813057e-01 7.92158186e-01 -1.05234999e-02 -2.58649141e-01 2.51201928e-01 -1.95118353e-01 -6.64510250e-01 -3.20530355e-01 -3.81234884e-01 -1.98573604e-01 8.32392156e-01 1.96010917e-01 7.65655190e-02 1.01095185e-01 -1.09780300e+00 3.59441698e-01 4.25623953e-01 1.25694573e+00 3.31961960e-01 7.43125796e-01 6.35248497e-02 1.09215260e+00 -3.03819716e-01 -5.26359737e-01 6.88749611e-01 -8.21544454e-02 -1.00846159e+00 6.80032521e-02 1.06032634e+00 -7.51303673e-01 -8.95132899e-01 8.47084343e-01 5.40767789e-01 2.19985202e-01 2.72952378e-01 -1.40358388e+00 -1.78480953e-01 5.38406670e-01 -7.65753329e-01 8.66366923e-01 3.51204455e-01 3.25824648e-01 9.18722153e-01 -8.63817990e-01 1.08722425e+00 9.43289459e-01 5.46227813e-01 7.76996434e-01 -5.38613260e-01 -8.71324539e-01 4.00938541e-01 8.37848008e-01 -1.46750081e+00 -5.93390346e-01 5.36590815e-01 -1.96275488e-01 5.20040631e-01 1.51698858e-01 8.41372907e-01 1.02401805e+00 -1.27404720e-01 1.32916701e+00 8.89258504e-01 2.37957507e-01 2.28702240e-02 -2.18869373e-01 8.85536298e-02 8.27401280e-01 -1.50473565e-02 5.10701314e-02 -4.99570101e-01 2.42895812e-01 6.05863571e-01 7.42066726e-02 -6.57794356e-01 5.63595444e-03 -1.27256608e+00 4.03898865e-01 5.23315966e-01 4.12952416e-02 -1.42177805e-01 6.41482890e-01 7.10823655e-01 3.80486012e-01 4.23931867e-01 4.95210290e-02 -4.10520285e-01 -6.34633422e-01 -1.39692461e+00 7.51391500e-02 4.95637625e-01 1.25803375e+00 2.53599733e-01 -4.61024567e-02 -3.43845993e-01 6.28242075e-01 4.22856182e-01 1.69486448e-01 1.25227258e-01 -1.73478949e+00 6.15775228e-01 -1.62015170e-01 1.27996862e-01 -9.23539579e-01 1.17956297e-02 -6.31343126e-02 -2.69730359e-01 -1.02867879e-01 4.18421507e-01 -1.77759841e-01 -8.33862007e-01 1.24853814e+00 1.91568464e-01 1.14066601e+00 -1.86484590e-01 1.12910390e+00 1.08340394e+00 9.16345775e-01 7.92432278e-02 -3.53970259e-01 1.20448589e+00 -1.61685121e+00 -7.06390798e-01 9.81925353e-02 6.06781662e-01 -4.88590986e-01 6.87173307e-01 3.35545599e-01 -1.32574666e+00 -5.87986350e-01 -1.10760939e+00 -5.67862466e-02 3.37221734e-02 -1.09118111e-01 4.56698328e-01 4.44861501e-01 -1.15962338e+00 5.24679363e-01 -1.06200171e+00 -4.34893072e-02 8.38142693e-01 2.34330922e-01 -2.24205047e-01 -4.31432158e-01 -1.11821163e+00 2.93201119e-01 3.94712716e-01 -7.32424408e-02 -8.95419657e-01 -6.72330081e-01 -8.06508720e-01 -1.93107218e-01 7.27433026e-01 -3.38489823e-02 1.14116645e+00 -5.25231957e-01 -1.20334649e+00 8.58805716e-01 -2.96195716e-01 -8.49496722e-01 7.67304540e-01 -2.41237253e-01 -4.10300255e-01 8.01661909e-01 1.50869926e-02 1.26158011e+00 6.63823664e-01 -1.09474266e+00 -9.71577764e-01 6.28756359e-02 3.80321562e-01 5.55710457e-02 -4.29712474e-01 -8.91006142e-02 -1.69624996e+00 -7.11969793e-01 -7.31778517e-02 -1.20273244e+00 1.21566363e-01 3.02935451e-01 -1.83686599e-01 -3.61414135e-01 1.29811728e+00 -8.37235868e-01 1.44768715e+00 -2.35005260e+00 -1.51012972e-01 -6.53663427e-02 6.22031629e-01 3.35265636e-01 -3.07924062e-01 -1.21735848e-01 2.44629249e-01 3.04935247e-01 2.55425990e-01 -3.48112673e-01 -2.19147444e-01 -5.89672662e-02 -1.06018841e-01 4.46221620e-01 -1.42338961e-01 1.05863881e+00 -1.11901724e+00 -8.11624825e-01 2.23011732e-01 3.35380584e-01 -4.81471062e-01 -3.05336323e-02 -1.85790017e-01 1.78514913e-01 -3.27779055e-01 6.40118599e-01 8.75322461e-01 -3.88218254e-01 6.74068853e-02 -7.73652494e-01 2.67145894e-02 -7.94069842e-02 -8.35999787e-01 1.81561065e+00 -1.19168147e-01 1.34763181e+00 -1.74547300e-01 -6.89372897e-01 2.37839594e-01 3.58223647e-01 9.09821033e-01 -8.33911240e-01 1.54958010e-01 1.29487395e-01 -3.02338630e-01 -6.09329581e-01 6.85111940e-01 6.34423077e-01 2.17409685e-01 6.87188748e-03 -1.13928989e-01 2.21540138e-01 6.97205424e-01 6.50336683e-01 1.13048089e+00 3.99048656e-01 -1.05556466e-01 -7.49233142e-02 3.77360433e-01 -1.29284054e-01 8.26370537e-01 4.89647150e-01 -1.09102547e+00 9.27352250e-01 3.92224371e-01 -4.88451004e-01 -7.65686572e-01 -8.84501338e-01 -1.24557994e-01 6.12721562e-01 6.49291813e-01 -8.27432811e-01 -6.60610557e-01 -1.36359191e+00 -2.31894568e-01 2.90411532e-01 -2.67469555e-01 1.82060555e-01 -7.96732068e-01 1.77660375e-03 6.67897224e-01 6.04881287e-01 8.13213766e-01 -7.69682050e-01 -2.73446083e-01 2.58101016e-01 -5.36804676e-01 -1.68348897e+00 -1.11380994e+00 -4.07322466e-01 -7.45490611e-01 -1.41689181e+00 -1.19574964e+00 -9.72383618e-01 3.34208876e-01 6.34774268e-01 9.34992015e-01 2.91551858e-01 -1.42390326e-01 5.53226352e-01 -6.49765134e-01 3.25039923e-01 1.42722294e-01 1.00213774e-01 -1.06943503e-01 -2.76344031e-01 2.48932362e-01 6.86265603e-02 -9.13208544e-01 6.40678704e-01 -6.91729128e-01 -7.09644482e-02 -2.18851477e-01 6.50795519e-01 7.50313103e-01 -1.09606288e-01 3.05692703e-01 -6.37289822e-01 1.25157416e-01 -5.10113120e-01 -5.86056828e-01 2.75879174e-01 -5.29818118e-01 -3.12681615e-01 4.41874981e-01 -3.71117800e-01 -6.24741197e-01 -5.40820174e-02 -2.57752955e-01 -9.75865722e-01 2.76234895e-01 2.30016485e-01 1.96104258e-01 -4.20599252e-01 4.18655463e-02 1.73767716e-01 -1.75961733e-01 -1.05489314e-01 2.33273476e-01 6.82756960e-01 6.01974845e-01 -4.22986358e-01 6.24699473e-01 3.96063149e-01 -2.52749413e-01 -6.48161888e-01 -7.35956848e-01 -5.64884722e-01 -1.66655138e-01 -4.63697582e-01 1.14584959e+00 -1.35895789e+00 -8.77689898e-01 4.06600744e-01 -1.15679240e+00 -6.22820854e-01 7.30048642e-02 4.77471620e-01 -4.93977875e-01 6.82958722e-01 -1.05214238e+00 -3.63898009e-01 -2.12138996e-01 -1.35765231e+00 8.63311887e-01 3.65389168e-01 1.70163348e-01 -9.54003274e-01 -8.41944218e-02 5.70932269e-01 2.16515541e-01 -1.73473023e-02 -1.16388179e-01 -1.94249794e-01 -9.16193187e-01 8.47571250e-03 -6.46028519e-01 4.75510329e-01 -7.50512406e-02 2.30169073e-01 -5.78075886e-01 -7.55461395e-01 -3.61311108e-01 -4.33140129e-01 1.10633981e+00 5.20199597e-01 1.80811226e+00 6.80198614e-03 -4.20928210e-01 9.86887753e-01 1.43360162e+00 4.69120711e-01 8.24555933e-01 3.61334085e-01 8.85166168e-01 6.56932369e-02 1.11116850e+00 1.31568745e-01 6.46207809e-01 1.05158961e+00 3.03872526e-01 1.89891718e-02 -4.02728617e-01 -3.90789174e-02 4.42074209e-01 9.37467217e-01 -3.07592213e-01 -5.89568973e-01 -6.26413882e-01 5.82060516e-01 -1.85822642e+00 -1.23280072e+00 -7.15080053e-02 1.84590280e+00 5.53837895e-01 2.72712588e-01 2.47911453e-01 -1.60599038e-01 6.11647904e-01 6.48689389e-01 -2.89305955e-01 2.23141924e-01 -5.81605136e-02 -4.69557308e-02 7.51315236e-01 4.35660839e-01 -1.23716092e+00 1.06212449e+00 6.56611681e+00 1.19040620e+00 -1.18261683e+00 3.42020363e-01 7.95808613e-01 -4.62709278e-01 -5.99619783e-02 -1.61577672e-01 -9.00332987e-01 1.06109607e+00 9.32216763e-01 -1.81942478e-01 1.99266091e-01 7.07533300e-01 3.24968159e-01 5.43822087e-02 -8.60386014e-01 1.42985666e+00 -2.93034445e-02 -1.94278789e+00 -1.33434474e-01 4.77760658e-02 7.98382759e-01 3.13051373e-01 4.57528941e-02 3.71623367e-01 -3.57147962e-01 -5.39199948e-01 6.39667809e-01 2.22512051e-01 1.13716745e+00 -6.68642163e-01 6.65552974e-01 -3.00560743e-01 -1.69824672e+00 1.02093458e-01 -1.51741832e-01 5.28433263e-01 6.00729525e-01 9.53382626e-02 -2.62579709e-01 3.11783999e-01 8.67938697e-01 1.37560713e+00 -5.52582026e-01 1.64520204e+00 5.66979572e-02 8.47111762e-01 -2.57208437e-01 1.88537732e-01 4.74081993e-01 1.93708509e-01 5.49036384e-01 1.21921253e+00 2.82169998e-01 2.55346984e-01 5.32886147e-01 1.27247155e-01 -5.52596331e-01 -2.12766334e-01 -3.77228588e-01 -1.02653310e-01 6.69321954e-01 1.10868573e+00 -8.37297320e-01 -5.87305129e-01 -4.84151900e-01 1.36929071e+00 -7.26676509e-02 5.06205440e-01 -1.47520244e+00 -1.92178175e-01 6.09392464e-01 3.83433439e-02 6.20129704e-01 -4.74399775e-01 3.20798457e-01 -1.50251603e+00 -4.55505587e-02 -6.68610096e-01 3.98691565e-01 -8.34589958e-01 -1.00579727e+00 6.72750413e-01 -3.46270576e-02 -1.48996150e+00 6.29445240e-02 -5.44630945e-01 -4.57714081e-01 1.96180530e-02 -1.63874316e+00 -6.50217533e-01 -6.34304345e-01 7.81596065e-01 1.08693802e+00 -1.20531917e-01 1.62683725e-01 9.27610397e-01 -7.63437927e-01 1.03653109e+00 9.46175829e-02 2.79483557e-01 8.70219350e-01 -9.79176581e-01 4.45980340e-01 7.69598067e-01 2.53991991e-01 2.24635482e-01 3.76543730e-01 -7.09349096e-01 -1.31193292e+00 -1.23516273e+00 2.67772585e-01 -2.66972780e-01 5.44277668e-01 7.44808391e-02 -6.06107056e-01 6.20066524e-01 4.33175981e-01 6.38717532e-01 3.97693634e-01 -5.72487056e-01 -2.48895764e-01 -1.25885859e-01 -1.08301008e+00 6.20911896e-01 1.19932365e+00 -4.28870261e-01 -5.84667549e-02 2.27709681e-01 1.04161859e+00 -6.96953177e-01 -9.18394387e-01 3.26025903e-01 4.34935153e-01 -6.70905650e-01 9.35875356e-01 -2.88206130e-01 4.69054878e-01 -4.69096988e-01 -1.83189496e-01 -7.98245549e-01 -6.65000305e-02 -8.65774810e-01 -7.14030683e-01 1.06765413e+00 2.66837418e-01 -1.52021974e-01 1.40991235e+00 4.17658120e-01 -1.89461280e-02 -1.01295054e+00 -8.34971368e-01 -1.06447828e+00 -4.33962673e-01 -6.26733720e-01 1.87793285e-01 8.73213112e-01 -3.16361189e-01 -2.72974759e-01 -7.17341483e-01 -1.49352476e-01 8.71182024e-01 -2.44965926e-01 7.06841111e-01 -6.21417344e-01 -3.07763904e-01 -1.99364677e-01 -7.12663054e-01 -1.78727055e+00 5.64050898e-02 -6.37878954e-01 -9.75400731e-02 -1.23552036e+00 2.49532565e-01 -4.31225836e-01 -5.45644939e-01 3.14552367e-01 -3.03764552e-01 1.05563939e+00 7.41005242e-01 2.80720234e-01 -1.35964787e+00 3.77195984e-01 1.22582626e+00 -3.21520180e-01 4.48865220e-02 -2.33988389e-01 -3.60686690e-01 3.96299988e-01 5.83058238e-01 -4.29391921e-01 -4.62118149e-01 -4.82772022e-01 -3.35799068e-01 5.39908968e-02 2.96772391e-01 -1.03086305e+00 2.32976466e-01 -9.10978988e-02 1.85112551e-01 -6.99921012e-01 3.37817252e-01 -8.33122671e-01 6.68299617e-03 4.12329942e-01 -6.83761910e-02 6.41797334e-02 3.26960146e-01 5.65318704e-01 -4.77274686e-01 -3.75511907e-02 6.45359397e-01 2.85984576e-01 -1.50977075e+00 9.26133573e-01 2.69174073e-02 3.59632581e-01 1.49693286e+00 -3.18682939e-01 -5.46174407e-01 -5.01850545e-01 -6.80250108e-01 8.26416671e-01 6.91125989e-01 5.75185299e-01 9.36566949e-01 -1.50302887e+00 -6.13862216e-01 -2.31460810e-01 -5.49014881e-02 -4.11808044e-01 4.58102763e-01 1.03119051e+00 -9.37497914e-01 2.08473518e-01 -2.66955823e-01 -6.73637271e-01 -1.49941361e+00 3.24719757e-01 1.17003478e-01 -3.72640900e-02 -9.07852590e-01 1.06560385e+00 -3.08307350e-01 3.03964347e-01 4.97375041e-01 1.27679676e-01 -1.65428683e-01 3.00008953e-02 7.83631742e-01 5.81222236e-01 -4.10406888e-01 -5.58215499e-01 -5.42399049e-01 5.00010431e-01 -3.66465867e-01 -7.75093958e-02 9.80166018e-01 -4.63561594e-01 4.61222678e-01 1.08425327e-01 1.72737908e+00 4.14146064e-03 -1.77893198e+00 3.96422669e-02 -2.01514930e-01 -8.01530957e-01 1.80989057e-01 -3.88762742e-01 -1.82508934e+00 6.75060689e-01 8.73611629e-01 1.47526339e-01 1.08395708e+00 -1.18209824e-01 1.53652596e+00 8.13281462e-02 6.54550612e-01 -1.01555383e+00 -1.56061566e-02 4.00647163e-01 3.87618810e-01 -1.49438477e+00 7.11570233e-02 -4.96971756e-01 -5.77729881e-01 1.15484178e+00 7.78308392e-01 -2.12982684e-01 4.81358469e-01 2.57063359e-01 3.10228076e-02 -6.28237501e-02 -8.49499285e-01 -3.67316082e-02 6.07873499e-01 3.83116663e-01 4.28761452e-01 -3.30795348e-01 -6.59942031e-01 1.48029834e-01 4.27996486e-01 2.99701959e-01 5.74936867e-01 6.96533024e-01 -1.31421134e-01 -1.05303705e+00 2.00266466e-01 4.48776096e-01 -6.68754041e-01 -1.44550189e-01 1.33706704e-01 7.78144956e-01 3.88164865e-03 8.06737363e-01 4.47128937e-02 -7.17357934e-01 -4.92549762e-02 -3.38871270e-01 5.12409091e-01 -1.92700356e-01 -2.55139738e-01 6.84251562e-02 2.58965492e-01 -1.27085686e+00 -5.86411417e-01 -4.70358521e-01 -1.46887481e+00 -6.01048827e-01 -2.08887860e-01 1.43806174e-01 6.21689737e-01 8.25077951e-01 4.72059011e-01 4.53867733e-01 6.60429001e-01 -1.05142522e+00 -6.01038337e-02 -3.35908830e-01 -3.67650837e-01 4.59906131e-01 2.39154205e-01 -5.69164336e-01 -3.52014482e-01 2.36496136e-01]	[9.171192169189453, 0.016688993200659752]
9f172c26-aec2-4972-9b41-bfa653e753c5	improved-low-resource-somali-speech	1907.03064	null	https://arxiv.org/abs/1907.03064v1	https://arxiv.org/pdf/1907.03064v1.pdf	Improved low-resource Somali speech recognition by semi-supervised acoustic and language model training	We present improvements in automatic speech recognition (ASR) for Somali, a currently extremely under-resourced language. This forms part of a continuing United Nations (UN) effort to employ ASR-based keyword spotting systems to support humanitarian relief programmes in rural Africa. Using just 1.57 hours of annotated speech data as a seed corpus, we increase the pool of training data by applying semi-supervised training to 17.55 hours of untranscribed speech. We make use of factorised time-delay neural networks (TDNN-F) for acoustic modelling, since these have recently been shown to be effective in resource-scarce situations. Three semi-supervised training passes were performed, where the decoded output from each pass was used for acoustic model training in the subsequent pass. The automatic transcriptions from the best performing pass were used for language model augmentation. To ensure the quality of automatic transcriptions, decoder confidence is used as a threshold. The acoustic and language models obtained from the semi-supervised approach show significant improvement in terms of WER and perplexity compared to the baseline. Incorporating the automatically generated transcriptions yields a 6.55\% improvement in language model perplexity. The use of 17.55 hour of Somali acoustic data in semi-supervised training shows an improvement of 7.74\% relative over the baseline.	['Thomas Niesler', 'Raghav Menon', 'Astik Biswas', 'Ewald van der Westhuizen']	2019-07-06	null	null	null	null	['acoustic-modelling']	['speech']	[ 5.59239030e-01 1.53274670e-01 2.16030017e-01 -4.48257864e-01 -1.47572494e+00 -4.82936710e-01 5.52095532e-01 1.59865230e-01 -9.44802284e-01 5.98375916e-01 6.36739314e-01 -8.87938857e-01 2.12712884e-01 -2.64956594e-01 -1.98849007e-01 -6.61847234e-01 -1.41501039e-01 6.02342963e-01 -1.28694579e-01 -4.23533112e-01 1.02932099e-02 5.12067497e-01 -8.33184183e-01 3.53045464e-01 7.24096179e-01 2.90008545e-01 5.75360477e-01 1.20956159e+00 -3.20171937e-02 4.06958222e-01 -8.81910384e-01 4.74836417e-02 5.76750375e-02 -5.03822505e-01 -7.35575020e-01 -7.52399862e-02 -8.35052058e-02 -3.63800138e-01 -6.18471622e-01 4.33891326e-01 8.75102937e-01 2.69089073e-01 3.32405567e-01 -3.35303009e-01 5.59239723e-02 5.72275162e-01 -2.15066478e-01 5.76156318e-01 3.49229932e-01 -5.81019633e-02 7.55926073e-01 -1.01500165e+00 2.49765307e-01 1.35986662e+00 5.26879430e-01 5.05509019e-01 -1.22995472e+00 -7.32102752e-01 -2.95454621e-01 -1.94039866e-01 -1.37551939e+00 -1.35886729e+00 3.24298054e-01 -1.50528088e-01 1.75921392e+00 2.40891144e-01 2.79009342e-01 7.26428211e-01 -1.57789707e-01 5.01936555e-01 9.16025698e-01 -9.43297207e-01 1.84061497e-01 1.15409054e-01 -3.26268584e-01 3.09861928e-01 -3.29337090e-01 3.71318981e-02 -4.98421758e-01 -3.81399751e-01 4.62160707e-01 -6.47171438e-01 -3.60841453e-02 6.09412909e-01 -1.20223677e+00 1.04586160e+00 -8.60898048e-02 4.74374294e-01 -5.29054642e-01 -1.93511657e-02 5.14552116e-01 3.37618530e-01 8.18439543e-01 3.95927548e-01 -5.35224319e-01 -6.08540893e-01 -1.29621029e+00 -1.15740351e-01 6.52825415e-01 5.23476958e-01 3.46738577e-01 7.28083432e-01 3.16163003e-02 1.61063254e+00 4.68456209e-01 1.04378140e+00 5.35893857e-01 -8.07545841e-01 7.83323646e-01 3.90204303e-02 -7.20119774e-02 -5.57838202e-01 -2.11967915e-01 -2.79987246e-01 -3.76689166e-01 -4.33296382e-01 2.33319342e-01 -4.47237432e-01 -1.49684763e+00 1.56575930e+00 1.97370630e-02 -1.20958544e-01 3.58323663e-01 3.88283581e-01 2.66794294e-01 1.36792195e+00 1.11324012e-01 -5.38830876e-01 8.59455466e-01 -7.42311895e-01 -9.00607646e-01 -4.44613963e-01 1.00621045e+00 -1.19815230e+00 7.40432680e-01 3.25217605e-01 -1.11699736e+00 -1.86872095e-01 -8.37716401e-01 3.98366362e-01 -1.74570858e-01 3.31893593e-01 4.98294048e-02 1.02328324e+00 -1.38645828e+00 7.24682286e-02 -9.72270608e-01 -4.61444616e-01 -5.09979986e-02 5.76456606e-01 -4.40395772e-01 -1.78794175e-01 -1.38310444e+00 1.06359875e+00 3.05878848e-01 2.96118438e-01 -8.76061916e-01 -7.61932582e-02 -1.20053673e+00 -2.65433669e-01 6.64584413e-02 3.06411803e-01 1.48083758e+00 -6.51491582e-01 -1.51024151e+00 7.06050634e-01 -5.55708110e-01 -5.11434138e-01 -1.03231110e-02 -9.35029984e-02 -7.25460172e-01 2.86148310e-01 3.51197943e-02 4.79576707e-01 5.41751564e-01 -9.08684790e-01 -8.24438870e-01 -1.43525362e-01 -4.31485385e-01 4.87463981e-01 -2.26163402e-01 7.89997697e-01 -5.48700750e-01 -8.02633524e-01 1.58812389e-01 -1.07153213e+00 -2.78590530e-01 -9.09439862e-01 -1.20777756e-01 5.28375059e-02 6.24343812e-01 -1.63183284e+00 1.69322526e+00 -2.10647583e+00 -1.41564891e-01 5.72285533e-01 -5.33230484e-01 7.52186775e-01 -2.70895630e-01 7.58022666e-01 -3.76582220e-02 1.02428071e-01 -4.75444585e-01 -5.38279533e-01 -3.23975235e-01 5.87972522e-01 -2.01236472e-01 3.58404279e-01 3.59338135e-01 4.16927069e-01 -7.68139601e-01 -1.86800838e-01 2.32522041e-01 6.47584081e-01 -3.14375222e-01 3.21083516e-01 2.63845563e-01 3.51991743e-01 7.12529570e-02 5.47280014e-01 1.53622553e-01 6.36689663e-01 1.46338642e-01 5.02460420e-01 -1.09756008e-01 1.00416577e+00 -8.47498834e-01 1.47074056e+00 -7.41588533e-01 6.64497554e-01 2.31294870e-01 -9.28703725e-01 1.11357701e+00 7.97584176e-01 2.11389393e-01 -8.52355719e-01 -2.00435907e-01 6.75466835e-01 4.53318208e-01 -3.38116765e-01 5.84788799e-01 -2.97245026e-01 -3.23319100e-02 4.47250038e-01 1.66548222e-01 -3.03260118e-01 1.05237864e-01 2.49639496e-01 1.07150948e+00 -2.32257739e-01 8.60763639e-02 2.80548856e-02 4.80674654e-01 6.37890324e-02 4.31048393e-01 6.74538851e-01 -5.98667562e-02 6.38872504e-01 -2.52838861e-02 2.37708800e-02 -1.46294534e+00 -6.62412882e-01 -2.48876009e-02 1.15206957e+00 -1.02827823e+00 -4.77822542e-01 -8.74741673e-01 -2.93166935e-01 -5.32899976e-01 1.04046464e+00 -1.08552210e-01 7.76114017e-02 -9.58946466e-01 -7.67091811e-01 1.26395595e+00 2.87037432e-01 7.58701637e-02 -1.22557831e+00 -1.25920758e-01 7.19403028e-01 -3.03920537e-01 -1.12873662e+00 -4.82935965e-01 5.34565747e-01 -6.78257048e-01 -3.72212261e-01 -1.14936697e+00 -8.14294219e-01 4.35540497e-01 1.75381720e-01 5.41289389e-01 -2.55562644e-02 -1.57930534e-02 2.34930798e-01 -4.31126595e-01 -2.93496490e-01 -1.08928716e+00 9.04334933e-02 4.38438475e-01 -2.82607973e-01 3.16976994e-01 -1.35056138e-01 -8.88409242e-02 1.58203259e-01 -9.13077354e-01 -1.68352902e-01 6.11292660e-01 1.05235350e+00 3.03084433e-01 -9.58869681e-02 7.94431090e-01 -5.92581570e-01 7.65151978e-01 -3.70220929e-01 -2.79689789e-01 1.32025287e-01 -4.59623069e-01 -9.80972201e-02 4.41678941e-01 -2.56713986e-01 -1.35243988e+00 9.32954997e-02 -7.56275535e-01 1.00486763e-01 -1.90184161e-01 9.40756679e-01 -4.53131832e-02 2.76781023e-01 5.86252093e-01 3.09484184e-01 2.13832617e-01 -4.53447342e-01 1.25313789e-01 1.61355007e+00 5.94509363e-01 -2.54678130e-01 7.03904033e-01 -1.36247694e-01 -7.22063005e-01 -1.51910186e+00 -2.57201046e-01 -8.93835962e-01 -5.07543445e-01 8.95100385e-02 4.55776721e-01 -9.16562021e-01 1.60197079e-01 5.94455361e-01 -1.23694301e+00 -6.41859591e-01 2.11618096e-01 7.18148053e-01 -3.38028640e-01 2.87641168e-01 -5.33969939e-01 -1.47785628e+00 -6.42121136e-01 -1.10474801e+00 9.50828671e-01 -1.47276402e-01 -4.21551615e-01 -1.06265366e+00 2.97428131e-01 4.87827390e-01 5.31183600e-01 -4.04852062e-01 7.41561472e-01 -1.12030005e+00 4.69404340e-01 -3.47136945e-01 9.63388532e-02 8.22589517e-01 3.39303613e-01 -2.24180639e-01 -1.10024834e+00 -4.74754810e-01 -1.35498658e-01 -1.79538354e-01 5.53247571e-01 2.44458467e-01 4.30463403e-01 -4.68330622e-01 -5.40556386e-02 1.02375537e-01 9.11694765e-01 9.34946537e-01 6.39792800e-01 1.65768191e-01 3.37713003e-01 6.47883415e-01 5.90233743e-01 2.86364943e-01 1.43527076e-01 7.86592782e-01 -4.33854282e-01 -2.19674006e-01 2.51143500e-02 -2.18449995e-01 8.11416924e-01 1.41988671e+00 2.78909653e-01 -2.47789428e-01 -1.56689775e+00 9.97327626e-01 -1.37864733e+00 -8.14399660e-01 1.97556876e-02 2.42815399e+00 1.07505071e+00 2.33541071e-01 1.83846548e-01 4.18783963e-01 6.83500290e-01 -6.10032752e-02 3.22931237e-03 -9.38623428e-01 -9.29999650e-02 6.11180782e-01 6.24140143e-01 1.01734173e+00 -8.04191709e-01 1.17508066e+00 6.25244093e+00 1.03598905e+00 -1.10616255e+00 1.76288217e-01 6.19212151e-01 -2.12955903e-02 -1.68058231e-01 -5.04081324e-02 -7.40410566e-01 3.60158682e-01 2.03401780e+00 1.19737759e-02 7.10028291e-01 2.17420489e-01 9.99494255e-01 -4.51579332e-01 -5.86621284e-01 8.56619716e-01 1.08653978e-01 -9.53698575e-01 -2.75104016e-01 1.61293313e-01 4.97154027e-01 3.94621909e-01 -1.72997802e-01 2.46919259e-01 2.14341968e-01 -1.24337339e+00 4.40428644e-01 -5.21789789e-02 1.11896288e+00 -1.07342005e+00 8.98804128e-01 4.74476010e-01 -1.00864875e+00 1.81167692e-01 -1.45107031e-01 5.20766564e-02 4.44596171e-01 2.34606281e-01 -1.56642103e+00 2.96739966e-01 3.48893702e-01 1.66038293e-02 -2.46250257e-01 9.01412666e-01 -8.09825882e-02 1.54987776e+00 -7.16617644e-01 1.32054538e-01 6.83908641e-01 1.10546432e-01 5.37525415e-01 1.86150825e+00 2.86099464e-01 2.32246116e-01 -8.54451358e-02 6.15367619e-03 2.50776619e-01 2.51895279e-01 -6.80742979e-01 -4.47433084e-01 7.64879584e-01 6.70621634e-01 -6.12820625e-01 -3.30947280e-01 -1.63576201e-01 1.12486100e+00 -6.63327053e-02 4.34689760e-01 -3.34444106e-01 -7.63990939e-01 3.22869033e-01 -7.47898966e-02 2.29853131e-02 -6.22980833e-01 -1.05151258e-01 -5.47202170e-01 -2.03699037e-01 -1.21508849e+00 1.76876694e-01 -5.73540211e-01 -5.46064436e-01 8.42109740e-01 9.48746949e-02 -6.00900590e-01 -8.34623992e-01 -3.11457902e-01 -5.00726938e-01 1.57148254e+00 -1.11337721e+00 -9.17082548e-01 5.26434839e-01 4.03175987e-02 1.12274337e+00 -5.13943851e-01 1.38604629e+00 3.93053859e-01 -4.08709735e-01 6.19754791e-01 4.33248937e-01 2.04378501e-01 4.03272003e-01 -1.04918182e+00 8.42204571e-01 1.01117325e+00 1.87545821e-01 8.08311164e-01 5.85856497e-01 -8.48609746e-01 -1.07236958e+00 -9.30908859e-01 1.65781927e+00 -1.91104844e-01 6.26744688e-01 -5.53611517e-01 -8.56759429e-01 4.55730557e-01 1.45878047e-01 -7.65904248e-01 9.04540479e-01 -2.08630472e-01 1.68467149e-01 8.73140469e-02 -9.59114254e-01 3.62147123e-01 4.75334972e-01 -9.46751654e-01 -6.27386868e-01 2.07150355e-01 7.46297300e-01 -2.00140700e-01 -5.79354644e-01 1.11842677e-01 4.77575123e-01 -6.25446290e-02 6.40680254e-01 -3.00454557e-01 -1.57464087e-01 -1.10742398e-01 -4.02150810e-01 -1.54990077e+00 7.88563862e-02 -1.14621985e+00 4.02145773e-01 1.24617696e+00 8.43566656e-01 -5.29612482e-01 4.68134224e-01 4.32922632e-01 -3.14909339e-01 -5.46637118e-01 -1.15086901e+00 -7.77078927e-01 -5.08523248e-02 -8.06482494e-01 7.13096512e-03 5.62997699e-01 8.66991356e-02 4.11317974e-01 -4.40143406e-01 1.53303683e-01 1.97547227e-01 -1.08388340e+00 4.10414070e-01 -5.56425571e-01 -2.09220931e-01 1.52473850e-02 -9.01564956e-02 -8.43958735e-01 -4.91993241e-02 -6.21125400e-01 4.67482030e-01 -1.65720069e+00 -2.83641338e-01 -5.64669549e-01 -1.13880873e-01 7.78059483e-01 -3.85100543e-02 3.08458328e-01 1.09782554e-01 9.67534035e-02 1.96215570e-01 3.93120438e-01 4.02406812e-01 -5.07431887e-02 -6.34251535e-01 2.19578728e-01 -4.55238104e-01 3.54349256e-01 1.13635194e+00 -5.62074482e-01 -3.95648599e-01 -4.80030656e-01 -2.29469031e-01 2.29906574e-01 -2.92232603e-01 -7.35807657e-01 3.44216228e-02 -1.26881599e-01 6.53177202e-02 -6.51510715e-01 5.39773226e-01 -4.77931768e-01 1.60528749e-01 5.68333507e-01 -3.53433549e-01 2.39572018e-01 7.09852815e-01 1.49525464e-01 -1.46317035e-01 -4.46116239e-01 6.17087364e-01 3.78186218e-02 -4.45486516e-01 -1.94527790e-01 -1.00441372e+00 -1.39784083e-01 3.82704556e-01 -3.75075996e-01 2.43888661e-01 -7.19627380e-01 -7.49407291e-01 6.08928129e-03 9.98476025e-06 2.88427830e-01 7.42037296e-01 -9.32290018e-01 -9.59493697e-01 3.25781137e-01 -7.48777837e-02 -1.86927006e-01 -1.11722484e-01 7.65774429e-01 -6.20127380e-01 8.89186561e-01 2.64511496e-01 -2.69130498e-01 -1.38597691e+00 -2.35820055e-01 1.89891607e-01 -1.83803439e-01 -2.44508713e-01 8.23591232e-01 -3.17928195e-01 -6.34019256e-01 4.32037592e-01 2.09718850e-02 -2.87887529e-02 -1.25827000e-01 5.40830553e-01 3.46169889e-01 4.73024756e-01 -1.09492576e+00 -4.20704931e-01 -1.50562942e-01 -2.09054932e-01 -1.15137327e+00 1.48911893e+00 -2.32902423e-01 1.62834957e-01 5.09576619e-01 1.11078310e+00 3.30491275e-01 -8.57407749e-01 -2.18001410e-01 2.44269967e-01 -4.10103425e-02 3.51706147e-01 -1.13218176e+00 -2.53058821e-01 9.40731168e-01 4.63886887e-01 -1.30445078e-01 9.09192860e-01 -3.29669148e-01 6.98064983e-01 5.49219191e-01 1.41161188e-01 -1.39414084e+00 -4.21384037e-01 9.53913450e-01 8.28238964e-01 -8.12218964e-01 -4.43346590e-01 5.27016521e-02 -6.45726144e-01 1.03015053e+00 6.46023974e-02 4.29614604e-01 3.04404408e-01 4.19661522e-01 4.12564963e-01 1.80925757e-01 -5.23674726e-01 -1.49561673e-01 1.06958762e-01 7.17784703e-01 6.49876356e-01 1.55896530e-01 -3.57262015e-01 1.30320877e-01 -3.46191853e-01 -4.00904328e-01 4.08078641e-01 9.67893720e-01 -7.01783001e-01 -1.32112837e+00 -5.94941378e-01 3.92548501e-01 -6.77728891e-01 -6.12290025e-01 -4.07178491e-01 3.87894243e-01 -3.78813803e-01 1.38028765e+00 5.35030216e-02 -3.11018735e-01 2.58070201e-01 3.39994818e-01 -3.44518684e-02 -1.00387788e+00 -5.45693696e-01 7.31893778e-01 7.70571709e-01 -8.21576919e-03 -9.87463295e-02 -6.22903168e-01 -1.40557277e+00 -1.38706908e-01 -5.50279975e-01 5.69750071e-01 1.22795463e+00 9.53813851e-01 2.10322626e-02 1.89451382e-01 8.80887210e-01 -8.16494226e-01 -5.99534810e-01 -1.45409238e+00 -2.12082222e-01 -2.12877154e-01 3.67279738e-01 -8.77132788e-02 -3.15064400e-01 1.04128629e-01]	[14.375754356384277, 6.857481479644775]
1e163303-6466-496c-82f1-e483ac8a1dd5	diverse-instance-discovery-vision-transformer	2204.10731	null	https://arxiv.org/abs/2204.10731v1	https://arxiv.org/pdf/2204.10731v1.pdf	Diverse Instance Discovery: Vision-Transformer for Instance-Aware Multi-Label Image Recognition	Previous works on multi-label image recognition (MLIR) usually use CNNs as a starting point for research. In this paper, we take pure Vision Transformer (ViT) as the research base and make full use of the advantages of Transformer with long-range dependency modeling to circumvent the disadvantages of CNNs limited to local receptive field. However, for multi-label images containing multiple objects from different categories, scales, and spatial relations, it is not optimal to use global information alone. Our goal is to leverage ViT's patch tokens and self-attention mechanism to mine rich instances in multi-label images, named diverse instance discovery (DiD). To this end, we propose a semantic category-aware module and a spatial relationship-aware module, respectively, and then combine the two by a re-constraint strategy to obtain instance-aware attention maps. Finally, we propose a weakly supervised object localization-based approach to extract multi-scale local features, to form a multi-view pipeline. Our method requires only weakly supervised information at the label level, no additional knowledge injection or other strongly supervised information is required. Experiments on three benchmark datasets show that our method significantly outperforms previous works and achieves state-of-the-art results under fair experimental comparisons.	['Hui Xue', 'Yuan He', 'Feihu Yan', 'Jingfeng Zhang', 'Haiwen Hong', 'Yin Zhang', 'Xuan Jin', 'Yunqing Hu']	2022-04-22	null	null	null	null	['weakly-supervised-object-localization']	['computer-vision']	[ 2.26220593e-01 -3.03305775e-01 -2.62527525e-01 -6.28679812e-01 -8.30838263e-01 -4.81901407e-01 5.19168198e-01 -3.81575227e-02 -3.64544004e-01 4.75942135e-01 -9.32434872e-02 1.71349812e-02 -9.83603075e-02 -6.80330634e-01 -7.79339910e-01 -7.60955572e-01 6.10953212e-01 2.43230760e-01 4.92488891e-01 5.81374429e-02 3.19680333e-01 4.14204776e-01 -1.57238662e+00 6.77831590e-01 7.18939602e-01 1.22857976e+00 4.82738286e-01 1.24758467e-01 -2.19737411e-01 1.22615314e+00 -2.56495297e-01 -3.20147067e-01 8.61610845e-02 -2.22455859e-01 -9.89386499e-01 4.96687293e-01 5.96730471e-01 -2.20090747e-01 2.54653413e-02 1.11096919e+00 3.78998756e-01 7.30524734e-02 5.41827857e-01 -1.05751657e+00 -9.25537884e-01 2.98159331e-01 -8.61578643e-01 1.35622457e-01 -6.07596114e-02 7.56855235e-02 1.09898913e+00 -1.22823679e+00 4.61254030e-01 1.22911775e+00 6.82374895e-01 4.42952871e-01 -1.03260612e+00 -6.27409279e-01 6.79263175e-01 4.12059516e-01 -1.46816564e+00 -3.03515762e-01 9.14618134e-01 -1.71521679e-01 7.88488030e-01 2.92097144e-02 4.44911450e-01 1.04687726e+00 -1.66350991e-01 1.00087333e+00 1.43438673e+00 -3.78384858e-01 8.80486816e-02 3.14946473e-01 9.01046619e-02 9.20569241e-01 -1.57507256e-01 -1.36992991e-01 -5.40307403e-01 5.03511168e-02 7.86934972e-01 4.50945884e-01 -1.44845575e-01 -5.07429659e-01 -1.35178399e+00 7.41456091e-01 9.11762893e-01 4.68904465e-01 -5.19934058e-01 7.08380714e-02 2.79945225e-01 1.10127412e-01 6.11109734e-01 1.43414363e-01 -5.63749075e-01 4.22279447e-01 -7.71890640e-01 -1.44497668e-02 2.83384740e-01 1.17042315e+00 1.02097511e+00 -4.97484714e-01 -3.09501022e-01 1.27593732e+00 3.98283780e-01 2.10010409e-01 4.45238680e-01 -6.26313269e-01 3.90936077e-01 1.01437974e+00 -1.91324607e-01 -8.45112205e-01 -4.71975535e-01 -7.89474308e-01 -8.74836802e-01 -1.18976392e-01 1.37884900e-01 3.38769674e-01 -1.06326425e+00 1.49953103e+00 3.77250910e-01 4.03777242e-01 -1.12375677e-01 1.11951149e+00 1.02805567e+00 3.36125404e-01 2.17090830e-01 -2.22251624e-01 1.56389797e+00 -1.53848839e+00 -4.59383279e-01 -3.83206695e-01 5.54860115e-01 -5.72666943e-01 1.03753567e+00 1.22551404e-01 -9.65843856e-01 -6.36099935e-01 -9.33023036e-01 -2.20625356e-01 -4.83009964e-01 2.89881170e-01 4.51678842e-01 2.14361012e-01 -9.14356589e-01 2.47621149e-01 -4.07421440e-01 -3.27481568e-01 7.29760468e-01 1.98813051e-01 -4.13273454e-01 -5.49135625e-01 -9.35078740e-01 7.56480455e-01 3.55416924e-01 1.75172254e-01 -9.60917592e-01 -5.81570983e-01 -8.52295399e-01 -1.19104898e-02 5.93285382e-01 -6.97909355e-01 1.09241927e+00 -1.09985173e+00 -1.11055160e+00 9.85897720e-01 -2.51405597e-01 8.98478776e-02 1.14281237e-01 1.16412707e-01 -2.19293669e-01 2.36778766e-01 4.45016891e-01 8.12854290e-01 6.43243253e-01 -1.56465948e+00 -9.47795093e-01 -5.15434444e-01 4.36703712e-01 4.72848058e-01 -4.51210201e-01 1.53377846e-01 -7.27142692e-01 -5.06235182e-01 3.69559437e-01 -8.70300174e-01 -1.37871787e-01 -2.27998763e-01 -4.46169674e-01 -4.92670655e-01 6.21479571e-01 -3.86658132e-01 8.95387650e-01 -2.05943489e+00 1.59830227e-01 -8.35916400e-02 2.86223829e-01 7.19618797e-02 -3.03452879e-01 8.84893686e-02 3.53845209e-02 1.23166069e-02 -1.93183571e-01 -5.55360675e-01 -1.93295226e-01 1.72634691e-01 1.05094248e-02 4.70113873e-01 3.92098159e-01 1.17771971e+00 -9.61458087e-01 -7.39498317e-01 2.15812564e-01 3.09052497e-01 -3.48409772e-01 1.30396068e-01 -2.88546473e-01 6.07779384e-01 -5.63391030e-01 1.12041795e+00 8.02662075e-01 -7.30468154e-01 1.13880530e-01 -6.60953343e-01 -2.31958732e-01 1.39887873e-02 -1.03005075e+00 2.00611067e+00 -7.44773328e-01 2.17990251e-03 6.78248471e-03 -1.27567220e+00 8.11316729e-01 2.25576520e-01 4.18046683e-01 -8.10744047e-01 6.58790618e-02 2.69614547e-01 -3.37114543e-01 -5.58708429e-01 1.88643113e-01 -2.59336323e-01 7.56740868e-02 3.67045343e-01 3.74841481e-01 4.71604764e-01 -1.27028199e-02 2.81638987e-02 9.53839064e-01 2.34374300e-01 2.60007888e-01 -1.19797841e-01 6.31291211e-01 -1.41812295e-01 8.10981929e-01 6.02439523e-01 -2.15283811e-01 7.62633801e-01 2.59128183e-01 -5.28106332e-01 -6.67827368e-01 -6.87185526e-01 -8.75208825e-02 1.42580223e+00 5.07289112e-01 -1.69870079e-01 -4.28667039e-01 -1.30831230e+00 -2.39366651e-01 1.97696447e-01 -7.65312076e-01 1.67617649e-01 -4.83680218e-01 -8.80317628e-01 3.11216861e-01 7.11863041e-01 8.52739751e-01 -1.20723963e+00 -2.17936695e-01 1.50533110e-01 -4.60420221e-01 -1.24898839e+00 -4.49126929e-01 3.51327062e-01 -6.40748739e-01 -1.03408861e+00 -7.93363214e-01 -1.00856316e+00 6.98198915e-01 7.55169749e-01 1.05634725e+00 1.74016058e-01 -1.88309252e-01 1.02813646e-01 -3.37738037e-01 -1.54095858e-01 2.49410510e-01 1.64909512e-01 -3.89774591e-01 4.62223768e-01 3.48861098e-01 -6.07743800e-01 -7.46348798e-01 6.07842624e-01 -7.96834230e-01 1.60980791e-01 1.04218400e+00 1.04746604e+00 9.56408858e-01 -1.08963504e-01 8.05270255e-01 -9.66029465e-01 9.84946266e-02 -6.14375830e-01 -3.62137556e-01 5.71102977e-01 -5.72392285e-01 -2.30555952e-01 6.13157392e-01 -4.55631226e-01 -1.08694637e+00 3.14739734e-01 8.97953808e-02 -7.35223234e-01 -4.78461772e-01 4.66688424e-01 -5.29690564e-01 -1.99012935e-01 3.85332108e-01 4.67357188e-01 -3.25377017e-01 -6.24648333e-01 4.83281463e-01 6.96091533e-01 2.34573752e-01 -3.98966819e-01 4.60378826e-01 7.01980293e-01 -1.94629759e-01 -3.71885478e-01 -1.54283547e+00 -9.16459322e-01 -8.58295023e-01 -2.41006494e-01 1.00525081e+00 -1.15271831e+00 -5.88411093e-01 5.60506046e-01 -1.12764132e+00 -4.31748852e-02 2.35012267e-02 2.14200824e-01 -4.30718988e-01 1.36198357e-01 -6.16739571e-01 -4.96180058e-01 -1.61343232e-01 -1.29768121e+00 1.37661684e+00 3.80431324e-01 5.42972744e-01 -9.74503338e-01 -1.88332543e-01 7.62052953e-01 4.16802824e-01 -4.90642972e-02 6.62238061e-01 -6.53415382e-01 -9.95967150e-01 1.29836574e-01 -8.33293557e-01 4.14777786e-01 1.44309968e-01 -6.39644563e-01 -1.31412315e+00 -2.23414049e-01 -5.76045066e-02 -6.09208286e-01 1.06320608e+00 1.46918550e-01 1.39366305e+00 -1.81356952e-01 -4.57807004e-01 6.95002675e-01 1.69405138e+00 -2.01302156e-01 3.53772432e-01 3.38759840e-01 1.20462847e+00 7.67028272e-01 7.76868820e-01 1.95921466e-01 9.08034086e-01 7.87751913e-01 6.87500358e-01 -2.45846197e-01 -2.35400259e-01 -1.09655224e-02 3.01891305e-02 9.28399503e-01 -4.28834073e-02 -8.00044686e-02 -6.89556718e-01 7.30002820e-01 -2.06911898e+00 -8.86303306e-01 3.24122161e-02 1.77240765e+00 7.39696801e-01 -1.30256146e-01 6.63129613e-02 -2.40145624e-01 8.02792490e-01 2.30295777e-01 -6.81815147e-01 8.46162289e-02 -2.47520935e-02 -1.30383998e-01 5.57656050e-01 8.74938816e-02 -1.37010491e+00 9.78541374e-01 5.27928352e+00 9.85001504e-01 -1.11394370e+00 4.97688055e-01 6.42813683e-01 -1.27623633e-01 -2.03522325e-01 -9.07142535e-02 -9.36874032e-01 3.26767534e-01 4.69913393e-01 5.18340945e-01 3.63298506e-01 7.89581418e-01 -1.48172408e-01 9.07476842e-02 -9.94225025e-01 1.09729290e+00 3.58306408e-01 -1.26455724e+00 -5.25598414e-02 9.52617675e-02 7.36297905e-01 1.58260569e-01 1.44438846e-02 3.12115073e-01 1.82157829e-01 -9.05936241e-01 6.69266105e-01 6.00867391e-01 7.93635428e-01 -7.06120431e-01 8.86353016e-01 5.04079103e-01 -1.51047540e+00 -2.93764681e-01 -5.19272208e-01 2.06470922e-01 -5.14988340e-02 6.42901421e-01 -5.21943331e-01 8.83731663e-01 7.96517491e-01 1.11997998e+00 -7.97276556e-01 9.00532484e-01 -1.13503061e-01 4.18529093e-01 -1.08787313e-01 1.52641535e-01 4.37667280e-01 9.60294828e-02 1.77555308e-02 1.11814427e+00 4.62081516e-03 1.06225729e-01 6.25009835e-01 8.63813579e-01 -1.80331036e-01 1.82348102e-01 -3.89167756e-01 4.93915707e-01 4.53228623e-01 1.73551130e+00 -8.29182923e-01 -3.04255843e-01 -8.49965990e-01 1.09493864e+00 7.74326622e-01 3.67599249e-01 -8.79981697e-01 -1.09422460e-01 3.47357064e-01 -2.12381840e-01 5.07981062e-01 2.40143552e-01 -2.76539683e-01 -1.33569276e+00 1.85242638e-01 -8.60581219e-01 4.58139360e-01 -7.23738730e-01 -1.56711650e+00 4.19936717e-01 -2.13823855e-01 -1.23939312e+00 1.16266936e-01 -3.94622415e-01 -3.17626864e-01 8.97585154e-01 -1.98395324e+00 -2.03212047e+00 -4.96664256e-01 7.86925972e-01 7.65531540e-01 -6.46925494e-02 6.83374703e-01 5.19286156e-01 -5.98293245e-01 6.39696538e-01 -1.63526982e-01 2.48579886e-02 7.38044858e-01 -1.24553514e+00 -7.63919055e-02 5.60453534e-01 3.58056962e-01 4.96817857e-01 -4.99809794e-02 -3.67632657e-01 -1.15069377e+00 -1.46501040e+00 7.65631855e-01 -5.23373067e-01 4.09836411e-01 -3.11889827e-01 -8.85377645e-01 6.46471083e-01 1.18942380e-01 6.26140594e-01 6.93748951e-01 3.19412172e-01 -5.72934568e-01 -2.62926191e-01 -1.06999743e+00 2.01148689e-01 1.27962768e+00 -6.01285934e-01 -3.62943262e-01 4.25516605e-01 7.51085341e-01 -2.10090071e-01 -9.67006624e-01 6.54121578e-01 3.42378229e-01 -9.80771244e-01 1.12728179e+00 -2.35956281e-01 4.07965243e-01 -5.24374247e-01 -3.05237055e-01 -1.16565394e+00 -7.14962542e-01 3.26425523e-01 1.72719777e-01 1.49008894e+00 3.56292397e-01 -4.41676617e-01 4.96568710e-01 -6.86145388e-03 -3.67682368e-01 -1.08212471e+00 -7.60887027e-01 -6.08448863e-01 -1.87636361e-01 -2.02954620e-01 5.42356014e-01 1.28320420e+00 -3.35353464e-01 6.07913077e-01 -4.10794705e-01 4.02132720e-01 7.57488728e-01 5.98966837e-01 4.19388145e-01 -1.11511004e+00 -2.20626801e-01 -3.17143857e-01 -2.57647425e-01 -1.12995493e+00 2.50689507e-01 -1.06774080e+00 1.89770848e-01 -1.75501812e+00 6.79367781e-01 -7.28698611e-01 -9.02573884e-01 8.01994383e-01 -2.31684774e-01 7.33952224e-01 2.01756239e-01 4.96886671e-01 -1.28605914e+00 5.03220320e-01 1.37671256e+00 -3.30545902e-01 1.90374225e-01 -1.60359457e-01 -8.77182662e-01 7.25356638e-01 5.16073346e-01 -3.98016661e-01 -4.19441670e-01 -5.25945723e-01 1.00661024e-01 -1.64370507e-01 5.63220739e-01 -8.61033380e-01 2.71790594e-01 -2.64837205e-01 3.34844112e-01 -6.84637189e-01 2.53280610e-01 -8.51698458e-01 -2.89130926e-01 -5.70280664e-02 -4.12348628e-01 -1.86860099e-01 -1.30068153e-01 7.29687929e-01 -3.90735507e-01 -1.62314489e-01 8.49421442e-01 -4.00981128e-01 -1.00818622e+00 6.40335917e-01 5.26060946e-02 -7.93663934e-02 1.07080483e+00 1.38593810e-02 -5.13198793e-01 2.43723825e-01 -6.60335779e-01 3.16856325e-01 5.65809190e-01 6.27641320e-01 6.72042549e-01 -1.54880047e+00 -5.76038957e-01 1.37899503e-01 6.75684988e-01 1.57771915e-01 5.04546940e-01 9.66526330e-01 -2.51272228e-02 4.18424785e-01 -1.28437653e-01 -8.23417425e-01 -1.11444628e+00 7.84979045e-01 2.68427372e-01 -4.20287728e-01 -4.76306289e-01 9.62705016e-01 6.32236183e-01 -7.11592138e-01 1.18436702e-01 -5.89955114e-02 -5.59357047e-01 2.25567251e-01 5.75282395e-01 -8.11323598e-02 2.27919549e-01 -7.07701743e-01 -5.18948555e-01 8.11078608e-01 -3.37457895e-01 1.96226805e-01 1.33601737e+00 -4.60149944e-01 -3.43696028e-01 6.93216205e-01 1.32282531e+00 -3.00102144e-01 -1.21518123e+00 -6.76103413e-01 -4.14783508e-02 -4.18902606e-01 1.83731928e-01 -9.28430259e-01 -1.38044500e+00 8.39113772e-01 5.92626393e-01 1.48142457e-01 1.30156124e+00 3.64310771e-01 7.11475134e-01 1.33658394e-01 3.93380970e-01 -9.52846706e-01 3.95640522e-01 4.97613668e-01 7.20688343e-01 -1.57552183e+00 -2.61501580e-01 -5.74372292e-01 -6.34859681e-01 7.25582838e-01 9.73948479e-01 3.63372117e-02 6.52502060e-01 -2.61818580e-02 1.36435404e-01 -2.95107991e-01 -9.08653140e-01 -5.39436102e-01 4.00087625e-01 5.54051340e-01 3.25979441e-01 -3.80968563e-02 -1.57701820e-01 5.28951764e-01 5.54730058e-01 -1.02586430e-02 5.06478548e-02 8.38394046e-01 -5.71189225e-01 -1.02665424e+00 -2.49480158e-01 4.18864906e-01 -4.77307945e-01 -2.09953368e-01 -3.34897712e-02 3.88245165e-01 5.97997487e-01 9.63400483e-01 -9.10204798e-02 -4.10317898e-01 2.40704685e-01 1.08980291e-01 4.51781154e-01 -7.74727106e-01 -6.87532544e-01 8.35628286e-02 -1.14970595e-01 -7.76291132e-01 -8.88813794e-01 -5.31526744e-01 -9.79038537e-01 9.02636796e-02 -5.56596935e-01 -3.66036415e-01 6.48548126e-01 1.12043512e+00 4.39938933e-01 7.00448751e-01 6.89165294e-01 -7.52756000e-01 -2.96758413e-01 -1.08896315e+00 -4.62765574e-01 4.83174086e-01 2.68102407e-01 -8.78768861e-01 -1.32732078e-01 -1.89201366e-02]	[9.793025970458984, 3.877256393432617]
2670277f-ac16-4677-bbaf-6c212a379224	language-guided-image-clustering	null	null	https://openreview.net/forum?id=-JW-1Fg-v2	https://openreview.net/pdf?id=-JW-1Fg-v2	Language-Guided Image Clustering	Image clustering methods have rapidly improved their ability to discover object categories. However, unsupervised clustering methods struggle on other image attributes, e.g. age or activity. The reason is that most recent clustering methods learn deep features that are designed to be sensitive to object category, but less so to other image attributes. We propose to overcome this limitation by introducing the new setting of language-guided image clustering. In this setting, the model is provided with an exhaustive list of phrases describing all the possible values of a specific attribute, together with a shared image-language embedding (e.g. CLIP). Our method then computes the subset of K attribute phrases that form the best clustering of the images. Differently from standard clustering methods, our method can cluster according to image attributes other than the object category. We evaluate our method on a attribute clustering tasks and demonstrate that our method significantly outperforms methods that do not use language-guidance.	['Yedid Hoshen', 'Niv Cohen']	2021-09-29	null	null	null	null	['image-clustering']	['computer-vision']	[ 5.77851459e-02 -1.82780743e-01 -2.77371287e-01 -6.55471683e-01 -6.78573549e-01 -6.45322919e-01 7.27394700e-01 5.86020887e-01 -6.33954287e-01 9.20788050e-02 2.45290026e-01 1.80358559e-01 -2.99505174e-01 -7.05375314e-01 -3.78717691e-01 -1.14139223e+00 2.68279146e-02 7.69052863e-01 -1.34963363e-01 3.78972858e-01 3.24819386e-01 1.48270175e-01 -1.70766222e+00 2.57214814e-01 6.42007589e-01 8.56968701e-01 2.76562601e-01 4.39857781e-01 -1.81331351e-01 4.81418043e-01 -2.65329838e-01 -2.51904249e-01 1.67517795e-03 -3.60488385e-01 -7.66670704e-01 5.05231798e-01 4.20888960e-01 -3.05136368e-02 -1.79705724e-01 1.09720111e+00 3.03681195e-01 1.09337009e-01 1.05099094e+00 -1.49445426e+00 -8.37292314e-01 8.13976228e-01 -4.64744031e-01 -2.36280426e-01 4.72569950e-02 -4.01932783e-02 1.29871476e+00 -7.06623554e-01 7.37517297e-01 1.21883261e+00 5.37253201e-01 6.40143871e-01 -1.70349646e+00 -5.17436326e-01 4.71728593e-01 2.41083398e-01 -1.67698908e+00 -1.95977077e-01 8.19815338e-01 -6.56392336e-01 3.82682025e-01 6.26650080e-02 4.87826407e-01 8.87985826e-01 -1.93447605e-01 6.08297288e-01 1.18514442e+00 -2.41575658e-01 5.11838675e-01 3.35622460e-01 2.17426896e-01 5.70123315e-01 3.09434999e-02 -5.34508049e-01 -3.12883258e-01 -2.07748830e-01 2.57444024e-01 1.79764494e-01 7.60784224e-02 -9.22141433e-01 -1.47011435e+00 1.01696432e+00 3.94603461e-01 3.46108377e-01 -2.51642168e-01 2.53522307e-01 2.42383450e-01 1.30408071e-02 3.61142725e-01 4.88021344e-01 -4.16580826e-01 1.28457010e-01 -9.44681346e-01 1.32673621e-01 6.69392049e-01 9.38487291e-01 1.26200581e+00 -6.33141160e-01 -2.78893471e-01 9.32067037e-01 2.63291776e-01 4.07706261e-01 2.98587143e-01 -1.24123836e+00 -4.34988700e-02 6.08452022e-01 -1.74089253e-01 -1.13072097e+00 -3.50739092e-01 -2.64740705e-01 -8.32028627e-01 -1.13915019e-01 3.95200729e-01 8.76349956e-02 -1.02046454e+00 2.17043352e+00 1.69643044e-01 6.06753565e-02 -1.67148650e-01 6.76760256e-01 5.86650431e-01 4.74873692e-01 4.81981337e-01 -3.52485776e-02 1.40023088e+00 -6.31149232e-01 -4.88677263e-01 -8.94446820e-02 6.46086335e-01 -4.04935271e-01 1.09420180e+00 3.13735664e-01 -6.84379339e-01 -4.45279300e-01 -6.09019637e-01 1.29618347e-01 -5.34367085e-01 1.15906157e-01 7.49403656e-01 6.53214753e-01 -1.34850121e+00 4.47951943e-01 -6.78159416e-01 -7.96313643e-01 5.52369297e-01 6.08456969e-01 -4.47982371e-01 -7.04192892e-02 -7.12466657e-01 4.14193392e-01 5.56174219e-01 -5.79119802e-01 -6.54280722e-01 -5.87762713e-01 -7.65704870e-01 1.47398055e-01 3.76012772e-01 -8.15516293e-01 8.68062437e-01 -1.24087048e+00 -1.03257024e+00 1.35325766e+00 -2.28021502e-01 -4.15503502e-01 7.08146915e-02 1.03569441e-01 -6.77973032e-02 4.92646337e-01 2.59084255e-01 1.22687125e+00 9.28624690e-01 -1.64229012e+00 -6.71769261e-01 -3.24808627e-01 -5.26558012e-02 2.28702635e-01 -8.88925254e-01 -4.59444299e-02 -8.99541259e-01 -4.05445695e-01 1.99531883e-01 -1.12451637e+00 -4.99174327e-01 1.68469809e-02 -6.02138519e-01 -4.75215614e-01 6.57887518e-01 -1.02858588e-01 1.11757314e+00 -2.37179804e+00 3.19973677e-01 5.25119722e-01 4.74929065e-01 -3.56191695e-01 -1.61800653e-01 3.57747823e-01 1.69278216e-02 4.34334636e-01 -3.75650197e-01 -6.28214598e-01 1.96453273e-01 3.53652537e-01 9.19221714e-02 5.61782479e-01 1.68323368e-01 7.06604064e-01 -1.00029624e+00 -9.72122431e-01 1.32277086e-01 4.68393207e-01 -7.20330060e-01 -9.37345438e-03 -3.28500867e-01 4.09155965e-01 -5.10019004e-01 4.87912804e-01 3.67523432e-01 -5.08063734e-01 2.82806337e-01 -1.38283700e-01 7.99719393e-02 -1.45124957e-01 -1.04049969e+00 1.75794518e+00 -2.84134150e-01 4.66578305e-01 1.14861587e-02 -1.11502111e+00 7.73450553e-01 1.38722390e-01 9.87052739e-01 -1.24706946e-01 6.63167462e-02 -1.11565128e-01 -5.17017096e-02 -3.21058005e-01 1.82842195e-01 6.25179633e-02 -3.39454770e-01 6.43060803e-01 1.15865335e-01 1.94202531e-02 1.58417344e-01 7.16993034e-01 1.04543638e+00 -2.31393427e-01 1.24121837e-01 -5.34780860e-01 4.96203929e-01 3.65355499e-02 5.38611352e-01 9.99188423e-01 -1.97981641e-01 8.09445083e-01 6.39838636e-01 -1.35102496e-01 -1.34054971e+00 -1.18041074e+00 -1.96813002e-01 1.19190538e+00 1.33083224e-01 -6.71446025e-01 -1.00568438e+00 -6.72513664e-01 3.84284370e-02 3.04802716e-01 -8.16650867e-01 -2.16283888e-01 -1.76961541e-01 -7.76010931e-01 1.57680571e-01 3.38898361e-01 2.59962410e-01 -9.97339666e-01 -3.00889909e-01 8.83099958e-02 -2.60244012e-01 -1.25283730e+00 -5.20673573e-01 2.41959199e-01 -6.78327262e-01 -8.59761357e-01 -3.52398843e-01 -9.99698460e-01 1.08685517e+00 1.74253285e-01 1.34438276e+00 3.61662954e-01 -3.39821100e-01 1.00145233e+00 -4.74729300e-01 -7.32816160e-02 -2.31219456e-01 3.35890204e-01 1.81849882e-01 5.76032043e-01 8.08427334e-01 -5.99123597e-01 -7.14400113e-01 1.25342190e-01 -8.78990054e-01 -8.70163292e-02 7.53682077e-01 5.98192453e-01 8.21544170e-01 3.05038601e-01 2.78044939e-01 -1.06786203e+00 3.41602415e-01 -6.85478806e-01 -1.73606381e-01 1.54949903e-01 -6.98754549e-01 1.82811514e-01 5.75429618e-01 -5.74340105e-01 -6.24213457e-01 6.77322090e-01 1.85803846e-01 -3.69748324e-01 -6.88527465e-01 5.10607660e-01 -3.98362517e-01 2.98784286e-01 3.47255349e-01 1.85238317e-01 -4.93012881e-03 -3.97858560e-01 6.20009542e-01 5.68080366e-01 6.96121156e-01 -7.77655602e-01 9.45080757e-01 7.85391152e-01 -3.04060336e-02 -7.99211025e-01 -7.99534798e-01 -7.81344056e-01 -1.01046956e+00 -1.75627708e-01 1.35672903e+00 -9.25744534e-01 -6.73172891e-01 1.98506892e-01 -7.66061306e-01 -2.45971262e-01 -8.34979117e-02 4.41630721e-01 -7.32497752e-01 4.14000809e-01 -5.99955320e-01 -5.31814635e-01 1.75078228e-01 -1.13142169e+00 1.02317095e+00 -1.18809044e-01 -2.90468633e-01 -9.48933840e-01 -1.58427909e-01 1.39916703e-01 1.72680482e-01 2.31415108e-01 1.28926706e+00 -7.88224697e-01 -6.48138940e-01 -9.84378010e-02 -1.90804064e-01 1.96214065e-01 1.20550431e-01 1.69589400e-01 -8.33664834e-01 -2.40999296e-01 -2.14356184e-01 -2.91864008e-01 1.18380141e+00 5.74815989e-01 1.63144124e+00 -3.15285802e-01 -6.70255005e-01 7.04494178e-01 1.54557550e+00 1.05897076e-02 4.19872791e-01 3.44524920e-01 8.50684226e-01 7.67579913e-01 4.21256572e-01 4.44019228e-01 4.95698839e-01 7.56564915e-01 2.92057961e-01 -1.82920381e-01 1.48404911e-01 -4.86722589e-02 2.66449749e-01 5.84084988e-01 1.50379419e-01 -1.47883028e-01 -1.04151154e+00 8.37940633e-01 -1.85958421e+00 -9.32765961e-01 -5.13778776e-02 2.09081411e+00 7.92034745e-01 -1.96748949e-03 2.81029701e-01 2.77939774e-02 8.00448060e-01 -5.72232269e-02 -5.14144897e-01 -7.66119584e-02 -2.25491822e-01 -9.04369354e-02 4.65714335e-01 3.45762819e-01 -1.59679818e+00 9.12736952e-01 6.06709957e+00 6.92733169e-01 -7.15625226e-01 -8.88888687e-02 8.24610353e-01 1.20308325e-02 -4.00046945e-01 8.87934119e-02 -6.77380383e-01 5.00940204e-01 7.54820108e-01 -2.22517997e-02 3.42353731e-01 8.19507480e-01 3.01907957e-02 -9.60313827e-02 -1.45476472e+00 9.55031216e-01 2.66591996e-01 -9.71171796e-01 1.83355138e-01 2.70978898e-01 7.02643633e-01 -2.46115357e-01 4.90629941e-01 -2.08505690e-02 5.35763621e-01 -1.15943694e+00 5.47352791e-01 3.55286837e-01 6.62285030e-01 -7.25572407e-01 4.72813636e-01 8.13727081e-02 -1.01164222e+00 -2.07248509e-01 -2.46935576e-01 2.62395591e-01 -2.97356427e-01 5.90292990e-01 -5.51499009e-01 6.41823560e-02 9.58968878e-01 9.60674524e-01 -9.33417737e-01 8.65127802e-01 -1.58308521e-01 6.77381456e-01 -2.77638018e-01 1.01329722e-01 4.80256945e-01 -2.55484283e-01 2.22204342e-01 1.26899111e+00 2.68469065e-01 2.03672826e-01 5.43945849e-01 7.96118259e-01 -1.48170903e-01 3.18917215e-01 -7.28300691e-01 -1.17847972e-01 6.85875237e-01 1.45717537e+00 -1.13917160e+00 -3.87660503e-01 -4.89359826e-01 9.45467889e-01 2.71215767e-01 3.75138760e-01 -5.94794631e-01 -1.79869950e-01 8.16053391e-01 1.10446312e-01 4.33875084e-01 -4.15587485e-01 -3.57427895e-01 -8.89573812e-01 -2.56297946e-01 -5.91344714e-01 5.46608448e-01 -5.89024842e-01 -1.57131696e+00 2.95823514e-01 -7.96769280e-03 -1.03982341e+00 -3.45553868e-02 -4.70695943e-01 -4.29726005e-01 3.62834573e-01 -1.15211689e+00 -1.30113614e+00 -2.29594484e-01 8.52992356e-01 2.71834165e-01 -1.93100959e-01 8.07554841e-01 2.39264593e-01 -2.94524163e-01 4.40163374e-01 2.36083269e-01 3.34919274e-01 9.67993259e-01 -1.58169734e+00 -2.16759309e-01 4.62206364e-01 4.41879839e-01 7.89681256e-01 7.13457346e-01 -3.21577340e-01 -1.09137273e+00 -1.05486631e+00 9.81058836e-01 -8.79229069e-01 6.04228318e-01 -7.95217335e-01 -7.48666465e-01 6.10829175e-01 3.05856764e-01 -2.55678982e-01 9.70572174e-01 3.54246676e-01 -6.33939147e-01 -2.35354796e-01 -9.52187777e-01 6.41882062e-01 8.70431244e-01 -4.93575275e-01 -1.65235177e-01 4.63060766e-01 7.30717480e-01 3.68418932e-01 -9.77931738e-01 -9.28922892e-02 1.87751144e-01 -7.43180156e-01 1.05700123e+00 -4.55219626e-01 5.45253158e-01 -4.42495823e-01 -2.82116979e-01 -1.16686392e+00 -6.16337240e-01 -2.85153747e-01 3.32878351e-01 1.47147250e+00 4.38820064e-01 -2.05493703e-01 8.54796052e-01 6.87037766e-01 3.60133424e-02 -4.76710051e-01 -5.88441968e-01 -6.94513023e-01 2.15493932e-01 -2.83663124e-01 5.35419762e-01 1.15618253e+00 -8.97093043e-02 4.10993248e-01 -2.75255352e-01 2.14352131e-01 9.66550529e-01 2.82322019e-01 6.15749657e-01 -1.65050077e+00 -3.06411125e-02 -6.04476213e-01 -5.84751546e-01 -5.76462448e-01 5.18758059e-01 -1.11444616e+00 2.09117308e-01 -1.43896151e+00 8.89943123e-01 -7.56810844e-01 -5.66294432e-01 6.17918611e-01 -2.23023921e-01 5.35606325e-01 1.65142968e-01 3.54445308e-01 -1.08916497e+00 1.59862667e-01 6.18183255e-01 -2.88228780e-01 -1.11736596e-01 -3.70049030e-01 -8.46791685e-01 6.25331342e-01 9.64246929e-01 -6.48174882e-01 -3.61955136e-01 -3.26463312e-01 1.73791587e-01 -4.58828539e-01 4.52361703e-01 -1.04472518e+00 5.20386338e-01 -2.51059502e-01 4.74901587e-01 -4.39355910e-01 1.88006237e-01 -9.58749831e-01 -5.26992651e-03 1.54056728e-01 -7.28455245e-01 -1.21183433e-02 -2.75079012e-01 7.66410351e-01 -2.35668033e-01 8.50343611e-03 7.78596759e-01 -2.21648142e-01 -8.36088598e-01 5.31506598e-01 -6.27087951e-01 -1.47241339e-01 1.11042500e+00 -2.76368171e-01 9.08528194e-02 -6.01986766e-01 -9.65555251e-01 3.44046384e-01 9.83688295e-01 3.36200148e-01 4.65450853e-01 -1.41416597e+00 -6.71917379e-01 2.04082150e-02 4.74152565e-01 -2.26140857e-01 2.82099308e-03 8.53471637e-01 1.46599039e-02 3.01791638e-01 -2.19390746e-02 -8.93102527e-01 -1.42620969e+00 1.05296218e+00 2.85171792e-02 -1.26940226e-02 -6.30457520e-01 6.72655344e-01 5.37500918e-01 -3.87173206e-01 3.15756500e-01 1.14260986e-01 -2.51569003e-01 1.99802920e-01 1.56474173e-01 -1.49735853e-01 -2.37524450e-01 -8.12730432e-01 -5.27064860e-01 9.51778948e-01 -1.56150967e-01 -2.07223922e-01 1.40378833e+00 -4.13673729e-01 -4.11453336e-01 5.22793531e-01 1.36064923e+00 -1.51590213e-01 -1.23474872e+00 -4.11193281e-01 3.29979330e-01 -3.41646761e-01 -4.25285399e-02 -6.47109807e-01 -1.25306094e+00 8.42661738e-01 6.14091456e-01 1.54842466e-01 1.25722861e+00 5.37979126e-01 3.55452985e-01 4.23585624e-01 2.15082526e-01 -1.29523349e+00 2.62615353e-01 1.76902235e-01 1.04066879e-01 -1.43839216e+00 -2.91539263e-02 -3.53955835e-01 -7.96172738e-01 8.52888107e-01 4.51398045e-01 -5.29248007e-02 7.66101241e-01 3.82966222e-03 2.68599898e-01 -3.81547987e-01 -7.59343803e-01 -5.41463077e-01 1.84990987e-01 8.39313090e-01 4.81120974e-01 1.34519652e-01 -3.18565160e-01 3.90036345e-01 -1.68866608e-02 -4.75866526e-01 4.35486704e-01 5.53201199e-01 -4.49793965e-01 -1.24475121e+00 -3.42902243e-01 6.97364748e-01 -4.21549529e-01 3.20998468e-02 -6.17551088e-01 6.05997503e-01 3.69144261e-01 1.06196856e+00 4.46915090e-01 -4.06287432e-01 -2.14474812e-01 1.57879457e-01 2.01951593e-01 -7.97886670e-01 -3.27442348e-01 4.40018587e-02 -4.70382035e-01 -5.11785865e-01 -7.47654915e-01 -9.83987153e-01 -1.13592839e+00 -2.23021120e-01 1.71758845e-01 2.57012695e-01 6.08330667e-01 6.68370545e-01 2.01339900e-01 2.23284960e-02 5.99711120e-01 -6.10685825e-01 2.16630697e-02 -4.27123606e-01 -7.12958097e-01 1.10144854e+00 2.92240947e-01 -5.05882323e-01 -4.08161223e-01 4.92482156e-01]	[9.288463592529297, 3.008788585662842]
c4e88a07-ddec-4753-9232-f0ccbdec0ea7	self-supervised-learning-of-a-tailored	2303.11837	null	https://arxiv.org/abs/2303.11837v1	https://arxiv.org/pdf/2303.11837v1.pdf	Self-supervised learning of a tailored Convolutional Auto Encoder for histopathological prostate grading	According to GLOBOCAN 2020, prostate cancer is the second most common cancer in men worldwide and the fourth most prevalent cancer overall. For pathologists, grading prostate cancer is challenging, especially when discriminating between Grade 3 (G3) and Grade 4 (G4). This paper proposes a Self-Supervised Learning (SSL) framework to classify prostate histopathological images when labeled images are scarce. In particular, a tailored Convolutional Auto Encoder (CAE) is trained to reconstruct 128x128x3 patches of prostate cancer Whole Slide Images (WSIs) as a pretext task. The downstream task of the proposed SSL paradigm is the automatic grading of histopathological patches of prostate cancer. The presented framework reports promising results on the validation set, obtaining an overall accuracy of 83% and on the test set, achieving an overall accuracy value of 76% with F1-score of 77% in G4.	['Valery Naranjo', 'Javier Oliver', 'Kjersti Engan', 'Adrian colomer', 'Zahra Tabatabaei']	2023-03-21	null	null	null	null	['whole-slide-images']	['computer-vision']	[ 3.05709302e-01 5.88649869e-01 -1.19441107e-01 -4.19829041e-01 -1.05857265e+00 -5.73520482e-01 7.03826487e-01 6.68219090e-01 -4.85742956e-01 6.24100208e-01 -1.06710248e-01 -2.36402959e-01 -7.81344846e-02 -8.13890874e-01 -3.21696550e-01 -9.73982036e-01 -1.51150912e-01 6.46984041e-01 -6.61252439e-02 4.21459526e-01 1.90065533e-01 6.44188344e-01 -9.13504303e-01 2.44034782e-01 8.78720343e-01 1.00557160e+00 1.68951795e-01 1.08916104e+00 5.50320698e-03 6.76539361e-01 -5.21940768e-01 -7.19821751e-01 2.02046350e-01 -1.49266034e-01 -7.51218736e-01 -2.18330044e-02 6.42606020e-01 -9.20634121e-02 -1.43918276e-01 1.17860007e+00 3.20569903e-01 -6.38549626e-01 9.14506316e-01 -7.47905433e-01 -2.58676231e-01 3.28857720e-01 -3.69088739e-01 6.85084164e-02 -3.31771076e-01 2.63960987e-01 8.13778102e-01 -5.91313124e-01 7.54110277e-01 5.36717176e-01 6.64670706e-01 7.54792929e-01 -1.27762318e+00 -4.56028879e-01 -6.97077930e-01 1.81699857e-01 -1.22232831e+00 -1.56004816e-01 6.27735913e-01 -3.86772245e-01 2.51209199e-01 5.01156330e-01 5.34482896e-01 6.50044739e-01 1.05780768e+00 6.23228729e-01 1.39516878e+00 -1.89965442e-02 6.30539000e-01 1.13467842e-01 3.88187803e-02 5.50029039e-01 5.14825284e-01 -1.43585920e-01 8.99161038e-04 1.90337785e-02 7.26189077e-01 1.08567864e-01 3.12184125e-01 -1.14419550e-01 -7.63439119e-01 6.91172123e-01 9.36890006e-01 5.17284751e-01 -2.93126523e-01 6.24411292e-02 6.25716329e-01 -1.12222694e-01 2.13704526e-01 1.44306332e-01 1.84096545e-01 3.31821948e-01 -1.12641394e+00 1.90330297e-01 5.15932202e-01 3.91353637e-01 4.42609608e-01 -3.43663126e-01 -2.04553932e-01 6.36594594e-01 2.87088245e-01 5.83272219e-01 6.63407385e-01 -5.58800220e-01 -7.01389983e-02 9.27157342e-01 -1.73082560e-01 -7.28427052e-01 -6.38654053e-01 -1.00643933e+00 -1.47398078e+00 2.65124977e-01 4.74413484e-01 1.88243479e-01 -9.23423946e-01 1.25707233e+00 1.09966815e-01 -1.41705960e-01 1.30331084e-01 8.49756539e-01 8.84000897e-01 4.18671548e-01 4.96403515e-01 1.87681183e-01 1.56624055e+00 -6.63989425e-01 -3.18021148e-01 -4.68677968e-01 4.07282442e-01 -4.17426616e-01 5.17210126e-01 1.17853865e-01 -8.25821280e-01 -5.08544981e-01 -1.04821742e+00 -1.07472494e-01 -1.93405405e-01 8.19448352e-01 7.33540833e-01 7.08248734e-01 -1.26818860e+00 3.96528542e-01 -1.03931534e+00 -5.40577114e-01 7.63840377e-01 5.22857726e-01 -8.46522689e-01 -2.18317240e-01 -5.14177084e-01 5.54121435e-01 1.66674137e-01 2.91110694e-01 -1.19352186e+00 -9.50593948e-01 -7.39762485e-01 3.99797499e-01 -3.24101627e-01 -4.18671012e-01 7.91347384e-01 -5.84443152e-01 -1.03097260e+00 1.22747552e+00 8.42334479e-02 -6.81497157e-01 6.59877300e-01 5.33448219e-01 -2.69050896e-01 2.50824600e-01 1.43342197e-01 6.75717592e-01 3.15891653e-01 -1.15306342e+00 -7.59927392e-01 -8.72240186e-01 -3.85934114e-01 -2.97934674e-02 -2.53782004e-01 -6.77993178e-01 -2.46888280e-01 -2.29732797e-01 -9.43334103e-02 -1.06868207e+00 -5.25228500e-01 -8.16463381e-02 -6.38538718e-01 -1.18190438e-01 4.85892832e-01 -8.31872284e-01 7.14892149e-01 -2.33486915e+00 -9.74556506e-02 3.75330120e-01 2.68689722e-01 2.68662453e-01 8.16676393e-02 7.53072947e-02 -4.06554714e-02 2.14041434e-02 -2.35746667e-01 -4.75757569e-01 1.16745345e-02 1.20382614e-01 2.04993010e-01 6.26169682e-01 5.79350114e-01 9.05646026e-01 -9.67978716e-01 -5.09081721e-01 3.23129714e-01 5.09128034e-01 -2.08011806e-01 2.13545009e-01 3.22680563e-01 6.08729184e-01 -3.86859238e-01 7.68087089e-01 8.04745853e-01 -4.59164351e-01 4.93295640e-01 -2.23209694e-01 9.72057283e-02 -3.03429395e-01 -6.83427751e-01 1.47716331e+00 -3.54877502e-01 6.64099872e-01 1.47651628e-01 -6.55284166e-01 1.17183352e+00 3.52700092e-02 5.89909792e-01 -7.58507013e-01 -4.93656285e-02 4.22780961e-01 7.63713047e-02 -3.17788064e-01 2.86622494e-01 -2.14054763e-01 -4.55566227e-01 -7.28758648e-02 1.11930050e-01 -2.13311563e-04 4.39925611e-01 1.83353275e-01 1.83879435e+00 -4.86928552e-01 4.35728580e-01 -6.20166004e-01 1.01465082e+00 2.39331722e-01 5.96540809e-01 3.92842948e-01 -5.62735200e-01 3.99197489e-01 7.42352545e-01 -6.42807603e-01 -1.12130773e+00 -1.25536191e+00 -4.81805056e-01 2.01440632e-01 -8.27206299e-02 2.05729399e-02 -6.66861117e-01 -7.16977954e-01 1.24961585e-01 2.47273475e-01 -7.82055914e-01 9.24218893e-02 -4.24929380e-01 -7.93367624e-01 5.34946740e-01 5.55782139e-01 7.36838043e-01 -9.24970329e-01 -3.92052352e-01 7.43648112e-02 2.10305676e-01 -8.57275665e-01 9.75093022e-02 3.56435537e-01 -9.63742018e-01 -1.32634854e+00 -9.66329813e-01 -9.52644110e-01 1.13126087e+00 -3.18952471e-01 8.59435141e-01 -1.43050313e-01 -8.98312807e-01 -2.12775603e-01 2.04344496e-01 -1.24042206e-01 -6.16157353e-01 1.72164217e-01 -5.17349064e-01 9.57585275e-02 3.20844889e-01 -1.20472901e-01 -9.27872181e-01 -1.99089333e-01 -9.74101901e-01 -8.42905194e-02 1.35642278e+00 7.39062786e-01 7.21283197e-01 1.13484509e-01 4.77720201e-01 -1.22056186e+00 2.28086382e-01 -2.88309306e-01 -5.39036393e-01 3.70668583e-02 -4.32078689e-01 -2.38535374e-01 9.89683509e-01 2.63482690e-01 -9.17248487e-01 4.17200118e-01 -3.40034902e-01 1.29992828e-01 -4.07688022e-01 3.67343783e-01 -1.80301890e-02 -6.97064847e-02 4.88086611e-01 3.91238391e-01 1.61906019e-01 -1.85450643e-01 -3.16718072e-01 6.79124832e-01 9.65102494e-01 4.76284549e-02 5.60632288e-01 6.43135428e-01 4.68359500e-01 -5.94622254e-01 -5.38119018e-01 -7.03702211e-01 -4.71267372e-01 -3.90628517e-01 9.20269608e-01 -9.06944036e-01 -8.53484929e-01 4.99449790e-01 -4.45097893e-01 -4.97675300e-01 -1.54959157e-01 1.65290684e-01 -3.23333591e-01 1.60249069e-01 -9.95801032e-01 -5.19612789e-01 -5.99058032e-01 -1.15942204e+00 1.37574041e+00 6.78701758e-01 -3.09792440e-02 -1.09250283e+00 3.04955363e-01 4.24457520e-01 7.11624861e-01 4.52456474e-01 1.02068746e+00 -7.32464373e-01 -3.62582058e-01 -7.58735895e-01 -5.57290792e-01 -3.03680915e-03 2.07730740e-01 -3.06448728e-01 -1.11388171e+00 -5.39330781e-01 -3.32914740e-01 -2.10299045e-01 8.01794410e-01 4.58238959e-01 9.88412738e-01 -6.22590855e-02 -4.85516280e-01 2.53746837e-01 2.07114196e+00 9.48065668e-02 7.48441279e-01 2.44706571e-01 2.39975363e-01 3.90779883e-01 3.23712379e-01 2.71620780e-01 4.34684724e-01 2.51995116e-01 6.38015568e-01 -1.84931070e-01 -4.82288331e-01 7.96803311e-02 -1.47829503e-02 3.67936850e-01 1.55251682e-01 8.06300640e-02 -1.14852858e+00 7.63191402e-01 -1.20612252e+00 -6.39704645e-01 -1.30208164e-01 2.30057192e+00 6.71367705e-01 8.46143812e-02 -3.91017616e-01 2.17979223e-01 6.73349082e-01 -3.04265786e-02 -4.86729771e-01 -2.84681499e-01 3.52198809e-01 3.39089930e-01 6.46709085e-01 3.66893470e-01 -1.40468001e+00 3.10857981e-01 5.29140282e+00 6.93947554e-01 -1.41510499e+00 -4.98480760e-02 1.39708877e+00 2.71585464e-01 1.99202508e-01 -2.82234490e-01 -7.47459650e-01 6.27116323e-01 1.07060385e+00 -1.48476437e-01 -1.75179362e-01 8.04319799e-01 -5.75061813e-02 -2.62162149e-01 -9.96979177e-01 5.97386837e-01 7.68534169e-02 -1.30756080e+00 -3.22118282e-01 4.26562965e-01 7.21993983e-01 -5.61312102e-02 8.09474066e-02 4.24042910e-01 1.44054890e-01 -1.19256949e+00 -1.79862767e-01 5.59021235e-01 1.03791738e+00 -7.82549322e-01 1.30343747e+00 1.80809632e-01 -1.01544523e+00 -1.53078064e-01 -3.67155224e-01 6.26395047e-01 -3.99967611e-01 7.75344789e-01 -1.22914076e+00 5.33030331e-01 5.81300139e-01 4.83657837e-01 -1.05540454e+00 1.19564092e+00 1.23150665e-02 5.20014226e-01 -6.67548403e-02 -1.67008787e-02 2.28111714e-01 2.73013651e-01 1.81500480e-01 1.43062925e+00 4.15457726e-01 -4.08715189e-01 -1.77116934e-02 5.51651657e-01 3.86826992e-02 -2.12217402e-02 -2.22060010e-01 8.77557546e-02 1.01911835e-01 2.01046395e+00 -1.10927749e+00 -1.97583154e-01 1.68476701e-01 7.94718564e-01 1.90701112e-01 -1.31147206e-01 -4.22702670e-01 -3.60043138e-01 -6.46250844e-02 1.91247575e-02 1.11703709e-01 3.50114971e-01 -2.38593519e-01 -6.58344448e-01 -2.86259115e-01 -4.16504890e-01 4.11479592e-01 -2.60415763e-01 -1.30418801e+00 5.94011426e-01 -6.88730121e-01 -1.16761351e+00 -2.77468145e-01 -9.12676096e-01 -7.47057259e-01 4.96219426e-01 -1.70095778e+00 -1.38324463e+00 -7.70351291e-01 2.37057023e-02 3.19469988e-01 -7.99218044e-02 1.14534140e+00 2.89776742e-01 -4.51151133e-01 4.92505103e-01 3.66133571e-01 5.04924178e-01 7.86818862e-01 -1.89861596e+00 -1.39887735e-01 5.41940570e-01 -6.77294075e-01 2.96068937e-01 4.45970833e-01 -6.77256167e-01 -1.62732136e+00 -1.58812129e+00 1.10934806e+00 9.66358632e-02 4.80367601e-01 -1.14436038e-01 -4.30615008e-01 3.57611299e-01 5.27921766e-02 4.57228631e-01 9.42305684e-01 -3.19497347e-01 1.93204451e-02 -4.59878892e-01 -1.58540940e+00 2.84104049e-01 3.69600616e-02 -2.29042381e-01 2.04386026e-01 2.79465616e-01 -4.21138644e-01 -3.92152756e-01 -1.45722568e+00 4.28234339e-01 5.36548376e-01 -9.46940899e-01 8.49741876e-01 1.97539240e-01 9.54131722e-01 -3.24909031e-01 1.13288499e-01 -1.27736676e+00 -4.19380933e-01 1.79886237e-01 2.84399122e-01 1.05939996e+00 2.09463492e-01 -1.20945759e-01 1.43390381e+00 4.79177207e-01 -1.17970847e-01 -7.10126638e-01 -1.16553271e+00 -3.80682290e-01 1.10485576e-01 2.84349293e-01 1.86889231e-01 6.22117996e-01 -1.75470799e-01 -3.92711237e-02 3.28489065e-01 1.97260067e-01 1.05441701e+00 1.18296288e-01 6.31618083e-01 -1.32042599e+00 -1.95091352e-01 -5.14667690e-01 -1.09160554e+00 6.16331398e-03 -6.97587878e-02 -1.14638054e+00 -5.04567437e-02 -1.51088619e+00 7.56308019e-01 -5.38402140e-01 -5.14386475e-01 4.23912942e-01 -1.59090966e-01 9.30744231e-01 9.42359716e-02 2.01455310e-01 -5.62318683e-01 -2.59265918e-02 1.27760959e+00 -6.64058328e-01 2.83409119e-01 -2.19533723e-02 -6.79137886e-01 3.44041944e-01 6.71128392e-01 -1.46454021e-01 1.06059097e-01 1.32226765e-01 -1.33916318e-01 3.15798491e-01 5.45155764e-01 -1.35605717e+00 3.25534075e-01 1.71690598e-01 9.85799491e-01 -6.78652227e-01 2.48657674e-01 -7.94182479e-01 4.11653668e-01 1.19362390e+00 -5.05552948e-01 -4.92479563e-01 1.86955363e-01 5.12388468e-01 -2.20929593e-01 -3.30018312e-01 7.19256818e-01 -4.37549874e-02 -5.98394096e-01 2.03575775e-01 -2.64514059e-01 -6.68123484e-01 1.11635029e+00 8.29391181e-02 -5.93983591e-01 1.57997876e-01 -8.02830994e-01 1.43400416e-01 4.58827615e-01 -7.51124769e-02 4.10757989e-01 -1.28300130e+00 -1.09539878e+00 1.77133411e-01 4.08004194e-01 1.70657992e-01 6.06515527e-01 7.51628578e-01 -7.95395732e-01 6.76908672e-01 -4.50348109e-01 -7.92878032e-01 -1.27092385e+00 -1.13620171e-02 2.49033123e-01 -1.20963502e+00 -5.78054190e-01 7.99651206e-01 5.63973561e-02 -4.51800406e-01 4.67139333e-02 2.59409100e-02 -5.49142718e-01 -2.83673942e-01 6.80805922e-01 1.28780305e-01 4.48738605e-01 -3.98150325e-01 -3.48504424e-01 9.91793126e-02 -5.17814875e-01 3.73165011e-01 1.44282484e+00 4.26625937e-01 -2.47493595e-01 2.79542524e-02 1.17799366e+00 1.00748695e-01 -1.15749574e+00 1.08263895e-01 1.40413180e-01 -3.34207684e-01 1.11095384e-01 -1.02112091e+00 -1.03290749e+00 6.28046036e-01 1.10894930e+00 1.56395987e-01 1.03331292e+00 -7.97153115e-02 4.84618813e-01 7.59073496e-02 2.45802730e-01 -6.15130246e-01 -2.15024784e-01 -3.66428606e-02 5.45530796e-01 -1.30684555e+00 7.12748840e-02 -2.13837773e-01 -4.85582411e-01 1.24658346e+00 2.73073822e-01 -7.78134763e-01 2.69846708e-01 4.66847360e-01 1.09113283e-01 -2.75139689e-01 -6.48999810e-01 2.16400146e-01 2.68616043e-02 4.00081009e-01 8.09339941e-01 4.53129917e-01 -4.79641914e-01 6.79257751e-01 -2.29412824e-01 1.84000790e-01 3.97656113e-01 8.11155915e-01 -2.62029707e-01 -9.88045454e-01 -2.02292323e-01 8.27657878e-01 -7.25375593e-01 4.05408025e-01 -4.17073965e-01 7.20225692e-01 -1.10013440e-01 5.50442219e-01 3.63905877e-01 -4.92015742e-02 -1.65028751e-01 -2.03578770e-01 3.22104067e-01 -4.13323194e-01 -8.72471988e-01 -1.75904075e-03 -2.40692183e-01 -2.85338491e-01 -2.26483330e-01 -5.58132946e-01 -1.20362687e+00 -1.82516277e-01 7.49098361e-02 1.49324268e-01 1.11207783e+00 8.05572748e-01 1.88940376e-01 8.52886081e-01 9.17144120e-01 -6.70997202e-01 -4.45976943e-01 -9.29039240e-01 -6.75831199e-01 4.36780125e-01 2.54604667e-01 -2.86575663e-03 -2.05151796e-01 1.44697458e-01]	[15.052544593811035, -2.895550012588501]
45392ec9-eb73-43cc-839a-444aaf5ddf1a	advest-adversarial-perturbation-estimation-to	2204.03848	null	https://arxiv.org/abs/2204.03848v1	https://arxiv.org/pdf/2204.03848v1.pdf	AdvEst: Adversarial Perturbation Estimation to Classify and Detect Adversarial Attacks against Speaker Identification	Adversarial attacks pose a severe security threat to the state-of-the-art speaker identification systems, thereby making it vital to propose countermeasures against them. Building on our previous work that used representation learning to classify and detect adversarial attacks, we propose an improvement to it using AdvEst, a method to estimate adversarial perturbation. First, we prove our claim that training the representation learning network using adversarial perturbations as opposed to adversarial examples (consisting of the combination of clean signal and adversarial perturbation) is beneficial because it eliminates nuisance information. At inference time, we use a time-domain denoiser to estimate the adversarial perturbations from adversarial examples. Using our improved representation learning approach to obtain attack embeddings (signatures), we evaluate their performance for three applications: known attack classification, attack verification, and unknown attack detection. We show that common attacks in the literature (Fast Gradient Sign Method (FGSM), Projected Gradient Descent (PGD), Carlini-Wagner (CW) with different Lp threat models) can be classified with an accuracy of ~96%. We also detect unknown attacks with an equal error rate (EER) of ~9%, which is absolute improvement of ~12% from our previous work.	['Najim Dehak', 'Jesus Villalba', 'Saurabh Kataria', 'Sonal Joshi']	2022-04-08	null	null	null	null	['speaker-identification']	['speech']	[ 4.37180161e-01 1.20358638e-01 2.38050863e-01 -3.26220877e-02 -1.09474659e+00 -1.11299717e+00 7.01683283e-01 1.90022904e-02 -1.49133757e-01 5.37847757e-01 3.91711533e-01 -7.43556857e-01 7.01707182e-03 -5.18464088e-01 -6.99173212e-01 -7.67633736e-01 -5.11212051e-01 1.46664202e-01 6.93576527e-04 -6.89220130e-01 1.34398937e-01 7.68962324e-01 -1.14048338e+00 2.78062224e-01 4.83267903e-01 8.34698439e-01 -9.16582763e-01 1.15191174e+00 1.65461063e-01 5.77409625e-01 -1.25532258e+00 -7.96131968e-01 6.49697661e-01 -2.59819478e-01 -6.96837902e-01 -6.78622127e-01 5.37356615e-01 -2.84676403e-01 -7.37230062e-01 1.24842918e+00 9.27342713e-01 8.24001282e-02 7.12799430e-01 -1.32842457e+00 -6.45875871e-01 7.38882422e-01 -3.49100858e-01 4.61265385e-01 7.89293110e-01 6.83586895e-02 4.66047764e-01 -5.12637973e-01 2.31730178e-01 1.43470287e+00 8.81456017e-01 1.04062104e+00 -1.10509193e+00 -1.09313393e+00 1.77640527e-01 2.86320806e-01 -1.37315881e+00 -7.72636712e-01 1.07752693e+00 -8.25874433e-02 6.64376080e-01 7.84024954e-01 -1.06396310e-01 1.79484606e+00 -1.72053173e-01 5.24298429e-01 9.89462614e-01 -6.38762832e-01 2.58516759e-01 2.92402208e-01 -3.64810601e-02 3.01156759e-01 -9.24602896e-03 7.42875993e-01 -3.86008710e-01 -6.85239911e-01 5.60192131e-02 -2.19351828e-01 -6.15951121e-01 2.30640948e-01 -5.38888276e-01 8.90505612e-01 3.82818490e-01 3.15670699e-01 -2.15216160e-01 -9.65879858e-03 6.81689560e-01 7.36307085e-01 3.06921124e-01 3.80764693e-01 -4.27708119e-01 -7.87568614e-02 -6.86613083e-01 1.42429784e-01 1.21285391e+00 3.94225180e-01 1.11133575e-01 8.18746746e-01 2.69553959e-02 7.68278718e-01 2.59717554e-01 7.21907973e-01 6.90256655e-01 -3.05048645e-01 5.74319661e-01 -1.11783996e-01 -1.57202244e-01 -8.57133567e-01 -6.13833554e-02 -4.42127734e-01 -8.21594775e-01 4.34318781e-01 4.56501096e-01 -5.10081470e-01 -7.87888646e-01 1.78079224e+00 2.34409228e-01 7.39634275e-01 5.01093030e-01 4.64996397e-01 3.22233379e-01 6.15671813e-01 -9.94042680e-02 -2.22155631e-01 1.00421536e+00 -4.54443932e-01 -6.33393884e-01 -9.11821797e-02 2.46213138e-01 -9.38845992e-01 6.54088557e-01 6.08951688e-01 -7.94603705e-01 -3.43134552e-01 -1.42607915e+00 6.74525678e-01 -6.88772678e-01 -4.69556451e-01 2.92180866e-01 1.75092375e+00 -9.46016669e-01 5.84184051e-01 -3.72309059e-01 -2.49780398e-02 2.32685953e-01 5.57713270e-01 -4.88797873e-01 1.60021871e-01 -1.69701409e+00 9.18760836e-01 1.84076577e-02 7.37596974e-02 -1.23786724e+00 -8.28742743e-01 -7.73593068e-01 -2.19205067e-01 -3.08366157e-02 -1.56925116e-02 1.09258032e+00 -8.13164890e-01 -1.88404930e+00 6.13730490e-01 3.16698104e-02 -9.26838934e-01 4.43758756e-01 -1.52950913e-01 -1.20990229e+00 1.74777433e-01 -6.68960512e-01 -2.85502315e-01 1.57748413e+00 -1.19326246e+00 -1.63770333e-01 -3.69131327e-01 1.22961961e-01 -2.30035439e-01 -5.78784645e-01 4.79767859e-01 3.52176279e-01 -9.31396008e-01 -1.02510437e-01 -8.55011225e-01 5.78278536e-03 -3.19893271e-01 -3.95384640e-01 2.79640973e-01 1.23457670e+00 -1.10008657e+00 1.20345235e+00 -2.24720669e+00 2.37859823e-02 5.40898144e-01 -8.71015638e-02 8.34979713e-01 -4.22168374e-02 5.57488263e-01 -6.89342201e-01 2.23543927e-01 -4.40767854e-01 -3.07119548e-01 3.08272034e-01 8.54840800e-02 -1.07313168e+00 8.61311197e-01 9.93039459e-02 5.31771183e-01 -6.91332638e-01 1.67937323e-01 2.61775464e-01 7.55667031e-01 -4.27710623e-01 3.66710991e-01 2.51043409e-01 4.07583177e-01 -1.78902000e-01 6.61870897e-01 8.99174869e-01 8.89838398e-01 3.75956409e-02 -7.44896531e-02 3.47201586e-01 3.69538099e-01 -1.52006352e+00 1.02195799e+00 -6.48033738e-01 6.62688494e-01 3.92877936e-01 -1.28060675e+00 9.29091930e-01 7.48111248e-01 7.27961883e-02 -2.01004654e-01 2.25618228e-01 1.44648701e-01 1.21026896e-02 -9.63652283e-02 -1.14162147e-01 -1.99094057e-01 -2.11240113e-01 4.24641222e-01 2.04197481e-01 1.64309129e-01 -5.11197090e-01 2.23936513e-01 1.34638250e+00 -4.33618426e-01 2.01280460e-01 -8.27323422e-02 1.10557997e+00 -7.99886048e-01 3.79475832e-01 8.56946945e-01 -4.76173520e-01 2.99232244e-01 3.55540931e-01 -2.27038562e-01 -5.70833623e-01 -1.37862289e+00 -1.40538901e-01 1.06948781e+00 -3.01179647e-01 -2.67196774e-01 -7.98914731e-01 -1.01676202e+00 8.18570405e-02 7.00311840e-01 -5.34235120e-01 -6.64960563e-01 -9.99976575e-01 -7.46677458e-01 1.41782641e+00 4.86528128e-01 2.28746966e-01 -6.93468928e-01 2.67064959e-01 4.17743772e-02 2.76974529e-01 -1.13898790e+00 -5.78708410e-01 3.94290626e-01 -3.25347215e-01 -1.00079381e+00 -3.84974360e-01 -6.47915006e-01 5.99208057e-01 -2.60981470e-01 7.85649955e-01 -1.39126405e-01 -2.67381370e-01 5.28677046e-01 -4.95913655e-01 -5.96855819e-01 -1.03924382e+00 -4.83692020e-01 7.68299043e-01 2.29337126e-01 3.01970780e-01 -8.40419471e-01 -2.70230144e-01 2.20714927e-01 -9.10688400e-01 -1.14335465e+00 2.76329517e-01 8.12150061e-01 -1.27273738e-01 2.18390062e-01 7.71483362e-01 -8.24202538e-01 7.38394916e-01 -5.60425878e-01 -4.04064924e-01 2.11895443e-02 -4.05006289e-01 1.11660741e-01 1.14035940e+00 -9.34773028e-01 -7.61695623e-01 -1.65322915e-01 -8.14363480e-01 -6.15756631e-01 -2.18096092e-01 -5.24606407e-02 -3.68863881e-01 -7.33870983e-01 1.17972493e+00 3.25958580e-01 -1.81651264e-01 -4.37052280e-01 5.24587214e-01 8.16617846e-01 8.19003761e-01 -6.01242065e-01 1.62463307e+00 2.29188919e-01 -9.21714827e-02 -8.34767044e-01 -4.01430339e-01 -2.49446064e-01 -3.59892726e-01 -6.44931346e-02 3.30617130e-01 -6.70437098e-01 -9.33655143e-01 5.09680152e-01 -9.14042115e-01 1.10996768e-01 -6.67075366e-02 3.58792245e-01 -2.55623281e-01 7.84975648e-01 -6.11318946e-01 -1.23570991e+00 -5.97263277e-01 -9.82307255e-01 7.54611611e-01 -1.74052030e-01 -4.56333123e-02 -1.13950646e+00 4.74672355e-02 1.60065442e-01 7.84477174e-01 6.61988437e-01 6.02615118e-01 -1.32811022e+00 2.12365702e-01 -7.54044831e-01 4.52671468e-01 8.59721661e-01 2.00357810e-01 -1.64815322e-01 -1.52152359e+00 -6.66084230e-01 5.53091645e-01 -1.34133428e-01 5.10365546e-01 -6.63763210e-02 1.00211132e+00 -7.74193108e-01 -4.01359983e-02 7.35744178e-01 1.21967161e+00 2.30988905e-01 7.23480046e-01 -2.62283627e-02 5.52670538e-01 3.63096207e-01 -7.06816092e-02 3.27797502e-01 -3.72105986e-01 8.83856773e-01 5.29648185e-01 2.98958123e-01 -1.84793904e-01 -2.36982062e-01 1.06956029e+00 6.91743612e-01 2.34063759e-01 -1.30439684e-01 -7.20168710e-01 2.18530715e-01 -1.13680911e+00 -1.22002256e+00 3.59399110e-01 2.60225534e+00 7.54730105e-01 4.59901243e-01 2.68518955e-01 9.81474876e-01 7.15362608e-01 3.05339932e-01 -4.03339237e-01 -1.02378702e+00 -1.05510786e-01 8.14412057e-01 6.70618415e-01 8.05550754e-01 -1.32297528e+00 8.30888748e-01 6.42552090e+00 7.04910994e-01 -1.28798270e+00 1.59871846e-01 1.28580645e-01 1.78249493e-01 4.34630848e-02 -3.26485574e-01 -7.18598962e-01 4.70102757e-01 1.61037564e+00 -2.47049466e-01 8.45265865e-01 7.45419979e-01 -2.40883023e-01 9.26652193e-01 -1.09093106e+00 9.23833549e-01 5.25854886e-01 -9.00816739e-01 5.37038967e-02 -5.74152805e-02 4.06714976e-01 -3.67188424e-01 4.58329201e-01 6.62389815e-01 6.70985281e-01 -1.06166351e+00 3.61529469e-01 1.22307643e-01 6.21068060e-01 -1.01753449e+00 7.55786002e-01 1.82286769e-01 -1.20271075e+00 -3.22804064e-01 -2.62320399e-01 4.64311503e-02 -6.57973811e-02 2.23607942e-01 -1.01539218e+00 6.76973462e-01 3.21297348e-01 1.47127345e-01 -1.84171990e-01 4.19639736e-01 -3.38314384e-01 1.11510384e+00 -3.99521708e-01 2.41213486e-01 4.86468412e-02 3.34406644e-01 9.71502960e-01 1.21137536e+00 7.20488727e-02 -1.92119792e-01 1.27679342e-03 3.30480456e-01 -2.74112970e-01 -1.06181368e-01 -7.11839557e-01 1.07367218e-01 7.75936365e-01 8.53155315e-01 1.21552549e-01 7.98792318e-02 -8.74199066e-03 1.16288269e+00 2.74346787e-02 3.65814954e-01 -8.98653984e-01 -6.49399877e-01 1.12096977e+00 -2.38506943e-01 2.96699822e-01 1.98703185e-02 1.94212452e-01 -1.00259483e+00 1.07051156e-01 -1.38799059e+00 4.53336209e-01 3.93649600e-02 -1.62899160e+00 7.50378549e-01 -2.07668260e-01 -1.32727659e+00 -6.68407023e-01 -7.65140295e-01 -1.03600430e+00 1.12457705e+00 -1.25965798e+00 -1.06589413e+00 3.18145543e-01 8.94104719e-01 2.77998835e-01 -6.16583407e-01 1.27720881e+00 3.87795031e-01 -6.11337006e-01 1.45970488e+00 1.65767923e-01 4.78580147e-01 6.63689613e-01 -1.35516429e+00 8.29107046e-01 1.12376964e+00 3.76397520e-01 6.89699173e-01 8.99634302e-01 -1.87306017e-01 -1.67256248e+00 -1.15987289e+00 3.50470096e-01 -5.85777760e-01 8.88022542e-01 -6.46555901e-01 -9.85805273e-01 5.94670951e-01 3.38667408e-02 1.68869048e-01 8.85951102e-01 -7.32209906e-02 -1.08068442e+00 -2.15071023e-01 -1.71778941e+00 5.67600548e-01 7.30520308e-01 -9.71926987e-01 -8.44191968e-01 4.35101062e-01 8.39016974e-01 -3.63195807e-01 -7.81584382e-01 3.94155741e-01 4.02620345e-01 -8.31868947e-01 1.49863565e+00 -8.64795089e-01 -4.21596199e-01 -3.20683151e-01 -4.95366931e-01 -1.38778090e+00 7.48254806e-02 -1.18977988e+00 -7.39673793e-01 1.34469366e+00 3.42864931e-01 -1.00210869e+00 5.58368266e-01 1.41579211e-01 -8.87905210e-02 -4.00854081e-01 -1.21158719e+00 -1.03980815e+00 3.66261929e-01 -7.43759751e-01 7.53464639e-01 1.09454226e+00 1.48154525e-02 -1.62897274e-01 -6.00349009e-01 1.03113699e+00 7.57548273e-01 -6.45347774e-01 6.51747823e-01 -1.03712177e+00 -4.64510918e-01 -3.24628949e-01 -9.64902282e-01 -5.35217524e-01 6.31830931e-01 -8.20856392e-01 -3.30119878e-01 -5.79246938e-01 -6.89454794e-01 -6.69643506e-02 -6.16711497e-01 2.83067346e-01 -1.65328756e-01 3.62629205e-01 -3.12236920e-02 -2.26599947e-01 1.10970199e-01 1.52791828e-01 1.96166128e-01 -4.03451055e-01 7.77612776e-02 3.97244573e-01 -8.37049365e-01 6.76787734e-01 8.69110286e-01 -4.70180303e-01 -2.06316039e-01 7.69793764e-02 -3.13608795e-01 -1.36039972e-01 1.98471099e-01 -1.32883215e+00 9.57361832e-02 3.32065523e-01 3.22857350e-01 7.18751922e-02 4.61796671e-01 -7.84838200e-01 -1.94880575e-01 6.71626806e-01 -4.65486437e-01 -3.83559875e-02 4.38179433e-01 6.59111559e-01 -1.95128173e-01 -3.22303891e-01 9.73565996e-01 3.10405701e-01 -3.38392854e-01 1.56325072e-01 -2.36132309e-01 -3.39727756e-03 9.52548087e-01 1.38559863e-01 -4.23622102e-01 -5.97911060e-01 -7.40741491e-01 -3.47353250e-01 2.71846633e-02 4.37946469e-01 7.20262825e-01 -1.18852413e+00 -1.05246115e+00 5.61476469e-01 -1.15216821e-01 -9.18375790e-01 2.32110426e-01 2.44045585e-01 -2.80913323e-01 1.71060875e-01 7.38139674e-02 -4.08795588e-02 -1.55338395e+00 9.12634730e-01 4.23821628e-01 -1.57255754e-01 -3.46721202e-01 1.06360638e+00 -3.46054584e-01 -5.85170448e-01 6.24637365e-01 1.37414709e-01 -1.55436590e-01 -2.79865831e-01 9.07277644e-01 4.04163629e-01 3.77575040e-01 -8.23852181e-01 -5.43824613e-01 4.29551661e-01 -2.26418078e-01 -5.93950562e-02 8.65213871e-01 4.06203151e-01 2.06883505e-01 -1.30979680e-02 1.57801712e+00 6.66888714e-01 -5.90776920e-01 -1.75655618e-01 -1.34524181e-01 -4.78700280e-01 5.68173639e-02 -7.79072106e-01 -9.08411801e-01 8.97594929e-01 1.09314322e+00 6.36957347e-01 1.20154035e+00 -2.98638672e-01 7.68369615e-01 2.59216607e-01 2.07798675e-01 -4.87157732e-01 -2.28204820e-02 4.20821249e-01 1.00852644e+00 -9.11772609e-01 -1.36191234e-01 -3.19480807e-01 -4.45656776e-01 9.07221019e-01 1.57160416e-01 -2.31711626e-01 8.85017991e-01 5.13966322e-01 2.59167045e-01 3.71096939e-01 -5.00547528e-01 1.71161383e-01 3.25201958e-01 1.12093663e+00 6.14657290e-02 4.99512218e-02 1.03245787e-01 4.84386683e-01 -5.64059138e-01 -8.72651458e-01 3.72499526e-01 7.90320456e-01 -6.43133223e-02 -1.43292511e+00 -8.53925526e-01 1.99446976e-01 -8.23950231e-01 -2.09435016e-01 -5.70508599e-01 2.83864707e-01 -1.05062574e-01 1.28553069e+00 -5.02436578e-01 -7.46736884e-01 4.55838382e-01 3.41237456e-01 2.20404491e-01 -2.96383739e-01 -9.27211702e-01 -5.83650708e-01 4.94786948e-02 -4.38712686e-01 1.13371713e-02 -4.11855906e-01 -7.81042993e-01 -4.24957454e-01 -2.63176978e-01 2.99222350e-01 9.73471224e-01 7.80258179e-01 1.48687974e-01 5.36653578e-01 1.39789963e+00 -8.36359560e-01 -1.26698577e+00 -1.03205431e+00 -3.93090904e-01 5.33784747e-01 7.01951623e-01 -5.28977990e-01 -1.20940888e+00 -2.51030087e-01]	[13.976191520690918, 5.820969581604004]
e50ff3a9-e3c2-45c1-afee-f126256bb488	pretraining-boosts-out-of-domain-robustness	1909.11229	null	https://arxiv.org/abs/1909.11229v2	https://arxiv.org/pdf/1909.11229v2.pdf	Pretraining boosts out-of-domain robustness for pose estimation	Neural networks are highly effective tools for pose estimation. However, as in other computer vision tasks, robustness to out-of-domain data remains a challenge, especially for small training sets that are common for real-world applications. Here, we probe the generalization ability with three architecture classes (MobileNetV2s, ResNets, and EfficientNets) for pose estimation. We developed a dataset of 30 horses that allowed for both "within-domain" and "out-of-domain" (unseen horse) benchmarking - this is a crucial test for robustness that current human pose estimation benchmarks do not directly address. We show that better ImageNet-performing architectures perform better on both within- and out-of-domain data if they are first pretrained on ImageNet. We additionally show that better ImageNet models generalize better across animal species. Furthermore, we introduce Horse-C, a new benchmark for common corruptions for pose estimation, and confirm that pretraining increases performance in this domain shift context as well. Overall, our results demonstrate that transfer learning is beneficial for out-of-domain robustness.	['Steffen Schneider', 'Thomas Biasi', 'Mert Yüksekgönül', 'Alexander Mathis', 'Matthias Bethge', 'Mackenzie W. Mathis', 'Byron Rogers']	2019-09-24	pretraining-boosts-out-of-domain-robustness-1	https://openaccess.thecvf.com/content/WACV2021/papers/Mathis_Pretraining_Boosts_Out-of-Domain_Robustness_for_Pose_Estimation_WACV_2021_paper.pdf	https://openaccess.thecvf.com/content/WACV2021/papers/Mathis_Pretraining_Boosts_Out-of-Domain_Robustness_for_Pose_Estimation_WACV_2021_paper.pdf	null	['animal-pose-estimation']	['computer-vision']	[-2.24781632e-02 -6.62568361e-02 9.67935622e-02 -4.60376680e-01 -7.14401364e-01 -6.11807466e-01 3.17328066e-01 -3.01265776e-01 -1.00946021e+00 7.21851766e-01 2.84761935e-02 1.32194445e-01 -8.88048112e-02 -5.23788154e-01 -1.24153006e+00 -3.84788811e-01 -3.50173414e-01 4.75810498e-01 5.28541565e-01 -8.29845071e-01 -2.91512143e-02 5.92438579e-01 -1.54079938e+00 1.64331689e-01 3.76936078e-01 7.53606796e-01 -1.65447265e-01 7.70439446e-01 8.06828797e-01 4.09309536e-01 -9.79800761e-01 -5.44474065e-01 6.18143559e-01 -5.77198416e-02 -9.10152674e-01 -2.74032354e-01 1.12587750e+00 -4.91539776e-01 -5.32126963e-01 8.97845685e-01 9.35783923e-01 3.47007513e-01 6.49430573e-01 -1.19200957e+00 -4.20590788e-01 1.77650765e-01 -5.22041142e-01 2.86193669e-01 3.11063200e-01 7.48662949e-01 5.82142532e-01 -4.89104718e-01 9.68908727e-01 1.36242068e+00 1.32299888e+00 7.42902160e-01 -1.17872953e+00 -8.23270559e-01 -8.92474428e-02 1.81465045e-01 -1.06212211e+00 -1.94674820e-01 4.02638435e-01 -4.11465496e-01 1.06462038e+00 6.11261092e-02 4.61719245e-01 1.74506485e+00 4.06552583e-01 6.23479486e-01 9.55368876e-01 1.60939202e-01 1.37715399e-01 -2.84347028e-01 -1.00195400e-01 2.10555404e-01 4.38191712e-01 4.73354667e-01 -5.41985989e-01 2.92871475e-01 8.28317285e-01 -4.99005646e-01 -1.33986190e-01 -8.18989098e-01 -1.39690828e+00 7.39151895e-01 9.11194503e-01 -1.07145116e-01 -1.62755355e-01 5.33178508e-01 8.63964379e-01 4.32501078e-01 4.38324034e-01 1.06200182e+00 -7.48793185e-01 -3.57077688e-01 -6.64436698e-01 6.94265723e-01 7.68980861e-01 9.83662546e-01 4.36729133e-01 1.72303021e-01 9.25545469e-02 8.37003648e-01 -2.26261705e-01 6.99248731e-01 4.54219878e-01 -1.00058353e+00 6.65518522e-01 2.10722074e-01 -4.60655093e-02 -9.84938741e-01 -9.12214637e-01 -7.75531232e-01 -8.15353751e-01 5.01598179e-01 7.94023991e-01 -1.92761138e-01 -1.10362625e+00 2.02723503e+00 2.46033788e-01 4.98236716e-02 -5.27226329e-02 1.01564682e+00 1.00157523e+00 8.59771892e-02 1.70036301e-01 5.47811508e-01 1.33325827e+00 -9.43214178e-01 -3.41829509e-02 -6.75906420e-01 7.71133721e-01 -6.55418515e-01 1.03053105e+00 4.36563730e-01 -7.74171710e-01 -7.95190811e-01 -1.28279936e+00 -7.90792033e-02 -4.98428881e-01 -1.02995588e-02 3.61760914e-01 6.02577865e-01 -9.98670042e-01 1.03986609e+00 -8.09289455e-01 -7.12297380e-01 4.74823177e-01 5.05275369e-01 -9.48706508e-01 1.40744343e-01 -1.17502391e+00 1.47494495e+00 6.26791596e-01 -5.92387579e-02 -8.97341192e-01 -7.88026810e-01 -9.21425164e-01 -4.40765530e-01 2.76605755e-01 -7.72465229e-01 1.32507908e+00 -7.47770965e-01 -1.12979698e+00 1.18454254e+00 5.69213748e-01 -8.19076478e-01 1.01765192e+00 -6.78616643e-01 -3.75535846e-01 6.27450570e-02 2.01527804e-01 1.23696554e+00 1.11637068e+00 -1.13594496e+00 -3.18556905e-01 -3.63601297e-01 2.00263634e-01 8.37279856e-02 -6.22774810e-02 -2.31686458e-01 -3.40308428e-01 -7.23476529e-01 -1.12647265e-01 -1.36379635e+00 -3.88334617e-02 2.03305315e-02 -2.76476204e-01 1.10457897e-01 8.67601693e-01 -7.64473021e-01 7.78895199e-01 -2.05024529e+00 1.03336237e-02 1.04009151e-01 2.54438013e-01 4.64612275e-01 -4.56243038e-01 1.69245303e-01 -6.09658957e-01 -7.39760697e-02 -2.78974455e-02 -1.59232065e-01 -1.31433845e-01 2.18870655e-01 -1.42775010e-02 7.60879934e-01 4.51946229e-01 9.95904922e-01 -8.26131642e-01 -3.03703755e-01 3.48718286e-01 3.01295042e-01 -9.09045577e-01 -5.62896803e-02 1.01315074e-01 8.22259724e-01 7.70985410e-02 3.09594959e-01 7.60420024e-01 -5.45346774e-02 -3.02383572e-01 -3.76814783e-01 3.95772368e-01 -1.23527244e-01 -1.03578603e+00 1.71622324e+00 -3.52446914e-01 7.73492992e-01 -2.09932879e-01 -6.92474425e-01 6.57536983e-01 -2.08413720e-01 3.01155329e-01 -7.84027755e-01 3.03227305e-01 2.40442470e-01 4.69746888e-01 -4.42965180e-01 9.74898696e-01 3.84082459e-02 -3.45882863e-01 5.53460093e-03 4.06526506e-01 -4.84326363e-01 3.67609829e-01 -1.57955721e-01 1.13466668e+00 4.12424982e-01 1.54376402e-01 -3.84875119e-01 2.08434779e-02 7.84238726e-02 3.52499723e-01 7.77014077e-01 -6.52764678e-01 1.15504193e+00 4.47137535e-01 -7.40273952e-01 -1.36315930e+00 -1.11985016e+00 -1.71388909e-01 1.38212562e+00 2.10004419e-01 -1.14345908e-01 -1.08783662e+00 -8.17009985e-01 3.06637764e-01 5.08613348e-01 -1.02545524e+00 -4.84918028e-01 -9.03130889e-01 -5.71980476e-01 1.19789732e+00 1.02868247e+00 7.02297926e-01 -1.09304881e+00 -8.38182092e-01 -7.90762603e-02 -1.17139466e-01 -1.24092269e+00 -4.76510286e-01 3.06474537e-01 -7.36540258e-01 -1.23385191e+00 -9.02902484e-01 -7.50151336e-01 3.67232859e-01 -1.07628100e-01 1.36942482e+00 -2.47577596e-02 -3.19648504e-01 3.02531004e-01 -2.15792462e-01 -4.27522749e-01 -4.71262366e-01 4.23354357e-01 4.55081552e-01 -6.31840169e-01 1.99073657e-01 -3.48079354e-01 -7.42586434e-01 8.11411738e-01 -7.01171339e-01 -4.11241502e-01 6.38032377e-01 9.72063422e-01 1.87268257e-01 -4.12769735e-01 4.64113325e-01 -6.37097418e-01 5.86958528e-01 -2.41667166e-01 -4.36197668e-01 -1.54631615e-01 -4.32199016e-02 1.58255398e-01 4.08973068e-01 -5.35708427e-01 -6.53932214e-01 6.67328164e-02 -2.96132118e-01 -4.13857162e-01 -3.72850895e-01 8.31446424e-02 1.26842290e-01 -4.14465457e-01 1.48679888e+00 -2.54328579e-01 3.12611133e-01 -3.88501495e-01 2.67336249e-01 2.07100749e-01 1.01961303e+00 -6.72242284e-01 1.04608321e+00 3.19217086e-01 2.59761781e-01 -8.60331833e-01 -4.27768171e-01 -4.72698152e-01 -8.59644711e-01 -6.24009371e-02 1.07252204e+00 -1.12317848e+00 -8.96582544e-01 7.04374433e-01 -1.04487455e+00 -5.94679117e-01 -2.23858103e-01 5.84136784e-01 -8.32414627e-01 3.39734584e-01 -4.21573669e-01 -3.02094985e-02 -7.96507224e-02 -1.45973885e+00 1.25100005e+00 -7.12834969e-02 -6.56307876e-01 -8.79730761e-01 8.53764266e-02 3.12558293e-01 3.67299795e-01 5.21444440e-01 3.64058554e-01 -8.16161990e-01 -7.40394667e-02 -3.81666392e-01 -2.41191477e-01 6.00032628e-01 -3.18599075e-01 -3.18392634e-01 -9.98326182e-01 -7.40358591e-01 -3.48650664e-01 -7.89284647e-01 9.24825013e-01 2.97501236e-01 1.09156096e+00 1.48770317e-01 -1.14444397e-01 8.26472104e-01 9.06142116e-01 -3.95034850e-01 8.89840066e-01 7.61631727e-01 7.35605240e-01 4.63698506e-01 5.93901098e-01 9.75823626e-02 2.01757804e-01 7.87456155e-01 6.18183434e-01 -1.85310289e-01 -3.86597157e-01 -2.77580053e-01 1.34186134e-01 2.26350397e-01 -4.00524914e-01 -1.17405862e-01 -1.03834426e+00 4.66571540e-01 -1.66163373e+00 -7.86622465e-01 1.08393051e-01 2.17108536e+00 7.72648036e-01 3.78722817e-01 5.14802098e-01 -1.83556825e-02 6.29644811e-01 4.00064141e-02 -6.55644596e-01 -2.63476998e-01 4.11989950e-02 4.43918586e-01 1.12740886e+00 -3.67615893e-02 -1.61763263e+00 1.02886903e+00 6.89468670e+00 8.67395222e-01 -1.16014051e+00 -2.72492394e-02 4.07131851e-01 -1.19313709e-01 5.29930711e-01 -5.62928975e-01 -6.31643116e-01 3.15357506e-01 7.24609852e-01 5.18745065e-01 1.07688479e-01 1.28253555e+00 -2.55327553e-01 -8.48006830e-02 -1.60717714e+00 1.17748189e+00 6.65265173e-02 -8.16727221e-01 -2.73147821e-01 -4.67848703e-02 8.13004553e-01 3.43650877e-01 3.70655417e-01 7.27976143e-01 2.70779520e-01 -1.35459781e+00 7.80991495e-01 -2.10316069e-02 9.94994104e-01 -8.51831913e-01 9.98053730e-01 1.95547134e-01 -1.02508616e+00 1.58344001e-01 -5.89999259e-01 8.73992965e-02 -1.91484824e-01 -1.01413861e-01 -1.00453615e+00 3.34344834e-01 1.19038081e+00 7.05769122e-01 -1.14284027e+00 1.15182197e+00 -2.30658233e-01 1.52970940e-01 -4.54630464e-01 5.15958667e-02 2.29078948e-01 4.83246863e-01 6.07012749e-01 1.21548498e+00 5.20487353e-02 -5.33315659e-01 1.54369036e-02 3.63846719e-01 -3.00648808e-01 -3.88925225e-01 -6.65860355e-01 3.97944570e-01 2.22662836e-01 9.14569259e-01 -6.29594862e-01 5.22753708e-02 7.42990300e-02 9.09218967e-01 3.68480265e-01 2.06703946e-01 -1.09444714e+00 -5.52398920e-01 9.90405083e-01 1.83317915e-01 3.72599781e-01 -2.25257516e-01 -3.65996122e-01 -1.07499409e+00 3.11575346e-02 -1.34571862e+00 2.93002665e-01 -7.68726587e-01 -1.26841652e+00 3.28767180e-01 3.06993902e-01 -1.71640778e+00 -6.09883428e-01 -1.04082763e+00 -2.41833001e-01 4.59710300e-01 -1.03128254e+00 -1.30181849e+00 -4.36978579e-01 5.56460559e-01 3.14947844e-01 -1.02911562e-01 6.75693929e-01 2.09946707e-01 -1.72447786e-01 1.17959809e+00 -7.05178455e-03 5.25071144e-01 1.16474128e+00 -1.10993469e+00 1.10075617e+00 7.60719895e-01 -2.21340656e-01 6.85479522e-01 9.77075815e-01 -6.40764713e-01 -1.24487579e+00 -1.24646986e+00 8.59761834e-02 -1.00209928e+00 5.37080050e-01 -3.04823309e-01 -7.07413077e-01 8.84296417e-01 -1.40936654e-02 1.87748909e-01 1.43220514e-01 2.48837247e-01 -6.88851416e-01 -7.43317744e-03 -1.28805852e+00 6.40441597e-01 1.47057557e+00 -4.39445317e-01 -6.93120360e-01 3.21780771e-01 6.80479884e-01 -1.18819141e+00 -9.65414703e-01 7.30469346e-01 8.14218998e-01 -9.48447347e-01 1.40979385e+00 -8.58376980e-01 7.86601543e-01 -1.85735315e-01 -5.76505065e-02 -1.70157421e+00 -2.40341559e-01 -2.25871563e-01 3.83711815e-01 6.01985216e-01 3.91621143e-01 -6.59944832e-01 1.01143920e+00 1.80830419e-01 -2.31900305e-01 -4.11766291e-01 -1.05184567e+00 -1.35399640e+00 4.52719420e-01 -6.10979021e-01 4.20750380e-01 7.33661950e-01 -4.05822963e-01 1.90923482e-01 -5.44203818e-01 5.28894663e-02 5.34691393e-01 -6.04831457e-01 1.42969561e+00 -1.01742816e+00 -3.02063018e-01 -4.45605159e-01 -1.14343739e+00 -8.89292896e-01 1.59741715e-01 -6.90856755e-01 3.46263260e-01 -1.02572131e+00 -1.35299325e-01 -8.79654009e-03 1.29902303e-01 5.19651532e-01 -1.52582690e-01 6.95591390e-01 2.72934556e-01 1.54913776e-02 -7.36521780e-01 2.00636894e-01 1.36472011e+00 2.96062417e-03 3.57630737e-02 -1.44820824e-01 -3.92899573e-01 8.83704007e-01 6.87683225e-01 -3.57624888e-01 -3.27485442e-01 -5.80960035e-01 1.12303920e-01 -4.47412461e-01 7.08942115e-01 -1.62236559e+00 -2.93435361e-02 1.56380951e-01 7.52407312e-01 -6.04432166e-01 5.37572682e-01 -6.93428636e-01 -6.40622387e-03 7.55525053e-01 -1.51126817e-01 1.77837148e-01 6.95682406e-01 4.57668215e-01 -1.56658310e-02 1.12992793e-01 1.14881146e+00 -1.16912253e-01 -1.18912470e+00 1.32821828e-01 1.09162830e-01 8.19181323e-01 9.68632519e-01 -4.64292407e-01 -5.52255452e-01 -4.39658135e-01 -6.59986019e-01 2.21990570e-01 6.55837476e-01 7.01392889e-01 4.52988416e-01 -1.23218894e+00 -8.32674921e-01 2.12539971e-01 5.90954304e-01 9.27953348e-02 3.25556517e-01 7.58572698e-01 -1.18405592e+00 1.94859326e-01 -6.62239015e-01 -1.08463502e+00 -1.23540986e+00 4.19492036e-01 5.03116548e-01 -1.25356108e-01 -2.93770909e-01 1.21179867e+00 3.21853191e-01 -9.39702332e-01 2.49368519e-01 -4.46753830e-01 1.94083720e-01 -9.98775139e-02 1.61589786e-01 5.81071615e-01 4.26226825e-01 -8.83266687e-01 -5.49175143e-01 6.93961501e-01 -7.26627186e-02 1.79982647e-01 1.27757847e+00 2.70815551e-01 3.14387918e-01 -1.96942165e-02 1.40507197e+00 -4.10025775e-01 -1.29441798e+00 3.28367800e-02 -1.46836221e-01 -3.72490466e-01 -4.98862207e-01 -6.94606900e-01 -9.14001465e-01 8.17749143e-01 8.18139851e-01 -3.52926970e-01 7.20144749e-01 -1.51941955e-01 7.12886631e-01 6.88834369e-01 5.15823007e-01 -1.41137624e+00 4.77070361e-01 8.55174363e-01 1.25488663e+00 -1.44652689e+00 1.85514539e-01 -1.42975211e-01 -7.27573693e-01 9.02590036e-01 1.10682273e+00 -4.37138706e-01 3.84349525e-01 7.55242854e-02 9.40644443e-02 -1.11999460e-01 -2.30765849e-01 -2.04339221e-01 4.89902437e-01 1.10322559e+00 2.93024778e-01 -3.04121263e-02 1.98860645e-01 2.17530414e-01 -9.81403768e-01 -1.88877672e-01 2.58219451e-01 9.74959075e-01 -2.10592493e-01 -6.77739561e-01 -6.31922841e-01 3.35070610e-01 -2.70024151e-01 1.05141230e-01 -4.86731499e-01 1.50942516e+00 2.24005312e-01 5.84485888e-01 -8.04408491e-02 -7.93367624e-01 7.47503877e-01 -1.60918236e-01 7.18705773e-01 -6.45989776e-01 -8.17230523e-01 -4.74479020e-01 2.93932140e-01 -8.64747882e-01 -3.39632621e-04 -3.06210101e-01 -8.63515139e-01 -6.49240792e-01 -2.48857751e-01 -5.48757970e-01 5.71169674e-01 7.17912436e-01 1.89798728e-01 6.98191166e-01 1.09230559e-02 -1.33239138e+00 -9.98430610e-01 -1.13519204e+00 -3.13691199e-01 8.92309070e-01 3.60933274e-01 -9.49272931e-01 -3.08314621e-01 -1.38160452e-01]	[7.466031074523926, -0.9637004137039185]
807f2394-3bfb-45fa-abb6-3018653f313f	alpha-matte-generation-from-single-input-for	2106.03210	null	https://arxiv.org/abs/2106.03210v3	https://arxiv.org/pdf/2106.03210v3.pdf	Alpha Matte Generation from Single Input for Portrait Matting	In the portrait matting, the goal is to predict an alpha matte that identifies the effect of each pixel on the foreground subject. Traditional approaches and most of the existing works utilized an additional input, e.g., trimap, background image, to predict alpha matte. However, (1) providing additional input is not always practical, and (2) models are too sensitive to these additional inputs. To address these points, in this paper, we introduce an additional input-free approach to perform portrait matting. We divide the task into two subtasks, segmentation and alpha matte prediction. We first generate a coarse segmentation map from the input image and then predict the alpha matte by utilizing the image and segmentation map. Besides, we present a segmentation encoding block to downsample the coarse segmentation map and provide useful feature representation to the residual block, since using a single encoder causes the vanishing of the segmentation information. We tested our model on four different benchmark datasets. The proposed method outperformed the MODNet and MGMatting methods that also take a single input. Besides, we obtained comparable results with BGM-V2 and FBA methods that require additional input.	['Alexander Waibel', 'Hazim Kemal Ekenel', 'Dogucan Yaman']	2021-06-06	null	null	null	null	['image-matting']	['computer-vision']	[ 6.86459839e-01 2.36090440e-02 -1.07150577e-01 -3.39982301e-01 -3.48906964e-01 -3.75665069e-01 5.17680705e-01 -1.68012291e-01 -1.04583569e-01 7.11563349e-01 -8.10273886e-02 -1.77581355e-01 3.74182492e-01 -1.09441113e+00 -9.79460597e-01 -8.11275423e-01 4.90919918e-01 2.99022108e-01 6.50527775e-01 2.63414234e-02 2.03012317e-01 1.88964352e-01 -1.42556024e+00 6.78768575e-01 1.18178976e+00 1.16379464e+00 4.69594538e-01 5.36737382e-01 -5.67062318e-01 9.50575233e-01 -6.88122451e-01 -5.03924668e-01 4.71072972e-01 -7.78597414e-01 -7.02617109e-01 4.78128433e-01 5.48288703e-01 -2.95599222e-01 -3.72656167e-01 1.13275945e+00 1.92851033e-02 -1.16256632e-01 6.10699296e-01 -1.37292778e+00 -3.84907991e-01 8.05342972e-01 -1.05259871e+00 1.51249409e-01 -2.33401190e-02 2.61165738e-01 7.90993750e-01 -7.63491929e-01 5.30289650e-01 1.32621944e+00 4.34344620e-01 3.65770310e-01 -1.09947872e+00 -6.66187763e-01 4.55299258e-01 2.53608048e-01 -1.33111095e+00 -1.95854053e-01 8.97378027e-01 -4.01710063e-01 3.65740687e-01 4.49174851e-01 8.98221254e-01 8.61353457e-01 2.73788065e-01 1.22453856e+00 1.43302011e+00 -2.01205090e-01 1.46537408e-01 1.35497198e-01 1.26114786e-01 7.11628139e-01 4.35663611e-02 -3.06037754e-01 -3.46256524e-01 2.79340148e-01 8.24500799e-01 2.10715160e-02 -2.67233998e-01 -7.64797479e-02 -1.30101931e+00 4.00542200e-01 6.15804374e-01 2.22826809e-01 -3.46744865e-01 2.52301246e-01 8.90330747e-02 6.45056590e-02 5.13692617e-01 6.26797080e-02 -2.49670818e-01 1.29479274e-01 -1.52541912e+00 -4.32395339e-02 6.69971526e-01 9.78984892e-01 1.11636508e+00 2.92249084e-01 -2.88569748e-01 5.51819921e-01 1.99435771e-01 5.64042807e-01 2.60283947e-01 -6.55933261e-01 6.99601710e-01 9.52772379e-01 -1.66413113e-01 -1.16953385e+00 -1.55951783e-01 -5.03269672e-01 -1.16796446e+00 2.62361884e-01 5.42806387e-01 -3.53880934e-02 -1.55915737e+00 1.40956676e+00 3.19871187e-01 4.07188773e-01 -1.73602670e-01 9.28215861e-01 7.98794150e-01 1.04482746e+00 -1.33102849e-01 -1.89232752e-01 1.14944899e+00 -1.38218534e+00 -9.23798501e-01 -4.10298228e-01 3.49641502e-01 -1.01089799e+00 1.02772844e+00 5.23686111e-01 -1.11962450e+00 -6.05627120e-01 -1.21204519e+00 7.97363222e-02 -3.29390258e-01 3.12020719e-01 4.85176802e-01 5.69531739e-01 -8.98543358e-01 6.68037117e-01 -8.31309795e-01 -1.65403977e-01 4.95586216e-01 3.83372694e-01 -4.19453606e-02 7.50934146e-03 -9.12821829e-01 7.61476576e-01 5.88383496e-01 3.01176339e-01 -1.04525292e+00 -5.68099678e-01 -7.18573928e-01 2.59087924e-02 5.32715857e-01 -6.97883308e-01 8.49606156e-01 -1.71209741e+00 -1.27503014e+00 6.64475441e-01 -2.41110802e-01 -4.79299873e-01 6.71096742e-01 -1.10570773e-01 -1.69872522e-01 2.40957998e-02 4.82022800e-02 9.06612694e-01 9.16237473e-01 -1.54971731e+00 -8.52176785e-01 -1.85815185e-01 1.97416618e-01 2.18900278e-01 -2.02910021e-01 -3.15385014e-01 -8.64677250e-01 -8.79951537e-01 2.02229366e-01 -7.44922400e-01 -2.64512151e-01 -3.08009293e-02 -7.71596670e-01 3.27155620e-01 1.06358159e+00 -9.68567133e-01 1.41754043e+00 -2.04327464e+00 2.59119064e-01 2.59111941e-01 2.67429680e-01 2.03473881e-01 9.13879052e-02 1.03124967e-02 1.40641928e-01 2.22285420e-01 -5.04359066e-01 -2.94676214e-01 -2.39519030e-01 2.53291309e-01 -2.94122458e-01 3.48828554e-01 1.26008824e-01 8.80308151e-01 -5.99215388e-01 -9.18320775e-01 4.10524994e-01 3.23848516e-01 -4.13419992e-01 8.27254951e-02 -5.24191499e-01 4.24756825e-01 -2.46411353e-01 9.21282291e-01 9.51495469e-01 -2.32396588e-01 -4.44012396e-02 -5.12490869e-01 -1.90493956e-01 -7.32791126e-02 -1.26039410e+00 1.50593460e+00 -2.94677675e-01 6.08753860e-01 1.78727046e-01 -7.85853744e-01 9.09594417e-01 -2.60810740e-02 5.07195771e-01 -6.35510802e-01 2.97550470e-01 7.50968084e-02 8.43514726e-02 -1.71320736e-01 4.18595165e-01 2.28229072e-02 2.51521263e-02 3.03446442e-01 -1.03346050e-01 9.48531404e-02 2.44234815e-01 2.08199814e-01 9.90430415e-01 2.89801151e-01 3.41053456e-02 -1.55218363e-01 6.23819530e-01 2.33297154e-01 8.00056875e-01 5.58884203e-01 3.19371372e-02 7.19386578e-01 6.56643152e-01 -5.91848016e-01 -1.07543111e+00 -9.11312521e-01 1.95733666e-01 7.89753914e-01 7.11326420e-01 -5.51507473e-01 -1.15222383e+00 -7.79626667e-01 -2.76315957e-01 5.90289712e-01 -8.39383423e-01 1.44488305e-01 -7.79235303e-01 -9.38977122e-01 3.15996140e-01 5.81424177e-01 1.11649883e+00 -9.19840217e-01 -3.42471778e-01 -1.37924533e-02 -3.74221295e-01 -1.06275630e+00 -5.71002185e-01 1.07919559e-01 -8.79205287e-01 -8.95770490e-01 -6.96966708e-01 -7.65136540e-01 9.27672684e-01 2.92048633e-01 1.04136491e+00 3.26736182e-01 -1.78842753e-01 -1.58144131e-01 -2.07168788e-01 -2.85897881e-01 -3.36604685e-01 2.24619672e-01 -4.12186772e-01 4.33392376e-01 7.24169016e-02 -5.20124495e-01 -6.42765522e-01 4.41331595e-01 -8.99857342e-01 8.70855033e-01 9.58233058e-01 6.28450394e-01 8.79886031e-01 1.39890626e-01 2.29328364e-01 -1.12477040e+00 5.72994389e-02 -2.82046974e-01 -5.19715488e-01 2.84410089e-01 -3.74851137e-01 -6.75439760e-02 8.17131698e-01 -3.56155217e-01 -1.19251645e+00 4.04500455e-01 -1.53980613e-01 -5.45833588e-01 -7.39761293e-02 2.50456959e-01 -6.60172224e-01 -4.74826582e-02 3.30405086e-01 3.71796250e-01 -2.55788624e-01 -4.81666535e-01 2.74792343e-01 4.90625829e-01 4.92208093e-01 -3.29990566e-01 9.76601243e-01 4.62168872e-01 3.25980745e-02 -6.04487717e-01 -8.96022797e-01 -8.54128897e-02 -9.02605295e-01 -4.07197565e-01 1.01314044e+00 -7.18227506e-01 -3.21490884e-01 5.23917615e-01 -1.09732842e+00 -6.16041124e-01 -5.31742349e-02 6.63671643e-02 -4.21714365e-01 3.96846384e-01 -6.06172144e-01 -6.19105101e-01 -3.93388540e-01 -1.23140383e+00 8.53972137e-01 3.21263820e-01 7.90598094e-02 -7.61261761e-01 -4.06617165e-01 5.36593318e-01 3.12926382e-01 5.01883566e-01 8.62523794e-01 -2.46556878e-01 -1.10894179e+00 1.54588282e-01 -3.84943068e-01 2.24938005e-01 3.29676181e-01 1.80296913e-01 -9.79952991e-01 4.84027807e-03 -9.54797715e-02 5.57099730e-02 1.16914332e+00 3.22182477e-01 1.40854025e+00 -3.99603277e-01 -3.61402601e-01 9.28056121e-01 1.34053683e+00 3.54543746e-01 9.24131989e-01 4.33031678e-01 1.09431148e+00 3.71205330e-01 8.70626271e-01 1.91750735e-01 1.91668361e-01 5.73410749e-01 5.81021547e-01 -5.64806163e-01 -4.54167932e-01 -3.78231585e-01 3.75766695e-01 8.21795166e-01 -1.83797643e-01 -3.36701572e-01 -7.47099936e-01 5.05207241e-01 -1.92768359e+00 -7.44546294e-01 -3.14221263e-01 1.86617756e+00 8.90512586e-01 3.75662386e-01 1.37976751e-01 1.99377805e-01 8.89690578e-01 2.40732953e-01 -6.69687629e-01 -9.81601998e-02 -3.61557454e-01 6.02917410e-02 6.37844622e-01 5.24825513e-01 -1.16360188e+00 1.23328471e+00 6.10976171e+00 1.04564631e+00 -1.17234874e+00 -4.19618934e-02 1.05703270e+00 -2.34301686e-02 -4.15182233e-01 4.11449969e-02 -8.00293446e-01 8.16047490e-01 5.89709282e-01 2.72277946e-04 2.99075633e-01 6.90956891e-01 8.97139534e-02 -2.72947609e-01 -8.51245522e-01 9.19360399e-01 2.32743397e-01 -1.19847322e+00 3.15720022e-01 -6.68766275e-02 7.84887135e-01 -5.06607950e-01 -9.03485715e-02 1.61072940e-01 2.82063615e-02 -1.06593847e+00 7.04285145e-01 4.53810126e-01 6.35788262e-01 -7.78694034e-01 7.68443108e-01 3.88073206e-01 -1.30439544e+00 1.36700556e-01 -4.05947089e-01 -9.26485881e-02 1.20239757e-01 6.57624602e-01 -8.33830416e-01 4.87440079e-01 4.77960885e-01 5.66112459e-01 -9.42657351e-01 1.14557540e+00 -3.52968872e-01 8.12763453e-01 -2.05743328e-01 2.08288893e-01 2.28058323e-01 -4.02988255e-01 2.37527445e-01 1.24835253e+00 2.34615132e-01 -1.50945812e-01 3.42422754e-01 9.83027935e-01 -1.94647223e-01 1.92327425e-01 -2.08550572e-01 4.06852700e-02 1.42709240e-01 1.46028900e+00 -1.28982258e+00 -6.15097106e-01 -3.62481087e-01 1.21664846e+00 -1.41542420e-01 2.55330741e-01 -1.17717373e+00 -2.37428814e-01 3.18378150e-01 4.42438334e-01 3.83476615e-01 2.91192606e-02 -7.14025021e-01 -1.17844045e+00 -1.05558403e-01 -9.39985693e-01 1.91390738e-01 -7.52703071e-01 -7.69657016e-01 6.47685111e-01 -1.55304489e-03 -1.12914705e+00 1.24759376e-01 -4.47016656e-01 -6.56921506e-01 7.98027098e-01 -1.21617794e+00 -1.52884054e+00 -7.55054712e-01 5.51585138e-01 9.15400743e-01 1.37361184e-01 3.06196779e-01 4.97171670e-01 -8.63357246e-01 4.90427077e-01 -2.46895626e-01 2.20099032e-01 6.01170778e-01 -1.40817153e+00 5.22878051e-01 1.15398550e+00 1.39134601e-01 2.60928214e-01 6.84022367e-01 -9.73491907e-01 -1.15927577e+00 -1.33725870e+00 4.55197930e-01 -2.32742503e-01 5.00363111e-01 -5.66435635e-01 -8.57762992e-01 8.04685354e-01 4.62644279e-01 -2.05782264e-01 2.68109798e-01 -3.37542504e-01 9.61490162e-03 -3.27788770e-01 -1.02790022e+00 7.37108231e-01 9.89335597e-01 1.20503888e-01 -3.23204011e-01 5.95196784e-02 7.60197163e-01 -6.30496919e-01 -6.07125163e-01 3.48240227e-01 4.13611144e-01 -7.91519225e-01 7.93031812e-01 -1.20904915e-01 7.66808927e-01 -7.18309164e-01 -2.10540090e-02 -1.14753032e+00 -4.11273211e-01 -3.99766505e-01 -1.81153446e-01 1.36749327e+00 4.28092003e-01 -1.25644296e-01 1.10931611e+00 4.94595379e-01 -1.65725395e-01 -9.55233991e-01 -5.56156576e-01 -3.03178668e-01 -2.01698527e-01 -8.91263708e-02 6.94802344e-01 7.38172114e-01 -6.24329865e-01 3.62223864e-01 -8.11313450e-01 1.23357311e-01 7.32674420e-01 4.64666098e-01 1.00123453e+00 -8.72209609e-01 -1.69869319e-01 -2.42286339e-01 -3.76072437e-01 -1.23892248e+00 -2.08061382e-01 -7.55438805e-01 4.65393029e-02 -1.70366347e+00 4.05594677e-01 -2.98064500e-01 -1.64378494e-01 5.81211030e-01 -4.02409166e-01 6.02510512e-01 3.96915257e-01 1.86846986e-01 -6.43493891e-01 2.94067562e-01 1.53349340e+00 -4.76625562e-01 -1.89987898e-01 1.04885004e-01 -6.79952502e-01 8.81122470e-01 9.03852999e-01 -3.84104401e-01 -5.15305340e-01 -4.72766757e-01 4.09500860e-02 -1.26394823e-01 2.76676893e-01 -1.14687324e+00 2.87127078e-01 -4.41233397e-01 8.92867804e-01 -1.11077499e+00 4.47753489e-01 -8.08516026e-01 2.97433853e-01 3.70842755e-01 6.68267347e-03 -4.48417440e-02 2.35354826e-01 3.96423012e-01 -2.11059883e-01 -1.85920492e-01 8.16701770e-01 -3.58836859e-01 -8.26025426e-01 3.76519024e-01 -3.06928277e-01 -1.82210073e-01 1.15308022e+00 -5.38303554e-01 -3.02275509e-01 -2.39614084e-01 -5.75637519e-01 2.29635224e-01 7.35279620e-01 1.83066532e-01 5.88657558e-01 -1.28393352e+00 -4.88145858e-01 1.30853832e-01 -4.85690445e-01 2.92645484e-01 2.75725245e-01 8.52507532e-01 -7.29185045e-01 5.48847951e-02 -5.16728163e-01 -6.87671065e-01 -1.43206656e+00 5.74215710e-01 2.38005310e-01 -2.55720764e-01 -6.34484291e-01 5.98324478e-01 7.78076947e-01 4.17346228e-03 2.56622702e-01 -4.12083477e-01 -4.30042520e-02 6.34737164e-02 5.04797876e-01 1.65002897e-01 -1.23946160e-01 -6.14716351e-01 -2.05684796e-01 4.91084009e-01 -2.09171161e-01 -7.21745240e-03 1.19837928e+00 -2.11922944e-01 -1.70503572e-01 4.69583899e-01 8.04148614e-01 -8.22811797e-02 -1.33290136e+00 -5.45408763e-02 -1.86119422e-01 -8.06630969e-01 7.49616399e-02 -7.89431334e-01 -1.45456266e+00 1.02068913e+00 3.74760389e-01 7.37287179e-02 1.49987614e+00 -2.25255117e-01 1.04637349e+00 -4.34436500e-02 2.04228565e-01 -1.12305164e+00 1.64138511e-01 1.42307609e-01 6.97329938e-01 -1.00919771e+00 2.18723431e-01 -7.96817601e-01 -7.70375073e-01 1.01275706e+00 1.13031209e+00 3.19305086e-03 2.81889379e-01 4.36338007e-01 9.22066569e-02 -4.16536927e-02 -6.82079792e-01 -1.83957115e-01 3.20769459e-01 3.55114400e-01 2.81333119e-01 -9.56133530e-02 -2.43403047e-01 6.17303431e-01 -2.90455550e-01 -3.98747846e-02 4.47125435e-01 8.37316573e-01 -6.07465804e-01 -1.03096724e+00 -5.85989416e-01 7.20200121e-01 -5.72407484e-01 -5.18007539e-02 -7.38470614e-01 7.77000487e-01 5.49424410e-01 6.10393703e-01 3.69265005e-02 -6.92061245e-01 1.56881228e-01 -1.40748531e-01 4.36711818e-01 -4.78382140e-01 -6.13750219e-01 3.04629087e-01 -1.13716694e-02 -4.82314706e-01 -3.96866471e-01 -2.99245954e-01 -9.63613570e-01 -5.30035913e-01 -3.63217086e-01 -2.35385969e-01 4.01156008e-01 8.71995032e-01 2.74516027e-02 8.27515185e-01 5.36114097e-01 -7.89644778e-01 1.12801053e-01 -9.62499321e-01 -5.24986565e-01 4.02485400e-01 5.22790849e-02 -4.88038301e-01 -2.54867017e-01 4.64181036e-01]	[10.659375190734863, -0.9397047162055969]
502b34aa-570f-4f1a-a7b4-4d957661c76e	temporal-relational-crosstransformers-for-few	2101.06184	null	https://arxiv.org/abs/2101.06184v3	https://arxiv.org/pdf/2101.06184v3.pdf	Temporal-Relational CrossTransformers for Few-Shot Action Recognition	We propose a novel approach to few-shot action recognition, finding temporally-corresponding frame tuples between the query and videos in the support set. Distinct from previous few-shot works, we construct class prototypes using the CrossTransformer attention mechanism to observe relevant sub-sequences of all support videos, rather than using class averages or single best matches. Video representations are formed from ordered tuples of varying numbers of frames, which allows sub-sequences of actions at different speeds and temporal offsets to be compared. Our proposed Temporal-Relational CrossTransformers (TRX) achieve state-of-the-art results on few-shot splits of Kinetics, Something-Something V2 (SSv2), HMDB51 and UCF101. Importantly, our method outperforms prior work on SSv2 by a wide margin (12%) due to the its ability to model temporal relations. A detailed ablation showcases the importance of matching to multiple support set videos and learning higher-order relational CrossTransformers.	['Dima Damen', 'Majid Mirmehdi', 'Tilo Burghardt', 'Alessandro Masullo', 'Toby Perrett']	2021-01-15	null	http://openaccess.thecvf.com//content/CVPR2021/html/Perrett_Temporal-Relational_CrossTransformers_for_Few-Shot_Action_Recognition_CVPR_2021_paper.html	http://openaccess.thecvf.com//content/CVPR2021/papers/Perrett_Temporal-Relational_CrossTransformers_for_Few-Shot_Action_Recognition_CVPR_2021_paper.pdf	cvpr-2021-1	['few-shot-action-recognition']	['computer-vision']	[ 2.25696906e-01 -3.58361185e-01 -6.83155656e-01 -3.11489522e-01 -9.77971852e-01 -4.63441133e-01 9.06897247e-01 3.79037336e-02 -3.53876024e-01 5.25090039e-01 5.49655318e-01 2.75814623e-01 -2.09682345e-01 -2.32389569e-01 -9.48234856e-01 -3.90954942e-01 -5.74127555e-01 3.95128965e-01 9.01728690e-01 -3.99937257e-02 1.78583562e-01 4.09926981e-01 -1.94076169e+00 1.02766359e+00 1.37392581e-01 1.24017429e+00 -1.81368832e-02 9.86628532e-01 1.22347325e-01 1.40196228e+00 -5.61468899e-01 -3.78620744e-01 3.86548519e-01 -7.32727885e-01 -7.83880711e-01 3.43590260e-01 9.31146502e-01 -5.02228141e-01 -8.06152940e-01 5.67366004e-01 3.31810087e-01 6.71181202e-01 4.98272657e-01 -1.67293739e+00 -3.87346089e-01 4.40878391e-01 -6.21172190e-01 9.52732444e-01 8.16308439e-01 3.31139952e-01 1.06601322e+00 -1.02243197e+00 1.19852805e+00 1.31772554e+00 5.67460001e-01 3.14442098e-01 -1.15396464e+00 -5.55636883e-01 1.93019673e-01 9.17720556e-01 -1.32157540e+00 -8.33939195e-01 3.85287791e-01 -4.63175297e-01 1.51543498e+00 2.67357528e-01 8.56427491e-01 1.37447369e+00 1.38914526e-01 1.06268477e+00 6.08702362e-01 -8.05717185e-02 2.83129066e-01 -5.19160450e-01 2.45298818e-02 6.59595013e-01 -1.90968305e-01 -6.03054725e-02 -1.18151927e+00 -4.11303341e-02 7.35283732e-01 4.16204274e-01 -4.70992595e-01 -4.94202882e-01 -1.48043180e+00 5.47107160e-01 7.14234337e-02 3.09357464e-01 -4.47604597e-01 4.09937173e-01 8.08408022e-01 5.02804637e-01 1.44321710e-01 2.29360461e-01 -3.41750354e-01 -5.05351424e-01 -1.03363287e+00 3.01820904e-01 5.49919367e-01 1.22641635e+00 4.36632782e-01 -6.59164833e-03 -6.61848426e-01 4.05453265e-01 -9.56503376e-02 3.93654220e-02 7.36466825e-01 -1.44421780e+00 5.10459244e-01 3.09569567e-01 6.06747484e-03 -9.46393490e-01 7.60652795e-02 2.82143652e-01 -2.31179684e-01 -6.89568818e-02 2.54610151e-01 2.85601288e-01 -8.87034893e-01 1.32688498e+00 2.21931115e-01 9.64989901e-01 -3.29270959e-02 7.34317422e-01 7.20315099e-01 4.65085059e-01 1.11887686e-01 -5.59627235e-01 1.39364803e+00 -9.96104240e-01 -6.07991397e-01 1.17274620e-01 5.13103545e-01 -4.21433866e-01 1.00090826e+00 1.78612798e-01 -1.21379042e+00 -8.10293138e-01 -1.02459049e+00 -2.56527830e-02 -1.86424866e-01 -3.02358001e-01 3.30388337e-01 9.78972390e-02 -8.73243570e-01 9.74165916e-01 -1.12064695e+00 -5.75300157e-01 7.03127563e-01 6.28893822e-02 -6.76126838e-01 -1.05612054e-01 -9.57745433e-01 5.64329565e-01 3.81461531e-01 -5.24325371e-01 -1.14950252e+00 -9.57947731e-01 -9.96998727e-01 1.41184211e-01 6.89281404e-01 -4.63581800e-01 1.27414012e+00 -9.87916410e-01 -1.17165172e+00 7.41032600e-01 -3.59918356e-01 -9.54044640e-01 4.96315598e-01 -4.47307974e-01 -5.79047918e-01 8.72202396e-01 1.38678208e-01 6.89457595e-01 8.81751120e-01 -5.63524723e-01 -7.94268966e-01 -1.37184069e-01 1.01390883e-01 2.83233017e-01 2.14562360e-02 4.04446721e-01 -8.36868227e-01 -6.45632148e-01 -9.98783857e-02 -6.71376646e-01 2.11693704e-01 1.06245659e-01 -1.20698012e-01 -2.28097379e-01 1.09138906e+00 -3.59521389e-01 1.20680928e+00 -2.27013659e+00 1.98097065e-01 -3.27503592e-01 -1.58821736e-02 1.44360766e-01 -2.09502488e-01 4.56070870e-01 -3.29286337e-01 -1.33779585e-01 2.09981278e-01 -2.86525041e-01 -1.94597125e-01 3.45475137e-01 -4.96098250e-01 5.68793178e-01 2.44152412e-01 1.00727427e+00 -9.07884181e-01 -7.31042325e-01 3.72595817e-01 3.72053742e-01 -3.64769757e-01 1.75029755e-01 -3.64785880e-01 5.39110228e-02 -6.60811216e-02 7.19429076e-01 -2.68824715e-02 -3.30767363e-01 9.00384113e-02 -3.11741382e-01 3.56787443e-03 1.53872326e-01 -1.00021410e+00 1.95699620e+00 8.84975567e-02 9.91604686e-01 -5.45391858e-01 -1.00880480e+00 4.81441528e-01 6.04342580e-01 1.05136335e+00 -7.44525015e-01 3.34012159e-03 -2.33528629e-01 -3.03661495e-01 -7.46599853e-01 3.45016479e-01 -1.14128925e-02 1.04674362e-01 2.96838760e-01 5.46715915e-01 1.88744545e-01 7.05636442e-01 5.16833246e-01 1.40429056e+00 5.02638400e-01 4.89682019e-01 2.76761174e-01 2.55600750e-01 -2.21908838e-02 5.92204213e-01 7.47101605e-01 -5.71355164e-01 8.77287745e-01 7.80973256e-01 -5.20464778e-01 -9.66184437e-01 -8.55001688e-01 2.93048531e-01 1.29354322e+00 1.76018730e-01 -9.11209345e-01 -3.55649680e-01 -7.86217988e-01 -2.33418494e-02 5.86354673e-01 -8.40171516e-01 -1.71299726e-01 -7.97703445e-01 -4.58681863e-03 4.19348001e-01 8.84394646e-01 3.31546873e-01 -1.12663233e+00 -1.25960946e+00 2.60334164e-01 -2.58037478e-01 -1.44688511e+00 -7.19207525e-01 1.67647451e-01 -8.30072582e-01 -1.35286868e+00 -5.91743648e-01 -4.48242188e-01 -3.96505482e-02 6.09463394e-01 1.17991114e+00 -2.22244829e-01 -5.40446877e-01 6.11698747e-01 -6.85385406e-01 2.97319535e-02 -4.95155565e-02 -5.00672758e-01 1.19510315e-01 1.68848783e-01 4.14048612e-01 -5.09989738e-01 -5.27107537e-01 4.95874435e-01 -7.23255217e-01 -1.59372032e-01 2.03514770e-01 7.23038793e-01 9.84422445e-01 -2.36150622e-01 2.20214546e-01 -4.50906157e-01 -5.95359914e-02 -5.21115601e-01 -2.64067709e-01 4.43659157e-01 -6.29771203e-02 -1.33760974e-01 4.22687352e-01 -7.19103515e-01 -8.29444885e-01 6.70089498e-02 5.72733998e-01 -1.49694860e+00 -1.24655493e-01 9.59233269e-02 2.11585715e-01 4.80290025e-01 6.30293071e-01 1.64930940e-01 -1.27548099e-01 -2.01485723e-01 4.32361692e-01 9.84559357e-02 9.13702726e-01 -3.56676072e-01 2.89411664e-01 6.93952203e-01 -5.63389175e-02 -7.92888403e-01 -6.23858213e-01 -8.35828245e-01 -8.33773732e-01 -3.80849987e-01 1.24692142e+00 -9.47444975e-01 -8.02427053e-01 6.00523800e-02 -9.96851146e-01 -3.73513013e-01 -6.76039755e-01 7.12056696e-01 -1.20591450e+00 3.45690012e-01 -6.43592358e-01 -6.32721722e-01 8.85428339e-02 -9.31938648e-01 1.26449835e+00 -1.61122009e-02 -4.88309383e-01 -3.72311532e-01 1.01092279e-01 3.29283148e-01 -1.99879885e-01 3.37218761e-01 5.85921884e-01 -9.45810437e-01 -6.93888187e-01 -1.31990075e-01 5.72583266e-02 -3.81917469e-02 2.66409628e-02 2.34132901e-01 -7.57626653e-01 -2.99229354e-01 -1.63335085e-01 -6.02994323e-01 1.11156392e+00 5.50602555e-01 1.08950591e+00 -2.30869964e-01 -4.01243538e-01 4.97408301e-01 1.15236330e+00 4.63298410e-01 7.27511108e-01 1.14515327e-01 5.44956446e-01 3.73466194e-01 1.02543461e+00 6.41971111e-01 1.11555215e-02 8.46604288e-01 3.56127650e-01 3.08873087e-01 -2.70064831e-01 -1.49653569e-01 5.54745018e-01 3.12511772e-01 -3.40572655e-01 -1.92568615e-01 -5.66701114e-01 6.19335949e-01 -2.24670792e+00 -2.04988337e+00 1.77287787e-01 2.29746079e+00 3.01632792e-01 2.25747496e-01 5.30849755e-01 4.37361225e-02 6.93786621e-01 5.15170813e-01 -6.86206758e-01 7.51139522e-02 -1.87295794e-01 1.68717295e-01 1.99265018e-01 6.58303648e-02 -1.35882282e+00 8.38888109e-01 6.57217216e+00 8.13090086e-01 -7.30454683e-01 1.01033077e-01 6.22542262e-01 -8.87618721e-01 3.02309126e-01 -7.80583769e-02 -5.53841054e-01 3.21335822e-01 1.03271055e+00 5.18290401e-02 3.55860472e-01 7.68061340e-01 1.18425116e-01 -2.09751576e-01 -1.40871215e+00 1.07578266e+00 3.90477270e-01 -1.68304265e+00 -1.12568408e-01 -2.98157871e-01 7.42010951e-01 1.88776970e-01 -2.14716867e-01 3.77533078e-01 2.39253283e-01 -8.25921953e-01 8.82729709e-01 6.38238966e-01 6.75437331e-01 -5.76011121e-01 3.61519605e-01 -1.53925478e-01 -1.58440471e+00 -3.99488568e-01 -3.58565301e-01 1.67994887e-01 3.02818120e-01 -6.25648201e-02 -7.46087193e-01 6.00064039e-01 1.06288815e+00 1.23319280e+00 -4.74718392e-01 9.20728266e-01 1.57246709e-01 3.65162492e-01 -1.43992960e-01 3.02467316e-01 1.41297773e-01 1.11954026e-01 4.59353805e-01 1.07973182e+00 3.66264254e-01 3.50092977e-01 3.60736698e-01 2.27515087e-01 7.13919923e-02 -1.57730982e-01 -7.11408079e-01 -1.35544613e-01 4.65497702e-01 8.08876216e-01 -8.62525940e-01 -8.92724454e-01 -8.06231916e-01 1.21216476e+00 2.00223163e-01 3.15230131e-01 -1.14390612e+00 -1.52770102e-01 9.19491351e-01 2.24707961e-01 1.03277516e+00 -2.30179727e-02 4.15172160e-01 -1.31128788e+00 1.38450218e-02 -1.05456352e+00 1.06975424e+00 -1.00667405e+00 -9.17406738e-01 5.45664728e-01 5.33656597e-01 -1.62553072e+00 -4.18100029e-01 -1.91346452e-01 -4.99026716e-01 9.73544344e-02 -9.16170597e-01 -9.72806215e-01 -2.56378382e-01 8.36804390e-01 1.27362931e+00 -2.58687496e-01 5.63645005e-01 2.08103135e-01 -3.62307012e-01 2.37042889e-01 -1.67105138e-01 5.53233037e-03 8.05142939e-01 -9.07236159e-01 5.59026241e-01 7.48070836e-01 6.65629983e-01 3.17130268e-01 7.46527731e-01 -7.61231124e-01 -1.33723032e+00 -9.50873137e-01 7.02647388e-01 -5.75733364e-01 7.82512367e-01 -1.57347977e-01 -9.58442867e-01 1.01156437e+00 2.56218731e-01 5.72848976e-01 5.69173515e-01 -2.85144180e-01 -7.16499329e-01 -5.37639968e-02 -7.68668592e-01 5.75150967e-01 1.49759138e+00 -7.98417032e-01 -7.30041206e-01 3.91766787e-01 7.70871997e-01 -2.13258728e-01 -8.14629674e-01 3.23990792e-01 8.08511198e-01 -1.36401892e+00 1.29260099e+00 -1.07362187e+00 5.62880635e-01 -3.07446718e-01 -5.23763657e-01 -8.52530062e-01 -2.21168250e-01 -8.27140987e-01 -6.68657064e-01 9.28149939e-01 -1.16571762e-01 8.25080499e-02 7.48498082e-01 2.59704143e-01 -1.47809640e-01 -8.02310884e-01 -1.08561623e+00 -1.07668567e+00 -6.35082603e-01 -5.86613894e-01 4.64872926e-01 8.99860442e-01 2.74732769e-01 1.94873005e-01 -6.32381499e-01 -1.36121690e-01 2.69967020e-01 2.79115915e-01 6.69741213e-01 -8.69968593e-01 -7.79909194e-01 -3.89633626e-01 -8.54880810e-01 -9.81551230e-01 8.38781074e-02 -5.11140168e-01 2.92676426e-02 -1.10630333e+00 2.83484131e-01 4.03396308e-01 -4.42916036e-01 5.03260791e-01 -8.77875462e-02 5.30593097e-01 3.70742023e-01 4.24001604e-01 -1.25295377e+00 5.04243016e-01 9.22709346e-01 -2.80580074e-01 -1.85800701e-01 -2.62473911e-01 -1.88148208e-02 5.52631021e-01 3.35213393e-01 -5.19289672e-01 -7.35616624e-01 -3.77307720e-02 -3.39219779e-01 5.78238964e-01 3.87605250e-01 -1.39366090e+00 2.23545656e-01 -2.99857765e-01 5.22729397e-01 -7.83863127e-01 8.17297101e-01 -6.18933260e-01 4.11539376e-01 4.15184855e-01 -5.80320597e-01 3.58395666e-01 -2.95527466e-02 9.28205729e-01 -2.10818246e-01 2.50154585e-02 6.99468672e-01 -2.72441059e-01 -1.16038299e+00 4.10567522e-01 -3.68457764e-01 2.09305301e-01 1.45937681e+00 -5.72628796e-01 -4.17287439e-01 -4.75870371e-01 -8.15175176e-01 9.49823186e-02 4.19175923e-01 4.70036447e-01 7.52311766e-01 -1.45418644e+00 -4.34899658e-01 4.10811119e-02 4.11863089e-01 -3.73212069e-01 4.69883263e-01 1.08216369e+00 -2.26478055e-01 2.89042205e-01 -5.08163512e-01 -6.91827536e-01 -1.68691659e+00 8.99140894e-01 1.11205369e-01 -3.94480601e-02 -9.69786167e-01 9.07811880e-01 1.59400895e-01 3.65705729e-01 3.37383509e-01 -3.03146601e-01 -1.19421214e-01 4.21226829e-01 8.32137287e-01 5.99222600e-01 -3.54755148e-02 -9.40174460e-01 -4.66723651e-01 3.33748847e-01 -1.00251608e-01 -1.63905129e-01 1.28787541e+00 1.01478145e-01 5.52843809e-01 7.30811298e-01 1.39818203e+00 -4.81266320e-01 -1.70758021e+00 -2.17937097e-01 -4.84843031e-02 -8.28285217e-01 -3.84812832e-01 -3.34109664e-01 -8.69401455e-01 6.18933797e-01 4.58372146e-01 -1.22986753e-02 9.17534053e-01 3.85808468e-01 6.53731406e-01 3.65889281e-01 3.19013029e-01 -1.04368758e+00 7.90091336e-01 2.92983025e-01 8.01444471e-01 -1.18374097e+00 1.07434347e-01 -1.67356208e-01 -9.76263285e-01 9.43270385e-01 5.00604331e-01 -2.10297570e-01 3.16856533e-01 6.58980608e-02 -2.34109789e-01 -4.16569948e-01 -1.26227736e+00 -2.93455213e-01 3.45898271e-01 6.33992970e-01 2.98855096e-01 -2.65269816e-01 1.11786060e-01 1.45416230e-01 3.78569633e-01 -5.88666927e-03 4.17869121e-01 1.12715280e+00 -4.10947114e-01 -5.90475738e-01 -1.23578556e-01 5.79107463e-01 -3.83226216e-01 1.20738484e-01 -2.00027198e-01 9.50936437e-01 -1.27450556e-01 8.48677933e-01 5.85923493e-01 -5.27647376e-01 4.09541696e-01 3.59753788e-01 6.71583116e-01 -4.98590887e-01 -4.55683231e-01 2.80573964e-01 2.09862396e-01 -1.26259363e+00 -7.93944716e-01 -1.05042040e+00 -1.22517288e+00 -1.55577093e-01 -6.17670044e-02 -1.48428872e-01 -2.40874644e-02 8.97647023e-01 6.06599748e-01 4.46111411e-01 3.55105907e-01 -1.07998550e+00 -3.30885112e-01 -6.91946089e-01 -5.31634927e-01 8.94790351e-01 3.57154608e-01 -9.24947321e-01 -1.49939552e-01 4.46200639e-01]	[8.630788803100586, 0.7941852807998657]
9d84e679-412a-443f-96c0-56a88e979ed0	joint-spatio-temporal-modeling-for-semantic	2212.05245	null	https://arxiv.org/abs/2212.05245v4	https://arxiv.org/pdf/2212.05245v4.pdf	Joint Spatio-Temporal Modeling for the Semantic Change Detection in Remote Sensing Images	Semantic Change Detection (SCD) refers to the task of simultaneously extracting the changed areas and the semantic categories (before and after the changes) in Remote Sensing Images (RSIs). This is more meaningful than Binary Change Detection (BCD) since it enables detailed change analysis in the observed areas. Previous works established triple-branch Convolutional Neural Network (CNN) architectures as the paradigm for SCD. However, it remains challenging to exploit semantic information with a limited amount of change samples. In this work, we investigate to jointly consider the spatio-temporal dependencies to improve the accuracy of SCD. First, we propose a Semantic Change Transformer (SCanFormer) to explicitly model the 'from-to' semantic transitions between the bi-temporal RSIs. Then, we introduce a semantic learning scheme to leverage the spatio-temporal constraints, which are coherent to the SCD task, to guide the learning of semantic changes. The resulting network (SCanNet) significantly outperforms the baseline method in terms of both detection of critical semantic changes and semantic consistency in the obtained bi-temporal results. It achieves the SOTA accuracy on two benchmark datasets for the SCD.	['Lorenzo Bruzzone', 'Bing Liu', 'Haitao Guo', 'Kai Zhang', 'Jing Zhang', 'Lei Ding']	2022-12-10	null	null	null	null	['change-detection']	['computer-vision']	[ 6.57178283e-01 -4.73022312e-01 3.31582613e-02 -6.48576915e-01 -4.19404984e-01 -5.83814502e-01 9.08869147e-01 3.03652465e-01 -3.45072836e-01 4.94879037e-01 2.67792553e-01 -3.12423110e-01 -2.83829629e-01 -1.04466617e+00 -7.17263997e-01 -7.16425359e-01 -2.38501444e-01 -1.36004239e-01 4.77157176e-01 -2.46718392e-01 1.21746317e-01 6.09912336e-01 -1.70793307e+00 1.84589028e-01 9.92374718e-01 1.24279368e+00 4.04200166e-01 4.47643936e-01 -8.27262625e-02 6.27873838e-01 -7.94210061e-02 3.33583891e-01 2.25639120e-01 -4.34216470e-01 -8.27905893e-01 -2.13091131e-02 3.93663317e-01 -2.38758773e-01 -2.09408730e-01 1.23494565e+00 3.56551707e-01 1.97223678e-01 3.61763507e-01 -1.10960317e+00 -2.94376075e-01 3.55616361e-01 -6.86195672e-01 7.73000419e-01 -2.59336140e-02 1.29579231e-01 1.16149068e+00 -7.50005424e-01 6.47767007e-01 1.20685530e+00 7.42751896e-01 -1.14256144e-01 -8.89452040e-01 -5.66460311e-01 6.86621308e-01 7.31783807e-01 -1.33478582e+00 -4.48571682e-01 9.44257438e-01 -6.67201519e-01 9.41073358e-01 3.31200540e-01 7.35298455e-01 7.29732692e-01 5.16786166e-02 7.67583191e-01 1.10471249e+00 -1.63691863e-01 4.28234011e-01 -5.51865816e-01 1.90415725e-01 2.06414580e-01 1.69745032e-02 1.10344552e-02 -4.66930509e-01 2.82480121e-01 4.62909400e-01 3.47986758e-01 -3.05593759e-01 -1.79996446e-01 -1.15041173e+00 5.65163314e-01 9.18769062e-01 5.65962791e-01 -6.12605870e-01 2.47296304e-01 4.48365539e-01 1.12040497e-01 7.96976089e-01 2.23788530e-01 -6.20636821e-01 8.64500627e-02 -1.00991940e+00 1.50007710e-01 1.71601340e-01 5.26511431e-01 1.01032424e+00 -7.76155964e-02 -2.98836857e-01 6.08591497e-01 2.42033198e-01 6.26914024e-01 1.52588829e-01 -5.65712452e-01 4.64190483e-01 8.57050300e-01 2.66673177e-01 -1.27079713e+00 -5.20525396e-01 -6.51639283e-01 -9.14469719e-01 -1.05983131e-01 3.91895920e-02 2.33562022e-01 -1.12652445e+00 1.72289908e+00 5.55582047e-01 4.27464485e-01 -1.39704958e-01 8.41661453e-01 7.16399610e-01 6.91775680e-01 2.22638607e-01 -3.70504782e-02 1.18653715e+00 -6.53656840e-01 -7.64774323e-01 -3.77622783e-01 3.63587022e-01 -2.28683233e-01 1.08693802e+00 -2.11038962e-01 -5.05581141e-01 -4.66918439e-01 -9.49537933e-01 8.29754099e-02 -6.91078901e-01 1.65420264e-01 4.98800367e-01 -2.09427431e-01 -1.09376228e+00 5.05739272e-01 -1.20744753e+00 -6.64794385e-01 7.16668367e-01 -3.45087126e-02 -1.13856465e-01 -7.67576098e-02 -1.47243845e+00 6.97840095e-01 4.85073209e-01 5.74396491e-01 -1.07175183e+00 -9.13692772e-01 -8.96060109e-01 1.52902946e-01 3.61197859e-01 -2.67668068e-01 8.48011017e-01 -1.19484818e+00 -9.23002601e-01 8.62340927e-01 -3.51774573e-01 -4.12375659e-01 5.50943375e-01 -1.68088034e-01 -5.43699622e-01 2.44510293e-01 3.19168597e-01 8.00230443e-01 7.08565772e-01 -1.35544062e+00 -1.14267302e+00 -5.43358564e-01 1.78187564e-01 2.82599956e-01 -2.43848160e-01 -9.84911621e-02 -4.27581161e-01 -8.08090806e-01 4.68281984e-01 -7.93144047e-01 -1.68567989e-02 2.56489098e-01 -3.12139541e-01 -1.08873956e-01 1.07517052e+00 -1.08806491e+00 1.35824251e+00 -2.26279259e+00 1.31577328e-02 1.31024361e-01 9.58500579e-02 3.74216497e-01 -1.02705315e-01 2.34224871e-01 -5.10130636e-02 6.54025152e-02 -8.33057821e-01 -6.29241671e-03 -7.83310086e-02 3.52193117e-01 -3.65091741e-01 4.17381048e-01 3.14958543e-01 9.90280330e-01 -1.00970960e+00 -1.58458725e-01 2.90588200e-01 2.73101896e-01 -3.76881599e-01 1.14340544e-01 -2.69203126e-01 7.45865464e-01 -5.06511748e-01 6.31981492e-01 8.79271924e-01 -1.57748654e-01 1.31970555e-01 -3.03341508e-01 -3.78181756e-01 3.68856311e-01 -1.04167318e+00 1.48308766e+00 -3.29711318e-01 7.21487641e-01 -6.25151992e-02 -1.17214060e+00 8.86501968e-01 -8.30046162e-02 6.58480465e-01 -1.05047691e+00 -3.31698656e-01 1.74028099e-01 -2.35614955e-01 -5.73644698e-01 3.41792911e-01 -8.69736969e-02 -1.47155039e-02 1.92368701e-01 -4.16397095e-01 1.30891100e-01 -6.18622601e-02 -1.83302417e-01 1.03379214e+00 2.43590936e-01 4.43678588e-01 -4.27619159e-01 5.50719321e-01 1.22642554e-01 8.74886870e-01 6.48266077e-01 -3.97013217e-01 2.58110106e-01 2.55775481e-01 -7.45058298e-01 -6.99457765e-01 -9.90648031e-01 1.34960643e-03 8.90352309e-01 6.48210287e-01 1.62302524e-01 -3.39351177e-01 -5.74932873e-01 1.46940559e-01 6.99345112e-01 -8.08564365e-01 -2.68977821e-01 -6.55770779e-01 -8.45212042e-01 3.18597436e-01 6.20512307e-01 1.12179291e+00 -1.02065265e+00 -7.86349118e-01 2.72770792e-01 -4.13024187e-01 -1.13900137e+00 -1.60659179e-01 9.92778167e-02 -7.06149340e-01 -1.16942954e+00 -1.08085841e-01 -5.43705523e-01 4.35435027e-01 5.81124067e-01 7.72751808e-01 -2.00314730e-01 -1.56043321e-01 8.48601982e-02 -5.64049184e-01 -1.53464735e-01 -9.34005529e-03 9.44254398e-02 -4.25807863e-01 2.72586346e-01 2.88029343e-01 -7.68765807e-01 -8.70230317e-01 3.29345345e-01 -1.21473551e+00 2.34939471e-01 3.76989126e-01 5.09959996e-01 6.31950617e-01 4.64690298e-01 5.12664080e-01 -5.76070666e-01 -4.88779927e-03 -5.49601734e-01 -3.57118458e-01 4.32005167e-01 -6.50368690e-01 -1.26016662e-01 3.66410255e-01 -9.30564478e-03 -1.29835081e+00 -6.97059035e-02 -9.87040550e-02 -1.29566535e-01 -3.28153372e-01 7.42775917e-01 -3.39607179e-01 1.95487067e-01 3.22567284e-01 4.92846578e-01 -6.03163958e-01 -6.01705790e-01 1.83820382e-01 6.22612894e-01 6.32491410e-01 -2.11055607e-01 6.75973237e-01 1.11063492e+00 -1.69675708e-01 -7.55788684e-01 -1.06528294e+00 -7.63904750e-01 -9.16948020e-01 -2.70719379e-01 8.16068828e-01 -1.12792885e+00 -7.25711929e-03 9.22995865e-01 -9.64188457e-01 -5.00824153e-01 -3.11466128e-01 3.01895738e-01 -2.64813513e-01 2.48255730e-01 -1.74863309e-01 -4.68823344e-01 -3.66128325e-01 -8.26555133e-01 1.33224833e+00 9.43149775e-02 4.48946990e-02 -1.07255042e+00 -9.27251857e-03 6.19041547e-03 6.27145827e-01 6.51767910e-01 1.05178916e+00 -3.09953541e-01 -4.79759723e-01 1.75247803e-01 -5.35488069e-01 1.61680549e-01 6.54379964e-01 -9.40772370e-02 -9.28229153e-01 -3.36807400e-01 -7.77181536e-02 2.93818146e-01 1.29336321e+00 4.95155931e-01 1.21282136e+00 -4.21365827e-01 -4.30971980e-01 7.58132756e-01 1.61140049e+00 3.20826292e-01 7.16977239e-01 5.38422525e-01 8.23023617e-01 5.52649379e-01 7.58225024e-01 6.59900129e-01 8.23224664e-01 8.61590564e-01 7.24207282e-01 -2.69535720e-01 -3.25038463e-01 -1.93011686e-01 2.86092997e-01 5.10536611e-01 1.31681114e-01 -3.01534291e-02 -1.19499922e+00 9.18178558e-01 -1.93869221e+00 -9.66395020e-01 -2.07093969e-01 1.95776713e+00 9.37306285e-01 -1.16388276e-01 -2.84754038e-01 8.38776454e-02 1.01836610e+00 6.93408728e-01 -9.53652799e-01 2.92606562e-01 -2.49544919e-01 9.76129249e-02 5.05237222e-01 4.91930723e-01 -1.51115787e+00 1.08123207e+00 5.48198462e+00 6.05730116e-01 -1.39389360e+00 4.50051665e-01 3.67650360e-01 1.58641383e-01 -4.51382548e-01 5.26686944e-02 -7.37839460e-01 4.36215401e-01 4.65203315e-01 2.02639297e-01 3.23991805e-01 3.01890850e-01 7.69993722e-01 -2.81271785e-01 -8.11094105e-01 7.01177239e-01 -8.32981169e-02 -1.10631883e+00 1.71970621e-01 -3.56976926e-01 1.04784966e+00 2.25711316e-01 -2.33987406e-01 7.18921348e-02 2.87353486e-01 -6.20390117e-01 9.89108741e-01 7.04943478e-01 8.21221828e-01 -3.69893640e-01 6.42728806e-01 1.36718065e-01 -1.62311673e+00 -1.90723851e-01 -4.32681702e-02 -4.50375937e-02 -4.12232094e-02 8.16400349e-01 -6.03586555e-01 8.21354687e-01 9.77513671e-01 1.55361521e+00 -7.59246111e-01 8.84347200e-01 -3.85712236e-01 7.74057388e-01 -2.47246087e-01 4.93881911e-01 3.90728623e-01 -1.41911373e-01 6.32485390e-01 1.10203230e+00 4.03257608e-01 4.39792173e-03 1.44170687e-01 9.50074255e-01 7.31530562e-02 -3.23157221e-01 -3.51824433e-01 8.43795538e-02 6.24933898e-01 1.00231946e+00 -8.29273045e-01 -2.69270480e-01 -9.62768495e-02 9.58474994e-01 5.06598689e-02 5.36535680e-01 -7.86211610e-01 -2.54123390e-01 8.81429911e-01 -4.33713943e-02 6.10690355e-01 -2.69114554e-01 -4.42380786e-01 -1.16084218e+00 3.37848157e-01 -4.75606501e-01 5.36935031e-01 -7.74031997e-01 -9.79501724e-01 2.43708134e-01 4.67227548e-02 -1.29437983e+00 2.53191739e-01 -6.00065961e-02 -7.53467977e-01 6.38416469e-01 -2.41949606e+00 -1.49457002e+00 -9.25932348e-01 5.18347561e-01 5.42387486e-01 5.07285297e-01 2.95864969e-01 3.27397883e-01 -5.98134995e-01 1.18137464e-01 2.17795283e-01 -2.91985199e-02 3.56596053e-01 -1.08869100e+00 4.18081015e-01 1.48603642e+00 -3.14328045e-01 9.31223035e-02 4.98176396e-01 -7.55785584e-01 -9.06324208e-01 -1.85188735e+00 9.10505295e-01 7.19409138e-02 7.51493037e-01 -7.24493042e-02 -1.08104956e+00 5.66597819e-01 -3.46876979e-01 4.84395474e-02 1.96921393e-01 -2.37757578e-01 -3.43727082e-01 -5.40907204e-01 -9.23309445e-01 3.56647521e-01 1.48202932e+00 -7.38613963e-01 -4.62209553e-01 1.97955698e-01 9.10353541e-01 -8.45004171e-02 -6.97577536e-01 7.25038111e-01 2.34429762e-01 -7.24105239e-01 9.14896071e-01 -4.04584169e-01 4.82133001e-01 -7.56462932e-01 -4.35106218e-01 -1.32084167e+00 -4.14486915e-01 3.15359235e-02 1.31727904e-01 1.18939018e+00 -4.03567217e-02 -6.97030067e-01 1.25221968e-01 -4.69594598e-02 -3.44144970e-01 -3.85830402e-01 -9.84343588e-01 -7.99463630e-01 -1.00743063e-01 -5.08975685e-01 1.02124667e+00 1.24345624e+00 -6.23740435e-01 -1.37921944e-01 -1.59105241e-01 6.75280571e-01 2.09010154e-01 4.81187791e-01 3.28597426e-01 -1.31285691e+00 2.74617851e-01 -6.65818155e-01 -5.09244323e-01 -8.71722996e-01 1.21505298e-01 -1.05691993e+00 2.26961553e-01 -1.85992241e+00 1.81807697e-01 -5.73205352e-01 -6.15661740e-01 9.28662956e-01 -6.02923870e-01 3.27940620e-02 -1.27530977e-01 5.59323967e-01 -6.04573548e-01 9.12451804e-01 1.09773684e+00 -4.46920305e-01 -2.68729538e-01 -1.69636264e-01 -5.09164572e-01 4.12108630e-01 8.12309206e-01 -3.68854910e-01 -2.70460635e-01 -5.89027345e-01 3.42043102e-01 -3.67870629e-01 8.52385223e-01 -1.09470427e+00 8.00376907e-02 -4.64064598e-01 1.41911069e-02 -8.86266887e-01 -2.92162985e-01 -8.26731563e-01 4.43768412e-01 5.52320778e-01 -4.84372914e-01 -2.83646673e-01 1.78188995e-01 7.24783719e-01 -4.89577651e-01 1.35444552e-01 8.89309645e-01 -2.03224011e-02 -1.42270505e+00 5.12614369e-01 -1.33099347e-01 -5.00536263e-02 1.04125834e+00 -1.07578613e-01 -3.72091323e-01 -1.33916542e-01 -6.70652270e-01 3.86217028e-01 4.05368209e-01 5.37270665e-01 4.42930192e-01 -1.36264396e+00 -6.17165446e-01 2.28246078e-01 4.96258587e-01 2.51422733e-01 4.71775532e-01 1.07221222e+00 -4.86291975e-01 3.22939336e-01 -1.13005295e-01 -8.04272234e-01 -9.59829569e-01 2.26032078e-01 6.18203998e-01 -3.13295335e-01 -7.87713408e-01 5.19761384e-01 2.53982902e-01 -5.88635325e-01 -1.60757259e-01 -6.28163099e-01 -3.76628935e-01 3.92590821e-01 3.74265641e-01 3.64817202e-01 3.17436129e-01 -5.71454942e-01 -6.05987728e-01 5.06314874e-01 3.89216900e-01 1.32909000e-01 1.82111859e+00 -5.61207354e-01 -2.85869300e-01 5.37574232e-01 1.23064375e+00 -6.17283940e-01 -1.66701603e+00 -8.55822980e-01 2.83007056e-01 -5.29936314e-01 3.20440888e-01 -9.23105180e-01 -1.19805074e+00 9.19196784e-01 9.48831141e-01 1.75379659e-03 1.44501507e+00 6.67215744e-03 7.56642103e-01 3.23200852e-01 2.40991309e-01 -1.26300335e+00 -1.50972262e-01 6.23005807e-01 9.25542891e-01 -1.36845994e+00 -1.61860853e-01 -3.82727861e-01 -3.99169922e-01 8.57572675e-01 3.95986944e-01 2.29799166e-01 8.48732173e-01 -2.09708139e-01 -1.62741676e-01 -5.49467742e-01 -3.99303526e-01 -6.75114989e-01 4.11403328e-01 4.10217285e-01 -1.14868484e-01 3.54828954e-01 -6.10652119e-02 2.08095998e-01 1.80120394e-01 3.99834588e-02 7.70515949e-02 1.07360530e+00 -6.63186014e-01 -4.88860071e-01 -1.15105718e-01 4.79729682e-01 5.10297753e-02 -8.48766789e-02 -2.53582209e-01 6.36148214e-01 4.76937950e-01 9.95371938e-01 3.74832124e-01 -3.74161184e-01 3.20407540e-01 -8.05969164e-02 -3.73038538e-02 -4.57897276e-01 -3.56575727e-01 -1.46184862e-01 -1.17224030e-01 -7.45770812e-01 -9.76275325e-01 -1.02687883e+00 -1.23275018e+00 -6.01019785e-02 -2.74494453e-03 -3.17170858e-01 5.97114861e-01 1.19063354e+00 5.45962989e-01 7.08850801e-01 1.02394736e+00 -6.20274723e-01 -1.35082439e-01 -1.02695966e+00 -6.21840894e-01 6.71889126e-01 5.14893472e-01 -8.65185797e-01 -3.74835402e-01 2.38535568e-01]	[9.683335304260254, -1.3023813962936401]
f125ffb0-d6b4-423d-9ada-88568c543ebd	automated-essay-scoring-using-transformers	2210.12809	null	https://arxiv.org/abs/2210.12809v5	https://arxiv.org/pdf/2210.12809v5.pdf	Data Augmentation for Automated Essay Scoring using Transformer Models	Automated essay scoring is one of the most important problem in Natural Language Processing. It has been explored for a number of years, and it remains partially solved. In addition to its economic and educational usefulness, it presents research problems. Transfer learning has proved to be beneficial in NLP. Data augmentation techniques have also helped build state-of-the-art models for automated essay scoring. Many works in the past have attempted to solve this problem by using RNNs, LSTMs, etc. This work examines the transformer models like BERT, RoBERTa, etc. We empirically demonstrate the effectiveness of transformer models and data augmentation for automated essay grading across many topics using a single model.	['Kshitij Gupta']	2022-10-23	null	null	null	null	['automated-essay-scoring']	['natural-language-processing']	[-2.49173135e-01 3.99430394e-02 -2.85308003e-01 -5.01988769e-01 -5.90989053e-01 -2.62118042e-01 5.20003021e-01 3.52801263e-01 -4.71503764e-01 1.03774476e+00 4.85966474e-01 -4.77260560e-01 -3.84556621e-01 -7.67641902e-01 -8.21621716e-02 -3.43229115e-01 3.12323749e-01 6.58867478e-01 2.26768598e-01 -6.55570924e-01 7.06729591e-01 2.47540712e-01 -1.47363997e+00 2.75573134e-01 1.07172608e+00 6.59121633e-01 -1.32122787e-03 7.40680277e-01 -7.54745722e-01 1.56651247e+00 -1.01598990e+00 -7.30227947e-01 -2.59846240e-01 -5.42365074e-01 -1.09158862e+00 -3.46069455e-01 6.23478472e-01 -2.27628663e-01 -1.68864459e-01 8.19963992e-01 4.71878558e-01 3.80457252e-01 7.79453635e-01 -1.01699364e+00 -1.20030379e+00 4.86235112e-01 -6.66497886e-01 1.51679575e-01 5.08381188e-01 -4.74616110e-01 1.13594818e+00 -7.68530071e-01 1.78774923e-01 8.94821048e-01 8.45504940e-01 5.89371145e-01 -8.16143513e-01 -7.93954551e-01 -2.43167236e-01 7.53542364e-01 -4.72933918e-01 9.20560732e-02 9.12143469e-01 -3.18632156e-01 7.80746222e-01 -4.41464186e-02 6.64840937e-01 8.89019907e-01 2.13203937e-01 1.27392876e+00 1.70411921e+00 -9.46018755e-01 -1.28567576e-01 3.06091398e-01 9.09978271e-01 9.47653890e-01 -6.96046054e-02 -2.83963919e-01 -7.58177459e-01 -1.95180147e-03 5.11248767e-01 -1.67388052e-01 -8.88526905e-03 2.00110823e-01 -6.61831379e-01 1.19707453e+00 1.45573720e-01 5.67992151e-01 -1.95766650e-02 -2.05534413e-01 5.51337779e-01 7.89619267e-01 8.25639248e-01 7.40191579e-01 -6.27217531e-01 -5.76332510e-01 -1.12812936e+00 2.25905523e-01 9.34314787e-01 4.56754029e-01 2.85535693e-01 2.62839228e-01 -3.02246302e-01 1.30363584e+00 9.93900895e-02 1.36164241e-02 1.10528314e+00 -6.37942910e-01 4.51521665e-01 9.97371018e-01 -6.38950318e-02 -9.63352978e-01 -3.72883379e-01 -2.73482531e-01 -8.13857913e-01 2.63567626e-01 6.32201314e-01 -2.45795697e-01 -7.69954801e-01 1.35716212e+00 -9.21961963e-02 7.06356242e-02 -3.09686452e-01 6.00295901e-01 1.12352180e+00 8.27777624e-01 1.68606907e-01 1.94131032e-01 1.16990066e+00 -1.26977742e+00 -9.88191485e-01 6.99728057e-02 7.54284143e-01 -8.34603429e-01 1.16809344e+00 8.34749997e-01 -1.35330689e+00 -5.77810287e-01 -8.81996632e-01 -3.66929591e-01 -6.75091743e-01 3.03492814e-01 8.04608226e-01 1.00486636e+00 -1.18725133e+00 5.99276602e-01 -3.41193199e-01 -4.65575069e-01 3.43585730e-01 5.23090899e-01 -4.97157797e-02 6.77472502e-02 -1.35026860e+00 1.45336986e+00 -7.01154992e-02 -2.68090844e-01 -2.74698377e-01 -5.48127234e-01 -5.77118754e-01 3.41589123e-01 -2.80043446e-02 -4.59141761e-01 1.55646217e+00 -6.06404185e-01 -2.12612629e+00 7.47194290e-01 -1.12633616e-01 -4.77069497e-01 8.07395056e-02 -3.17496628e-01 -2.74663389e-01 -2.53571481e-01 -2.07347095e-01 4.00115907e-01 2.22254366e-01 -4.84861106e-01 -6.53796315e-01 -6.20277107e-01 5.03998362e-02 3.27582151e-01 -1.20260036e+00 5.15826404e-01 3.91044557e-01 -7.22126186e-01 -1.02323823e-01 -6.53609991e-01 -3.24298814e-02 -5.81356108e-01 1.57248080e-01 -9.71175373e-01 8.07120800e-01 -9.71995354e-01 1.32441509e+00 -1.33725572e+00 9.20536369e-02 -3.22358049e-02 2.14928780e-02 5.43480039e-01 7.13538798e-03 4.38640952e-01 6.95052072e-02 1.12474509e-01 1.97066128e-01 -5.10278642e-01 -2.69638523e-02 -3.18770250e-03 -2.29315907e-01 3.01034395e-02 -2.44230896e-01 9.88981009e-01 -7.69038856e-01 -6.20658457e-01 1.72712222e-01 1.67075634e-01 -1.70904025e-01 3.03431332e-01 -2.80069113e-02 -1.04997177e-02 -3.61564487e-01 5.69639862e-01 1.27654836e-01 -8.53672475e-02 -3.10887128e-01 5.18121183e-01 -2.35169843e-01 6.85753047e-01 -8.09786081e-01 1.48850822e+00 -4.09727097e-01 1.16563165e+00 -2.01436877e-01 -1.01120186e+00 1.26026285e+00 6.13425612e-01 2.86652714e-01 -8.24359059e-01 2.58004248e-01 1.67576790e-01 2.10416242e-02 -6.47969484e-01 1.02569497e+00 -2.85431564e-01 1.27542973e-01 9.37921584e-01 2.01376274e-01 -2.20493704e-01 2.68729120e-01 5.56294769e-02 9.76967692e-01 2.59379774e-01 3.37407649e-01 -5.68725429e-02 7.43260145e-01 1.27131671e-01 1.18498862e-01 5.38377523e-01 -2.17693612e-01 4.23738658e-01 2.44191021e-01 -6.00098252e-01 -9.25824106e-01 -5.68000317e-01 3.63803096e-02 1.61887753e+00 -5.25257766e-01 -1.44617716e-02 -8.26635957e-01 -5.19367218e-01 -2.25051090e-01 6.35774612e-01 -6.61012650e-01 -2.74818968e-02 -5.29678702e-01 -7.50047982e-01 7.57163703e-01 7.63712764e-01 9.16607141e-01 -1.43872190e+00 -2.91036010e-01 4.45980519e-01 -2.19873965e-01 -7.10308194e-01 -1.09251007e-01 2.37104714e-01 -1.12385142e+00 -9.29475427e-01 -8.26735556e-01 -1.04678273e+00 3.29260647e-01 3.64236355e-01 1.31310236e+00 3.07922006e-01 -2.05686837e-01 3.47854882e-01 -3.94986480e-01 -1.02199662e+00 -1.52642757e-01 5.26698947e-01 -1.79043785e-01 -3.70169371e-01 9.75289643e-01 -3.79809946e-01 -1.66711677e-02 7.57861063e-02 -5.21394372e-01 -1.94423765e-01 5.30497372e-01 1.06163120e+00 -1.64636910e-01 -2.68257022e-01 1.08654368e+00 -1.12087071e+00 1.72147369e+00 -2.01022819e-01 -3.61265928e-01 3.60454559e-01 -6.85530186e-01 -8.12696591e-02 6.60303712e-01 -3.22652578e-01 -1.24940848e+00 -3.69773239e-01 -1.06032126e-01 1.68383792e-01 -3.84185284e-01 9.15365696e-01 4.01722193e-01 -3.32209647e-01 6.44350290e-01 3.30959037e-02 -4.20824848e-02 -4.65604097e-01 -6.49364218e-02 7.65423298e-01 1.43964693e-01 -7.00641751e-01 4.23877031e-01 -1.24991119e-01 -2.63358355e-02 -9.04262602e-01 -1.34911656e+00 -6.47679150e-01 -8.32915485e-01 -4.25739080e-01 6.46751106e-01 -4.44270939e-01 -8.84484768e-01 6.51839972e-01 -1.13569808e+00 -2.07570821e-01 -1.47651136e-01 3.72988164e-01 -1.70406342e-01 1.60237417e-01 -1.08364284e+00 -8.47703576e-01 -6.14809930e-01 -8.10787022e-01 2.17636585e-01 6.87666953e-01 -4.83375728e-01 -1.36445832e+00 5.49021184e-01 7.83020794e-01 6.26257241e-01 -2.62245297e-01 1.09077787e+00 -8.21811497e-01 -3.89234312e-02 -3.75482976e-01 -1.14571325e-01 3.42074811e-01 -7.21902475e-02 3.20142843e-02 -1.08706403e+00 1.47160918e-01 1.43165275e-01 -7.13041008e-01 8.94412875e-01 4.23582673e-01 1.33184302e+00 -1.02546744e-01 2.30336875e-01 -9.62181538e-02 9.29262936e-01 2.57973582e-01 5.96444845e-01 6.10449553e-01 3.82382691e-01 7.35597432e-01 5.38797140e-01 -1.52181359e-02 6.65924728e-01 3.07188660e-01 1.14511043e-01 -9.65668187e-02 -7.57872760e-02 -9.49124917e-02 3.22073877e-01 1.34773648e+00 -4.01118785e-01 -7.89169073e-02 -8.88815224e-01 6.50534987e-01 -1.81652677e+00 -1.08160579e+00 -5.95521927e-01 1.86988688e+00 8.58178735e-01 -8.39115307e-02 2.86957979e-01 7.37721741e-01 4.12780970e-01 6.17594533e-02 -1.18311614e-01 -1.04711890e+00 1.54134203e-02 7.52709627e-01 1.25655621e-01 4.66361254e-01 -9.07971263e-01 1.00762022e+00 6.61725187e+00 5.38569689e-01 -7.87709475e-01 2.20156953e-01 8.10971916e-01 1.55893013e-01 1.72101632e-02 -3.17935914e-01 -8.27505231e-01 3.21623325e-01 1.09429419e+00 -1.42107144e-01 8.88887346e-02 7.54667342e-01 1.79801553e-01 -2.60066718e-01 -7.02909827e-01 4.57646817e-01 4.37901229e-01 -1.07433164e+00 2.57274453e-02 -4.18715514e-02 1.08098054e+00 -2.58360654e-01 3.48179221e-01 9.12808001e-01 6.16625667e-01 -1.28708041e+00 2.29706746e-02 3.67674798e-01 1.52662829e-01 -9.97532308e-01 1.12671995e+00 4.70490217e-01 -7.16977656e-01 -2.48414785e-01 -7.40301251e-01 -8.97936225e-01 -2.77885258e-01 4.32752609e-01 -9.66320872e-01 1.48204759e-01 5.84127367e-01 6.11420810e-01 -7.76250362e-01 1.32958710e+00 -7.29615450e-01 1.22329056e+00 1.54585019e-01 -9.20687735e-01 5.33644736e-01 -6.17915750e-01 -7.29616284e-02 1.20560706e+00 2.51837969e-01 2.95282692e-01 -2.46693869e-03 4.82964784e-01 -3.21108639e-01 4.63985950e-01 -5.41794717e-01 1.72670662e-01 2.29357451e-01 1.38720095e+00 -4.00949031e-01 -3.88756692e-01 -3.71780962e-01 5.07711828e-01 5.89568555e-01 8.20686892e-02 -5.27359366e-01 -5.44683278e-01 1.85870990e-01 -6.59120306e-02 -4.30326402e-01 -1.44900963e-01 -8.74214113e-01 -8.52142215e-01 -1.68810144e-01 -9.88273561e-01 5.31105459e-01 -9.20490921e-01 -1.25431585e+00 2.56499946e-01 -4.49806333e-01 -9.19471085e-01 -1.70473397e-01 -8.37050080e-01 -1.01078010e+00 8.75339389e-01 -1.67406952e+00 -9.34888184e-01 -4.62753505e-01 5.29411495e-01 1.00430453e+00 -7.22233891e-01 1.24846029e+00 2.91966051e-01 -5.07347524e-01 5.73280931e-01 2.95968145e-01 3.25490355e-01 9.71436799e-01 -1.71983767e+00 9.81284752e-02 4.16734964e-01 2.33372107e-01 4.42738205e-01 5.10489345e-01 -2.82083333e-01 -8.51991534e-01 -5.91236591e-01 1.45695746e+00 -5.36425531e-01 8.39718759e-01 8.82837772e-02 -9.97811615e-01 5.96659601e-01 7.70365477e-01 -7.52512157e-01 9.03769732e-01 6.53889239e-01 -1.38286352e-01 -9.57693383e-02 -1.09240162e+00 6.89628541e-01 1.34658009e-01 -2.93789595e-01 -9.59033132e-01 2.92581409e-01 1.79685965e-01 -2.71082371e-01 -8.03130448e-01 2.98140515e-02 5.75403094e-01 -1.04561150e+00 6.78685069e-01 -9.01609480e-01 1.11442721e+00 4.46221054e-01 6.75679445e-01 -1.75214207e+00 -4.38008785e-01 -4.02307302e-01 -2.63774633e-01 1.33973551e+00 4.45558757e-01 -3.32192957e-01 1.36665297e+00 7.85980642e-01 -1.83181450e-01 -9.35366750e-01 -7.15912580e-01 -4.60601360e-01 5.36268413e-01 -3.57056707e-02 4.56809282e-01 1.29349017e+00 6.18864298e-01 8.10642958e-01 -3.93695772e-01 -5.38117647e-01 3.58227223e-01 -9.80827510e-02 6.74935639e-01 -1.90146101e+00 1.98119313e-01 -9.68498290e-01 -1.73632681e-01 -9.60601389e-01 3.62411529e-01 -6.37180448e-01 -2.26725116e-01 -1.98223567e+00 2.99710393e-01 -2.19870135e-01 -3.55902046e-01 7.16551125e-01 -3.36339116e-01 2.30513334e-01 -1.63235869e-02 -2.06064165e-01 -4.10987288e-01 4.56594259e-01 1.26091611e+00 -2.60665715e-01 -9.29528549e-02 2.36326009e-01 -7.18826532e-01 8.79210293e-01 1.43012035e+00 -5.13442218e-01 -3.33335370e-01 -5.54054499e-01 4.98865277e-01 1.21139176e-01 -2.66595334e-01 -9.00758743e-01 7.01904893e-01 -2.31653191e-02 5.10850906e-01 -6.44842684e-01 4.75594968e-01 -5.99418938e-01 -8.54772508e-01 2.06158236e-01 -7.01078296e-01 5.44569373e-01 1.12401627e-01 7.37909926e-03 -5.76052725e-01 -9.86145258e-01 6.45729125e-01 -4.12046790e-01 -3.83385420e-01 -1.30726427e-01 -5.82993090e-01 -8.08494911e-02 7.98836350e-01 -2.66391009e-01 -4.44976240e-01 -6.57812715e-01 -2.55188763e-01 3.43578905e-01 -2.09451020e-01 6.00674570e-01 6.62439823e-01 -9.67261553e-01 -9.37687755e-01 -1.87633976e-01 -3.45051199e-01 -3.86275232e-01 -7.18489289e-02 6.35416150e-01 -4.90905315e-01 8.39133501e-01 -4.91737992e-01 -4.24858145e-02 -1.80377197e+00 -2.66889725e-02 1.20986648e-01 -9.48894441e-01 -2.42395446e-01 8.74900401e-01 -5.03943682e-01 -7.72307098e-01 4.60229307e-01 -1.49894804e-01 -1.14221334e+00 2.34272897e-01 6.63933098e-01 8.74424517e-01 4.08756673e-01 -1.96185801e-02 4.60085422e-01 3.62173051e-01 -5.10674477e-01 -6.57777339e-02 1.67604208e+00 3.42130303e-01 -2.58869916e-01 6.58349395e-01 5.62857985e-01 -7.84353986e-02 -2.46752605e-01 -1.61645040e-01 4.95444089e-01 -1.42294466e-01 1.62671745e-01 -1.37307823e+00 -8.33389640e-01 1.31718194e+00 7.92967081e-02 3.61767739e-01 7.49861419e-01 -7.60062099e-01 6.73799336e-01 7.43016541e-01 2.63158411e-01 -1.42951894e+00 3.56487870e-01 1.12399375e+00 5.72685182e-01 -1.35510242e+00 2.58334637e-01 1.39693350e-01 -5.23273528e-01 1.39633536e+00 8.68219614e-01 -1.41045451e-01 3.15522790e-01 2.58510441e-01 1.08408920e-01 -1.01482339e-01 -6.99612796e-01 2.68100888e-01 3.37093532e-01 3.03677648e-01 1.34058380e+00 9.26271603e-02 -6.50611460e-01 6.23572350e-01 -5.30836880e-01 2.08785951e-01 1.02522731e+00 7.72341013e-01 -7.99841583e-01 -1.40366483e+00 -6.30656064e-01 8.03778470e-01 -8.34281147e-01 -2.33517930e-01 -8.93207192e-01 8.34885955e-01 -3.95930976e-01 1.10044193e+00 -4.08132434e-01 -1.85144633e-01 1.89290732e-01 5.76473236e-01 5.58231890e-01 -6.22509062e-01 -1.45614815e+00 -6.92024231e-01 1.24828577e-01 2.36208007e-01 -3.85456204e-01 -5.89412689e-01 -8.53369296e-01 -3.88273418e-01 -5.31468511e-01 5.63398361e-01 8.49718869e-01 1.17289388e+00 -1.22018762e-01 8.43729615e-01 2.60960937e-01 -5.42033911e-01 -8.73093545e-01 -1.62417006e+00 -6.06497467e-01 -5.93189336e-02 -3.03881951e-02 -5.10128260e-01 -1.39441445e-01 -1.67450592e-01]	[11.293671607971191, 9.318516731262207]
cfb8697e-ee80-4363-a951-3d22c1adecf2	mutual-learning-for-domain-adaptation-self	2203.09430	null	https://arxiv.org/abs/2203.09430v1	https://arxiv.org/pdf/2203.09430v1.pdf	Mutual Learning for Domain Adaptation: Self-distillation Image Dehazing Network with Sample-cycle	Deep learning-based methods have made significant achievements for image dehazing. However, most of existing dehazing networks are concentrated on training models using simulated hazy images, resulting in generalization performance degradation when applied on real-world hazy images because of domain shift. In this paper, we propose a mutual learning dehazing framework for domain adaption. Specifically, we first devise two siamese networks: a teacher network in the synthetic domain and a student network in the real domain, and then optimize them in a mutual learning manner by leveraging EMA and joint loss. Moreover, we design a sample-cycle strategy based on density augmentation (HDA) module to introduce pseudo real-world image pairs provided by the student network into training for further improving the generalization performance. Extensive experiments on both synthetic and real-world dataset demonstrate that the propose mutual learning framework outperforms state-of-the-art dehazing techniques in terms of subjective and objective evaluation.	['ErKang Chen', 'Sixiang Chen', 'Yunchen Zhang', 'Yun Liu', 'Tian Ye']	2022-03-17	null	null	null	null	['image-dehazing']	['computer-vision']	[ 2.40392298e-01 -2.51713395e-01 1.94248900e-01 -3.29810679e-01 -5.29431403e-01 7.39191100e-02 4.28575486e-01 -7.61147067e-02 -6.66484118e-01 8.22809935e-01 -1.30955487e-01 -4.15088870e-02 -1.10188015e-01 -1.17858875e+00 -7.59164333e-01 -1.03533792e+00 2.63321429e-01 1.27044842e-01 4.85317856e-01 -1.84025168e-01 7.05615282e-02 6.38047382e-02 -1.39320326e+00 -1.58185050e-01 1.63801062e+00 8.42929423e-01 2.71441460e-01 3.62968147e-01 9.96914804e-02 8.85643780e-01 -9.66365874e-01 -3.62955689e-01 3.75422269e-01 -5.82323194e-01 -3.46836179e-01 2.42103249e-01 1.87788442e-01 -5.11453748e-01 -6.55761719e-01 1.32094836e+00 6.03184164e-01 5.40677607e-01 9.26779926e-01 -1.17694461e+00 -1.14526343e+00 1.24300487e-01 -7.87956715e-01 1.80041432e-01 -2.01853424e-01 3.03236783e-01 4.62125808e-01 -8.62986088e-01 6.42695948e-02 1.11344969e+00 5.05782306e-01 7.50740647e-01 -1.07090354e+00 -1.16431987e+00 2.65169963e-02 4.69607174e-01 -1.54273129e+00 -1.13020442e-01 1.08511603e+00 -3.20999742e-01 3.16928834e-01 -1.49371237e-01 3.31666082e-01 8.75858486e-01 7.22915977e-02 9.68234181e-01 1.13720536e+00 -3.16508263e-01 3.13648641e-01 5.87494075e-01 -4.39289063e-01 3.95165026e-01 2.14449480e-01 2.41062611e-01 -3.10692817e-01 1.80108055e-01 7.99365163e-01 8.21692869e-02 -4.06092823e-01 -5.91992497e-01 -8.13076079e-01 8.73175979e-01 6.87898278e-01 8.33130851e-02 -2.48745352e-01 -2.39447534e-01 3.72087546e-02 4.98645276e-01 8.57023418e-01 3.09773207e-01 -1.18322466e-02 4.74393487e-01 -1.00173151e+00 2.78866649e-01 4.51390445e-01 8.51584911e-01 1.11504018e+00 3.19710255e-01 -6.27939701e-02 1.03439498e+00 2.45849565e-01 4.78749841e-01 7.51990497e-01 -4.80531186e-01 3.88304472e-01 4.84851629e-01 1.11083202e-02 -1.04344094e+00 2.26050556e-01 -6.08308494e-01 -1.08627057e+00 5.50917983e-01 4.72694337e-02 -3.63690585e-01 -1.24262977e+00 1.49631584e+00 4.34979349e-01 8.40630829e-01 4.41057622e-01 9.74711835e-01 6.86232567e-01 9.92106199e-01 1.74076661e-01 -2.19628122e-02 7.59535551e-01 -1.14644456e+00 -7.83543706e-01 -1.90667003e-01 5.43148339e-01 -7.05750823e-01 9.17109728e-01 4.69939023e-01 -1.02519429e+00 -6.59431636e-01 -1.42420483e+00 9.72142592e-02 -5.14261723e-01 -1.14960343e-01 -4.75268550e-02 5.73133588e-01 -7.71750629e-01 4.96261299e-01 -5.55140376e-01 -3.27393897e-02 6.49390757e-01 2.95536071e-01 -1.99883983e-01 -2.36677989e-01 -1.51368773e+00 8.04915905e-01 7.56351233e-01 -1.39575914e-01 -1.43121433e+00 -8.85031700e-01 -9.05852675e-01 -9.35080647e-02 2.69141793e-01 -6.74093962e-01 8.49186897e-01 -1.08419538e+00 -1.68651056e+00 7.16454148e-01 4.82559592e-01 -7.12000251e-01 7.40820110e-01 -4.37800467e-01 -6.86938643e-01 1.10369138e-01 -1.40442699e-01 7.94560552e-01 1.38920987e+00 -1.47177947e+00 -6.60083532e-01 -1.98693722e-01 -6.01774901e-02 5.65937221e-01 -8.02346289e-01 -4.69694018e-01 -8.72159004e-02 -1.11246324e+00 -1.60946205e-01 -4.06113774e-01 -1.20021746e-01 1.16162233e-01 -9.52273756e-02 -5.50647415e-02 1.29550147e+00 -7.03182042e-01 1.17583787e+00 -2.40183902e+00 1.19454667e-01 -1.45990942e-02 2.73788631e-01 8.72451305e-01 -5.20746224e-02 7.50202537e-02 -1.04310855e-01 -8.34168792e-02 -6.19560182e-01 -4.24391389e-01 -3.41422647e-01 2.50158608e-01 -2.81289816e-01 5.65948963e-01 2.91499972e-01 3.51848692e-01 -1.04126370e+00 -6.60300612e-01 4.33283120e-01 6.05483592e-01 -5.83290517e-01 7.19763100e-01 -2.58944958e-01 5.45744002e-01 -2.07141668e-01 2.38630116e-01 1.19709933e+00 2.98053287e-02 -5.04012883e-01 -9.53985285e-03 1.01323627e-01 -1.25367790e-01 -8.45927656e-01 1.60492802e+00 -5.78631997e-01 5.02263427e-01 -3.02429926e-02 -1.12845075e+00 1.14268959e+00 1.97918832e-01 -4.88143489e-02 -7.53944695e-01 1.50309637e-01 5.35804294e-02 -1.46848708e-01 -5.09969115e-01 3.02057832e-01 -4.86367822e-01 4.82613772e-01 2.00725034e-01 2.97004789e-01 -4.63871151e-01 -1.66321054e-01 2.52159461e-02 6.26593828e-01 2.20337021e-03 -5.01341335e-02 -1.16372339e-01 9.03513074e-01 -2.27908567e-01 6.90858483e-01 3.65214288e-01 -4.12410647e-01 6.92415833e-01 5.88359348e-02 -1.38404518e-01 -1.07825184e+00 -1.16307151e+00 -3.52703147e-02 7.42479742e-01 6.24926746e-01 6.96467832e-02 -9.26077068e-01 -7.60483384e-01 -1.42055750e-01 9.30063069e-01 -7.75753796e-01 -8.44611943e-01 -3.40076298e-01 -7.73347735e-01 3.95330846e-01 7.88850039e-02 1.37763381e+00 -7.47902691e-01 -2.00498939e-01 1.09351389e-01 6.24392852e-02 -1.02987385e+00 -3.97774369e-01 -2.12993637e-01 -7.22960353e-01 -7.27356136e-01 -1.17783761e+00 -1.01809120e+00 4.46097285e-01 5.65245688e-01 5.53510845e-01 -6.18707016e-02 -2.50606481e-02 -2.91822366e-02 -3.65406156e-01 -5.39930165e-01 -4.51577902e-01 -2.28460524e-02 8.43652617e-03 4.93413270e-01 3.93742621e-01 -9.99108613e-01 -9.02610481e-01 4.31870818e-01 -1.32255149e+00 -9.32198614e-02 8.63510668e-01 8.37551057e-01 2.90659547e-01 6.95540607e-01 5.80273390e-01 -8.57105017e-01 5.03731906e-01 -7.13983655e-01 -6.56402826e-01 1.49343371e-01 -8.30677748e-01 -2.73587573e-02 5.94692051e-01 -5.54127157e-01 -1.56215346e+00 -4.10092115e-01 5.79108447e-02 -8.72115612e-01 -2.08391890e-01 3.37019116e-01 -2.70382941e-01 -2.97565877e-01 8.86377156e-01 7.05485463e-01 2.43311360e-01 -4.14916694e-01 1.94042444e-01 6.99712813e-01 9.26345825e-01 -3.96055639e-01 1.52569425e+00 5.22518992e-01 -3.53080273e-01 -7.87084520e-01 -8.98848712e-01 -4.29267496e-01 -3.80099475e-01 -2.92312596e-02 1.07839799e+00 -1.30703139e+00 -9.91087705e-02 9.86448407e-01 -8.76605392e-01 -3.96762997e-01 -8.87578428e-02 5.92722535e-01 -3.14800620e-01 3.79797101e-01 -2.67461389e-01 -7.66398132e-01 -2.46249601e-01 -1.00861657e+00 6.60580218e-01 5.95061481e-01 5.94551027e-01 -1.28740013e+00 1.26939118e-01 2.86173970e-01 4.69687641e-01 8.77850205e-02 8.98904920e-01 -6.07979298e-01 -4.97771502e-01 9.28064659e-02 -4.06076789e-01 1.06859195e+00 3.49123597e-01 -3.24527949e-01 -8.99276495e-01 -4.00277108e-01 3.48065287e-01 -3.73878717e-01 1.00116992e+00 1.42371625e-01 1.22757161e+00 -3.25154454e-01 -1.70056403e-01 9.94072735e-01 1.43804312e+00 2.28174403e-01 9.34348762e-01 5.22403479e-01 7.57456481e-01 5.83204985e-01 6.87063932e-01 3.29095066e-01 2.63433754e-01 3.02088588e-01 7.24277675e-01 -3.40666801e-01 -2.15645075e-01 -3.95209908e-01 1.90846816e-01 6.18911266e-01 3.04672390e-01 -2.59400189e-01 -7.29224324e-01 8.13400984e-01 -1.63520908e+00 -6.23733222e-01 3.60420853e-01 2.22622919e+00 9.85972166e-01 1.40821978e-01 2.33090017e-02 -6.95030540e-02 8.07207048e-01 3.52763116e-01 -4.95781213e-01 1.58831954e-01 -1.42060399e-01 1.15446366e-01 5.40438533e-01 3.87515873e-01 -1.45204401e+00 1.04563951e+00 5.30068588e+00 1.14735675e+00 -1.16057587e+00 1.25851870e-01 6.37091160e-01 9.57656652e-02 -2.29070783e-01 -1.62044078e-01 -4.11393464e-01 7.40389466e-01 7.68704653e-01 -2.79428005e-01 3.55935693e-01 7.37458289e-01 5.04282676e-02 2.97373049e-02 -5.94708979e-01 1.06330931e+00 1.77847564e-01 -1.17515779e+00 2.93636590e-01 -8.82018432e-02 1.16349506e+00 -2.64913350e-01 5.33402205e-01 3.69071573e-01 3.51099432e-01 -9.45661902e-01 3.25716943e-01 4.01083022e-01 5.20869970e-01 -1.03327620e+00 8.32167089e-01 5.11397779e-01 -6.33680642e-01 2.62033530e-02 -5.94006538e-01 2.27490723e-01 -2.44097501e-01 6.41659975e-01 -8.69076312e-01 6.92704976e-01 8.18913698e-01 9.30740535e-01 -5.17036080e-01 1.25053644e+00 -3.78120959e-01 8.25491130e-01 -1.44795412e-02 3.66331607e-01 3.80628943e-01 -3.89446884e-01 5.42285204e-01 9.70091343e-01 2.23322019e-01 1.93594277e-01 1.21431379e-02 9.70955372e-01 -2.34545305e-01 -1.59669429e-01 -6.03418946e-01 1.57848746e-01 5.77832520e-01 8.35553050e-01 -8.63333717e-02 -3.57082963e-01 -3.40914905e-01 1.19617200e+00 1.45208672e-01 5.83987117e-01 -9.81669068e-01 -8.47857654e-01 6.35411322e-01 -4.55502160e-02 3.75464588e-01 -7.76837096e-02 -9.95848104e-02 -1.15000653e+00 -1.51458472e-01 -7.60027349e-01 2.24562630e-01 -6.91651344e-01 -1.45030272e+00 2.87849188e-01 9.88059118e-02 -1.52759564e+00 1.19387351e-01 -3.07620764e-01 -8.54670465e-01 1.05680990e+00 -1.98361623e+00 -9.18963552e-01 -7.54223645e-01 8.83903265e-01 5.34132659e-01 -3.85567904e-01 2.77878344e-01 5.75703502e-01 -6.67326152e-01 9.10840273e-01 3.54936868e-01 1.86348364e-01 1.08139813e+00 -1.12059271e+00 1.34849623e-01 8.73715162e-01 -2.30016187e-01 2.80000418e-01 8.92148256e-01 -4.02585179e-01 -7.15718865e-01 -1.50738716e+00 8.14619288e-02 -2.51667909e-02 5.30568302e-01 -2.16802835e-01 -1.44650948e+00 5.80849230e-01 4.36383843e-01 9.06349495e-02 4.63574827e-01 -5.60367644e-01 -3.60478997e-01 -3.43438298e-01 -1.36217546e+00 6.14343345e-01 5.72220445e-01 -4.49506283e-01 -6.79506242e-01 3.62767756e-01 1.02649033e+00 -3.82753253e-01 -7.37275600e-01 5.55818260e-01 1.77769408e-01 -9.66485620e-01 9.45383787e-01 -2.29020268e-01 4.62516785e-01 -4.91662443e-01 1.57105595e-01 -1.77261102e+00 -9.55505967e-02 -6.12627685e-01 -5.61523139e-02 1.23704958e+00 4.78057675e-02 -7.62459815e-01 9.91132021e-01 1.07225381e-01 -1.45894766e-01 -4.86849576e-01 -5.49075186e-01 -9.92936254e-01 4.53713000e-01 7.74156302e-02 6.87444329e-01 1.16195190e+00 -5.75225532e-01 1.77677915e-01 -6.48777425e-01 4.07649368e-01 9.48724568e-01 -3.64144772e-01 9.42390919e-01 -8.79549325e-01 -2.24103078e-01 -1.40882298e-01 -6.23328686e-01 -1.05526543e+00 1.81771636e-01 -4.17250484e-01 1.49119705e-01 -9.76625562e-01 -6.36510104e-02 -3.13874573e-01 -4.93113995e-01 3.09225302e-02 -4.82405752e-01 1.81558147e-01 -1.17996326e-02 2.88909167e-01 -3.51181418e-01 1.17874455e+00 1.42154288e+00 -4.94088322e-01 -2.20584378e-01 7.85415899e-03 -4.66636956e-01 6.22543156e-01 8.55244517e-01 -4.41324115e-01 -9.07363713e-01 -7.12194920e-01 -4.53261942e-01 -2.54649132e-01 3.69239926e-01 -1.49592793e+00 4.38622534e-01 -3.03089797e-01 3.74036431e-01 -3.92194659e-01 2.68349707e-01 -7.94744790e-01 -3.67560714e-01 2.39424765e-01 -1.72095090e-01 -4.27631766e-01 8.78977701e-02 9.66265142e-01 -5.75977862e-01 -1.61240816e-01 1.31836188e+00 2.80609168e-03 -8.55007112e-01 5.22468805e-01 -5.77577129e-02 3.35400924e-02 1.33928931e+00 -2.62457430e-01 -3.78622830e-01 -4.97163355e-01 -5.76558352e-01 3.64552557e-01 4.89899158e-01 2.29275882e-01 9.13523018e-01 -1.21652818e+00 -8.68536353e-01 4.05973643e-01 2.06395954e-01 4.59277958e-01 4.56732273e-01 6.04580522e-01 -7.21538782e-01 -9.79348421e-02 -3.67151052e-01 -4.32736456e-01 -9.92054522e-01 6.34284258e-01 4.72869098e-01 -7.56313577e-02 -4.02675718e-01 1.08976364e+00 7.72857487e-01 -4.98534441e-01 1.64702907e-01 2.29053110e-01 -6.27750009e-02 -2.94920325e-01 6.43193245e-01 3.22555691e-01 -3.55704464e-02 -3.73787731e-01 -5.62576279e-02 2.91491419e-01 -4.83331323e-01 -1.28884882e-01 1.08947551e+00 -1.70718521e-01 -1.82798151e-02 6.77424967e-02 1.33480632e+00 -3.31081480e-01 -1.64446771e+00 -5.88315904e-01 -3.03977638e-01 -7.18442202e-01 2.91955203e-01 -6.41030312e-01 -1.09682751e+00 1.15535188e+00 1.01632226e+00 -6.50191978e-02 1.39684975e+00 -4.39852148e-01 1.03774786e+00 3.22961569e-01 6.42063990e-02 -1.03255880e+00 5.00598252e-01 2.56513029e-01 5.19366801e-01 -1.32901287e+00 -1.74833070e-02 -2.49791205e-01 -5.30287087e-01 6.57594979e-01 1.12245476e+00 -5.51968694e-01 9.03212070e-01 -3.58321637e-01 1.99676052e-01 1.56924203e-01 -4.60506290e-01 5.73393404e-02 1.26915917e-01 9.13733184e-01 2.01904085e-02 -1.36038229e-01 -2.94048842e-02 7.69849345e-02 -5.89382015e-02 -1.33085385e-01 4.70597714e-01 8.92684639e-01 -7.39201128e-01 -9.66504991e-01 -3.90122324e-01 1.45855248e-01 -1.36394501e-01 -6.43714964e-02 -8.14474523e-02 8.99647474e-01 2.74679720e-01 8.32880437e-01 -1.11096194e-02 -6.16885066e-01 1.14663407e-01 -2.22502261e-01 4.67280954e-01 -6.92778111e-01 -1.77968860e-01 -1.97595447e-01 -5.34105599e-01 3.42831276e-02 -3.46307099e-01 -3.05092156e-01 -9.62693512e-01 -2.46946692e-01 -3.39790612e-01 4.16374862e-01 3.44385415e-01 9.94954646e-01 1.70911208e-01 6.18147373e-01 8.94698560e-01 -7.38778591e-01 -6.38221145e-01 -1.10231173e+00 -8.71354640e-01 4.16229635e-01 6.96085036e-01 -7.79029369e-01 -6.27195120e-01 1.24002494e-01]	[10.934354782104492, -3.080382823944092]
809922a6-bf61-429a-8477-61f742f8cd70	connective-cognition-network-for-directional	null	null	http://papers.nips.cc/paper/8804-connective-cognition-network-for-directional-visual-commonsense-reasoning	http://papers.nips.cc/paper/8804-connective-cognition-network-for-directional-visual-commonsense-reasoning.pdf	Connective Cognition Network for Directional Visual Commonsense Reasoning	Visual commonsense reasoning (VCR) has been introduced to boost research of cognition-level visual understanding, i.e., a thorough understanding of correlated details of the scene plus an inference with related commonsense knowledge. Recent studies on neuroscience have suggested that brain function or cognition can be described as a global and dynamic integration of local neuronal connectivity, which is context-sensitive to specific cognition tasks. Inspired by this idea, towards VCR, we propose a connective cognition network (CCN) to dynamically reorganize the visual neuron connectivity that is contextualized by the meaning of questions and answers. Concretely, we first develop visual neuron connectivity to fully model correlations of visual content. Then, a contextualization process is introduced to fuse the sentence representation with that of visual neurons. Finally, based on the output of contextualized connectivity, we propose directional connectivity to infer answers or rationales. Experimental results on the VCR dataset demonstrate the effectiveness of our method. Particularly, in $Q \to AR$ mode, our method is around 4\% higher than the state-of-the-art method.	['Yahong Han', 'Linchao Zhu', 'Aming Wu', 'Yi Yang']	2019-12-01	null	null	null	neurips-2019-12	['visual-commonsense-reasoning']	['reasoning']	[ 7.54970163e-02 4.33334373e-02 2.41117895e-01 -4.25319850e-01 2.27799132e-01 -4.56017405e-01 6.84906423e-01 1.94500640e-01 -1.30138263e-01 4.05987918e-01 5.89284778e-01 -3.57303590e-01 -9.54045132e-02 -9.31643486e-01 -5.20068228e-01 -3.09690475e-01 4.28048939e-01 -2.39995122e-01 1.95086703e-01 -4.45963621e-01 6.99511766e-01 8.33392665e-02 -1.34608984e+00 7.77246952e-01 1.01820910e+00 1.02581561e+00 4.80001003e-01 3.88833046e-01 -4.49672431e-01 1.39106953e+00 -5.66664815e-01 -4.14443642e-01 -4.10192698e-01 -9.39510763e-01 -1.07475531e+00 -1.70409679e-02 1.69534773e-01 1.73858355e-03 -2.86783248e-01 1.49238455e+00 1.14222847e-01 2.95204878e-01 5.45400798e-01 -1.17614174e+00 -1.76746607e+00 7.71183491e-01 -4.80483115e-01 8.00845861e-01 3.63644183e-01 4.43636119e-01 1.01926410e+00 -9.04067338e-01 7.20566213e-01 1.54598451e+00 1.67303812e-02 4.52174604e-01 -1.24874377e+00 -5.55507123e-01 6.83521330e-01 8.62874627e-01 -1.18652833e+00 -5.36888745e-03 1.16945767e+00 -4.63368177e-01 9.14373994e-01 1.37420967e-01 1.33275747e+00 9.40893471e-01 2.49826342e-01 6.63899660e-01 1.40931571e+00 -2.28488877e-01 4.02972519e-01 -1.25881419e-01 3.87138188e-01 7.75913119e-01 2.03487396e-01 -2.00591534e-01 -7.23914921e-01 3.28772813e-01 6.73331201e-01 2.33213067e-01 -4.04860914e-01 3.48954126e-02 -1.37738228e+00 7.57906914e-01 1.19804323e+00 4.88368481e-01 -5.22383630e-01 4.87391323e-01 2.01497301e-01 5.61885647e-02 1.06257796e-01 3.42950672e-01 1.41892016e-01 3.21478397e-01 -5.96175849e-01 -7.39028230e-02 3.57117802e-01 7.13354290e-01 7.35557973e-01 1.17135122e-01 -5.20112276e-01 7.08725154e-01 6.72601044e-01 3.92839283e-01 3.25311780e-01 -1.04538131e+00 2.65764862e-01 1.09857976e+00 -5.22309899e-01 -1.45948732e+00 -2.51509279e-01 -3.98534864e-01 -8.85334074e-01 8.82012546e-02 -2.59711388e-02 2.32762545e-01 -7.55572379e-01 1.94293249e+00 9.44791287e-02 1.04105189e-01 6.73118457e-02 1.13969254e+00 1.27722847e+00 5.23420095e-01 3.34962249e-01 -1.00554548e-01 1.70918155e+00 -7.15753913e-01 -1.01656008e+00 -4.85891372e-01 -1.69395227e-02 -1.61198780e-01 1.41452694e+00 1.14735320e-01 -9.25090015e-01 -5.99381030e-01 -1.07393348e+00 -4.55642968e-01 -6.43939197e-01 -1.67766452e-01 6.91603184e-01 2.27918610e-01 -1.21991026e+00 3.94790247e-02 -2.76600599e-01 -4.42189962e-01 9.42722857e-01 -2.97953039e-01 -5.45218261e-03 -1.41358152e-01 -1.25294101e+00 9.28337038e-01 4.97050494e-01 3.73413354e-01 -7.79258549e-01 -6.98177874e-01 -6.84255123e-01 3.37348908e-01 4.40233499e-01 -1.12376845e+00 8.29167545e-01 -7.86009252e-01 -1.10158098e+00 9.61344242e-01 -3.25697184e-01 -1.99848026e-01 -7.72139058e-02 4.20580506e-02 -3.99984181e-01 6.44449353e-01 4.60462458e-02 8.59894216e-01 4.53680009e-01 -1.62823105e+00 -1.66117147e-01 -5.34025788e-01 4.97532010e-01 2.70369589e-01 -1.97083652e-01 -1.74956590e-01 -4.81967479e-01 -5.13734341e-01 3.98513556e-01 -4.24914211e-01 6.03980990e-03 3.07997912e-01 -4.64456648e-01 -3.03311616e-01 5.71032405e-01 -6.77377582e-01 1.23638046e+00 -2.02529383e+00 3.17448705e-01 1.93334490e-01 6.94704294e-01 -2.23599151e-01 -1.83171593e-02 3.16113740e-01 -2.85794050e-01 3.87903750e-01 -2.83373475e-01 8.02815035e-02 8.91408399e-02 1.72894120e-01 -6.50977552e-01 1.87009901e-01 1.99826434e-01 1.43459821e+00 -1.15587032e+00 -5.72735786e-01 6.75971434e-02 4.57262516e-01 -6.30720198e-01 -1.18481569e-01 -3.80512416e-01 3.43141645e-01 -2.40202233e-01 6.74646139e-01 5.21075070e-01 -5.21598637e-01 3.15454125e-01 -5.10047853e-01 8.47428665e-02 1.35801956e-01 -6.54315472e-01 1.73323321e+00 -2.66821057e-01 9.42183971e-01 -2.79677570e-01 -1.11518610e+00 1.06728470e+00 -1.29794955e-01 -3.83595347e-01 -1.05033326e+00 3.18774045e-01 -3.67704660e-01 3.06880653e-01 -6.67633712e-01 3.18005905e-02 -2.06296623e-01 1.37785912e-01 3.58263940e-01 -3.25818621e-02 -2.23359764e-01 6.95126057e-02 6.45909250e-01 7.31001258e-01 5.09634567e-03 3.14913511e-01 -4.05133367e-01 6.02539361e-01 -8.07233825e-02 3.10992897e-01 4.12864864e-01 -4.07190263e-01 2.02106372e-01 8.52352083e-01 -1.87967315e-01 -3.89408916e-01 -1.20214295e+00 2.26405531e-01 9.64630008e-01 6.49439096e-01 -3.63455862e-01 -9.05454636e-01 -2.27615625e-01 -3.16584736e-01 1.19440281e+00 -1.17095661e+00 -5.12989104e-01 -2.57410318e-01 -5.28002143e-01 2.09680915e-01 7.61932671e-01 1.06978667e+00 -1.59877789e+00 -7.77881384e-01 -1.95946798e-01 -4.34362650e-01 -1.11728823e+00 -2.17689425e-01 -1.99262887e-01 -7.63189316e-01 -1.28518224e+00 -1.05452903e-01 -8.23370814e-01 8.10514688e-01 5.48020244e-01 1.19470346e+00 5.69612563e-01 -3.75908166e-01 5.52409232e-01 -3.17540318e-01 -8.97443146e-02 1.72178298e-02 -7.27924764e-01 -5.07923067e-01 7.98868164e-02 4.54504550e-01 -8.20404708e-01 -9.73202646e-01 -2.43447006e-01 -1.07570517e+00 3.13327193e-01 3.76904279e-01 6.46457434e-01 6.36025548e-01 -1.46774143e-01 7.27309704e-01 -6.98369384e-01 1.14507484e+00 -6.21459067e-01 -2.67508253e-02 5.48393428e-01 -4.62448537e-01 1.18190527e-01 4.02947903e-01 -3.02993089e-01 -1.42090201e+00 -6.16074800e-01 2.46094882e-01 -4.48433936e-01 -1.23427629e-01 8.00736785e-01 -3.36231023e-01 2.52748907e-01 5.54424345e-01 5.80197275e-01 -2.27018028e-01 4.93595935e-02 9.62495089e-01 2.05035031e-01 7.79676139e-01 -6.51668549e-01 4.48340118e-01 8.36832285e-01 -1.28034666e-01 -5.32598376e-01 -9.62634146e-01 8.56873579e-03 -5.94904304e-01 -6.07222736e-01 1.37896252e+00 -5.87724209e-01 -1.26992917e+00 4.49462757e-02 -1.41988134e+00 -1.35089457e-01 -2.39788443e-01 1.78339690e-01 -5.58282554e-01 4.11162108e-01 -4.92025405e-01 -6.37204170e-01 -3.82642657e-01 -9.02044594e-01 4.68514949e-01 4.64413047e-01 -1.61856264e-01 -1.20943677e+00 -1.58033296e-01 5.13941348e-01 3.34836274e-01 3.02637011e-01 1.42433095e+00 -2.34361380e-01 -6.72718108e-01 3.78294766e-01 -8.74341488e-01 4.90978323e-02 -1.15833737e-01 -6.60562515e-02 -9.31791186e-01 3.94578785e-01 2.13283241e-01 -3.06208700e-01 1.23090959e+00 2.13817179e-01 1.61528695e+00 -1.83471113e-01 -2.13973314e-01 3.60080063e-01 1.52442515e+00 2.67873555e-01 9.42218840e-01 1.72963232e-01 5.75084507e-01 6.97755337e-01 1.92831710e-01 2.61469156e-01 9.82285321e-01 7.27702230e-02 8.49919438e-01 8.75525177e-02 -5.37459910e-01 -4.13555682e-01 1.05769701e-01 7.72914052e-01 -2.78884560e-01 1.41304612e-01 -9.66288388e-01 5.20507634e-01 -1.86804318e+00 -1.45449102e+00 -2.40124047e-01 1.40458262e+00 8.31719637e-01 5.05458083e-05 -5.40576637e-01 -1.64863378e-01 7.95645654e-01 2.35342115e-01 -7.02802658e-01 -4.90270168e-01 -1.91019580e-01 6.30757511e-02 -5.07076502e-01 3.39394480e-01 -4.25103098e-01 1.19608414e+00 6.16374683e+00 5.95219493e-01 -8.85654211e-01 1.75131172e-01 4.43589896e-01 5.95656000e-02 -8.69350433e-01 1.65329352e-01 -1.88267589e-01 2.69536257e-01 2.61057317e-01 -1.45175233e-01 8.60885203e-01 5.19943058e-01 1.86887205e-01 -3.25556129e-01 -9.03325260e-01 1.10937202e+00 5.43454587e-01 -1.64440095e+00 4.65762287e-01 -1.01732492e-01 5.78598976e-01 -5.14829755e-01 1.83711126e-01 3.85518581e-01 3.10214043e-01 -1.06868970e+00 9.66549575e-01 9.89252448e-01 5.08336246e-01 -6.24090850e-01 3.18716735e-01 3.83884251e-01 -1.34770548e+00 -1.56913385e-01 -4.42757338e-01 -2.36291617e-01 -3.98328789e-02 4.34076697e-01 -1.90669388e-01 2.43009984e-01 8.57742667e-01 7.50219584e-01 -9.15283084e-01 7.18267381e-01 -9.07478511e-01 3.65884781e-01 4.24383074e-01 -4.59843725e-01 -1.07283682e-01 -1.04810178e-01 2.37112924e-01 1.13289189e+00 -5.50271347e-02 7.50214040e-01 -1.43853262e-01 1.80521953e+00 -1.09753065e-01 -1.08105049e-01 -4.05027181e-01 4.28129472e-02 6.56521797e-01 1.34959185e+00 -1.05558050e+00 -5.25127590e-01 -2.19556779e-01 9.16391850e-01 7.20558107e-01 6.69620752e-01 -9.85288560e-01 -3.51210535e-01 5.00351727e-01 -2.98290521e-01 3.20732623e-01 -1.81329161e-01 -6.25927925e-01 -1.30030525e+00 -6.63650930e-02 -4.34429437e-01 2.12762296e-01 -1.61206925e+00 -1.36670661e+00 6.09723806e-01 -4.74007986e-02 -7.74376094e-01 2.95689136e-01 -7.17666805e-01 -9.45991397e-01 8.53891730e-01 -1.54691446e+00 -1.09854305e+00 -6.22165978e-01 8.90804887e-01 4.30535942e-01 2.86067724e-01 7.05365002e-01 -3.40759128e-01 -4.51163739e-01 2.90315822e-02 -8.31617475e-01 1.46311671e-01 2.09071487e-01 -1.18464744e+00 1.09939985e-02 9.17956173e-01 1.89059049e-01 1.13309181e+00 4.45980668e-01 -7.69320488e-01 -1.22437787e+00 -7.65292883e-01 6.60980105e-01 -4.91783917e-01 8.14034641e-01 -3.57630491e-01 -1.06287181e+00 4.82281685e-01 8.14653754e-01 4.11347523e-02 1.07168961e+00 1.08153455e-01 -8.57140899e-01 -5.67594031e-03 -1.02567506e+00 1.02689469e+00 1.38892043e+00 -8.98406327e-01 -1.36026406e+00 4.14580517e-02 1.02514672e+00 1.35861784e-01 -5.38897812e-01 -1.27759770e-01 2.13533074e-01 -1.06026864e+00 1.00076175e+00 -6.29174531e-01 8.04677129e-01 -6.13718987e-01 -4.54787582e-01 -1.35363603e+00 -6.79475665e-01 1.02736481e-01 -2.27011845e-01 1.15266097e+00 1.52767301e-01 -3.99072766e-01 6.92183748e-02 5.30511200e-01 -1.75308332e-01 -6.05610967e-01 -6.97758853e-01 -2.48529404e-01 -5.65400384e-02 -6.05213642e-01 4.83165652e-01 1.14353347e+00 4.75751728e-01 6.86952829e-01 2.65806526e-01 7.84461796e-02 5.45624971e-01 4.56195325e-01 6.81933686e-02 -1.15569293e+00 -1.13619067e-01 -6.93141997e-01 -2.06365615e-01 -8.90386760e-01 2.92242080e-01 -1.27873147e+00 -1.78366855e-01 -2.25357199e+00 7.92418182e-01 3.87430996e-01 -4.46847022e-01 6.02118254e-01 -4.75012273e-01 1.94538549e-01 5.96780479e-01 1.98723879e-02 -7.19903767e-01 5.44898272e-01 1.83464897e+00 -2.27465540e-01 9.11992136e-03 -9.63900864e-01 -1.34817469e+00 9.22656059e-01 7.12244093e-01 6.64050924e-03 -6.15005970e-01 -4.50455248e-01 6.97510839e-01 -3.60507108e-02 1.14847994e+00 -7.43474007e-01 3.91210914e-01 -3.99345487e-01 6.60618961e-01 -6.15491927e-01 2.26492748e-01 -6.44143283e-01 -3.53445292e-01 5.65797329e-01 -5.60045004e-01 4.28254642e-02 3.30088176e-02 7.03679800e-01 -2.41503045e-01 1.95246730e-02 6.11201942e-01 -4.20478910e-01 -1.06220555e+00 -3.96085307e-02 -3.88017178e-01 3.75882715e-01 8.67978334e-01 -2.85910577e-01 -9.57295716e-01 -2.89654523e-01 -7.99223840e-01 2.04783157e-01 -8.52558166e-02 4.04414415e-01 1.18999171e+00 -1.42617047e+00 -3.53259891e-01 -2.54738629e-01 2.33041599e-01 -2.41658136e-01 5.90526044e-01 7.70841599e-01 -3.48828375e-01 3.19132686e-01 -3.27920914e-01 -4.59585190e-01 -8.38517308e-01 9.63668883e-01 2.23780826e-01 2.77247667e-01 -6.23550475e-01 9.14420843e-01 5.21362305e-01 -6.53283149e-02 -1.79800645e-01 -7.84021020e-01 -8.13671112e-01 2.34099746e-01 7.43545651e-01 8.51533562e-02 -5.44826686e-01 -4.89858240e-01 -5.44996381e-01 6.04642987e-01 2.94815958e-01 -2.25088105e-01 1.12483501e+00 -2.51971602e-01 -6.27520978e-01 5.69834411e-01 8.00384223e-01 -1.35733142e-01 -1.14848542e+00 -1.36422589e-01 -3.06799263e-01 -2.85478026e-01 -2.41454467e-02 -1.11202168e+00 -1.20209920e+00 1.41986549e+00 2.28318274e-01 2.54414231e-01 1.31578279e+00 4.29893732e-01 2.01118723e-01 3.68259311e-01 6.09659702e-02 -9.10879612e-01 6.79400682e-01 4.95210499e-01 1.41076910e+00 -9.48724329e-01 -9.55955237e-02 -4.24652725e-01 -8.80903721e-01 1.20881116e+00 8.94129336e-01 -3.15785289e-01 7.01447785e-01 -2.72653282e-01 -1.79229900e-01 -6.39484167e-01 -9.47955787e-01 -4.18879747e-01 4.14632201e-01 6.63528502e-01 2.96578586e-01 5.44645190e-02 -3.04158837e-01 1.12449992e+00 -1.89499840e-01 -5.63858375e-02 3.95477563e-01 5.74649811e-01 -7.60612369e-01 -2.01737061e-01 -3.04483801e-01 4.01033238e-02 1.47880927e-01 -5.66120863e-01 -6.39635444e-01 5.57403028e-01 4.62139308e-01 1.19576573e+00 1.24336742e-01 -4.15953755e-01 2.55732924e-01 1.67016163e-01 5.35677373e-01 -6.05920911e-01 -5.15828311e-01 -2.55425274e-01 -2.96845019e-01 -6.83260381e-01 -6.63730145e-01 -3.43496084e-01 -1.94983625e+00 -1.66640893e-01 7.73367211e-02 -2.58939654e-01 4.65614617e-01 1.21180809e+00 2.68132269e-01 1.10384130e+00 5.50546013e-02 -3.22669685e-01 -5.00137694e-02 -7.05516756e-01 -3.57664198e-01 4.91048217e-01 6.49318239e-03 -7.01358557e-01 -2.54399776e-01 4.02001083e-01]	[10.761784553527832, 1.799759864807129]
52d35329-26f0-4d51-ab73-7bbc8bb8cd17	classification-of-12-lead-ecg-signals-with-bi	1811.02090	null	http://arxiv.org/abs/1811.02090v1	http://arxiv.org/pdf/1811.02090v1.pdf	Classification of 12-Lead ECG Signals with Bi-directional LSTM Network	We propose a recurrent neural network classifier to detect pathologies in 12-lead ECG signals and train and validate the classifier with the Chinese physiological signal challenge dataset (http://www.icbeb.org/Challenge.html). The recurrent neural network consists of two bi-directional LSTM layers and can train on arbitrary-length ECG signals. Our best trained model achieved an average F1 score of 74.15% on the validation set. Keywords: ECG classification, Deep learning, RNN, Bi-directional LSTM, QRS detection.	['William Wee', 'Junye Luo', 'Ahmed Mostayed', 'Xingliang Shu']	2018-11-05	null	null	null	null	['ecg-classification']	['medical']	[ 2.53182948e-01 -1.39038056e-01 8.76531973e-02 -3.59075963e-01 -9.90897298e-01 -1.50458470e-01 -6.16768837e-01 -2.17856869e-01 -3.40017378e-01 8.30683053e-01 1.10684916e-01 -5.91619968e-01 -3.59740742e-02 -1.91831976e-01 -2.90403754e-01 -6.02173030e-01 -6.08811855e-01 -1.24363795e-01 -3.76892656e-01 -6.67685494e-02 -1.90530509e-01 5.48237503e-01 -5.24449229e-01 7.22741246e-01 5.61600327e-01 1.31779325e+00 -2.64961571e-01 1.21704376e+00 6.89761817e-01 1.18546271e+00 -1.17960083e+00 -2.04899199e-02 -3.96541685e-01 -1.11536384e+00 -8.81484032e-01 -7.62134135e-01 -4.32744354e-01 2.27174461e-02 -3.94855797e-01 6.15576744e-01 1.49068511e+00 -2.57174134e-01 1.98538408e-01 -5.86981654e-01 -8.88027400e-02 1.07067144e+00 -3.61290038e-01 1.00633717e+00 3.10508788e-01 -2.07259014e-01 4.56912994e-01 -6.53214335e-01 4.85931635e-01 1.94759741e-01 1.58605468e+00 5.16441643e-01 -1.07732582e+00 -7.30425239e-01 -6.45259559e-01 3.52931648e-01 -1.38238835e+00 -2.88559496e-01 1.05602181e+00 -1.91028938e-01 1.19867551e+00 3.72385263e-01 8.18947911e-01 1.57458091e+00 6.98005617e-01 7.89493024e-01 8.85578036e-01 -2.76840180e-01 -2.46180743e-01 -3.63056928e-01 4.13287222e-01 3.27113688e-01 -2.13110715e-01 1.69764496e-02 -4.07007366e-01 -3.28742027e-01 7.74016082e-01 7.15088611e-03 -3.34236562e-01 6.04127347e-01 -1.22576451e+00 4.24949020e-01 3.50031316e-01 7.95361698e-01 -9.66034055e-01 1.78353339e-01 1.23526561e+00 7.09833801e-01 4.95161474e-01 6.87377810e-01 -8.12224507e-01 -6.27371132e-01 -9.09236252e-01 -2.39527211e-01 5.86073041e-01 3.45177323e-01 -2.02001229e-01 6.19078755e-01 -3.37230295e-01 1.16736114e+00 -1.86559707e-01 4.79907930e-01 1.01702845e+00 -7.37686157e-01 1.84322938e-01 -1.24677261e-02 -1.85596198e-01 -9.54252064e-01 -1.09024715e+00 -1.07399690e+00 -1.37911534e+00 -5.10821998e-01 -1.70475557e-01 -9.51227844e-01 -7.20078349e-01 1.38177276e+00 -4.44633275e-01 1.95948407e-01 2.75521815e-01 8.00746799e-01 1.34265649e+00 4.25227910e-01 1.73996344e-01 -5.45828044e-01 9.71397221e-01 -3.47838104e-01 -1.03883946e+00 6.62508458e-02 9.10990596e-01 -2.14013293e-01 4.35519993e-01 6.89119756e-01 -1.23989475e+00 -7.72892892e-01 -9.86348033e-01 3.29368770e-01 1.70776576e-01 7.61516571e-01 1.23856135e-01 7.08800375e-01 -9.57866371e-01 8.84973407e-01 -9.50634360e-01 -1.22848496e-01 3.77386302e-01 4.64858145e-01 -2.38855314e-02 5.10935664e-01 -1.92214370e+00 7.78905749e-01 3.49581987e-01 8.75984669e-01 -7.50121653e-01 -5.13547540e-01 -5.87545216e-01 -1.70415640e-01 -4.95197147e-01 -4.47447032e-01 1.23154247e+00 -8.09985518e-01 -1.38843071e+00 9.84234273e-01 -1.06723480e-01 -9.43558574e-01 3.90857548e-01 -3.49736452e-01 -1.07022679e+00 3.77167352e-02 -2.17993975e-01 1.48781938e-02 5.83397508e-01 -4.08346325e-01 -1.99312106e-01 -2.89582282e-01 -5.85620999e-01 -1.66891262e-01 1.77924082e-01 3.53820384e-01 3.85481298e-01 -8.20058942e-01 1.55635551e-01 -7.51634240e-01 -2.32444212e-01 -1.01144373e+00 -5.80107510e-01 -7.85507485e-02 5.59699237e-01 -1.21207833e+00 1.61846542e+00 -2.46090102e+00 -2.10310724e-02 2.51913071e-01 4.50330555e-01 3.93990546e-01 1.69143695e-02 1.20271839e-01 -6.70906842e-01 2.53793985e-01 -1.56469181e-01 1.29850596e-01 -4.73093510e-01 6.33182079e-02 -2.79209703e-01 5.11622012e-01 -9.86093469e-03 1.34795570e+00 -5.76349676e-01 -8.27164128e-02 -5.17724790e-02 7.75680661e-01 3.40810061e-01 1.71025276e-01 6.44205213e-01 4.85554457e-01 -9.41054448e-02 3.43464166e-01 2.90545851e-01 -1.93092778e-01 4.35618013e-01 -4.92918551e-01 2.28253096e-01 6.29618227e-01 -6.18939459e-01 1.89898908e+00 -2.42325202e-01 9.81769025e-01 -6.37749970e-01 -1.23904252e+00 1.31218851e+00 1.05539215e+00 6.31730199e-01 -7.13746548e-01 4.64581817e-01 4.07622963e-01 5.73038518e-01 -7.61419654e-01 -2.45866567e-01 -1.61317021e-01 -2.26504907e-01 2.99528569e-01 -5.36486171e-02 6.10446334e-01 -2.75149316e-01 -2.99373478e-01 1.45463395e+00 -7.57831186e-02 1.17737353e-01 -1.23063527e-01 3.56236309e-01 -5.65746546e-01 9.27805722e-01 1.02522433e+00 -3.65611136e-01 7.73400307e-01 6.06023490e-01 -9.97330606e-01 -4.21551973e-01 -8.27852607e-01 -2.76704490e-01 8.88965726e-01 -6.19593978e-01 -2.36627951e-01 -5.49756587e-01 -4.74467486e-01 -3.82631779e-01 3.80977243e-01 -7.63904035e-01 -5.41908741e-01 -8.75524104e-01 -9.71483111e-01 1.71220911e+00 1.03466880e+00 2.85136670e-01 -1.88263428e+00 -1.18404996e+00 7.52146423e-01 -5.32324016e-01 -1.00140440e+00 1.37166297e-02 7.27106094e-01 -1.23944628e+00 -8.34475636e-01 -9.44676936e-01 -8.80688250e-01 -1.13128595e-01 -8.43148351e-01 1.35755324e+00 -3.19399945e-02 -5.27103305e-01 -9.27657913e-03 -3.56893033e-01 -6.44749522e-01 -1.64653048e-01 3.84018660e-01 -5.49117401e-02 -7.23017007e-02 4.38538134e-01 -6.43755972e-01 -6.21865690e-01 9.36478451e-02 -1.35534212e-01 -3.78061503e-01 3.89099061e-01 1.00066411e+00 5.59087038e-01 -5.53793371e-01 1.49018538e+00 -8.62600446e-01 1.19605613e+00 -1.65839165e-01 2.11422220e-01 9.05506611e-02 -3.86266708e-01 -5.65810621e-01 5.18504083e-01 -3.80056232e-01 -4.65574533e-01 2.29080375e-02 -7.20263481e-01 -3.62296909e-01 -1.21440746e-01 8.63604426e-01 4.12125766e-01 2.21118048e-01 8.44587028e-01 3.26375157e-01 -2.47342736e-01 -2.61169195e-01 -3.03633630e-01 6.34178221e-01 7.38409698e-01 -1.55780524e-01 -2.27121621e-01 -2.41410613e-01 -3.06476533e-01 -7.11542010e-01 -6.62781239e-01 -1.44408867e-01 -8.39380920e-01 -2.65326589e-01 9.29813027e-01 -8.15433025e-01 -7.85845220e-01 6.59124494e-01 -1.36131704e+00 -4.29614604e-01 -3.58539730e-01 7.39826679e-01 -4.07334358e-01 -2.96504162e-02 -1.24272346e+00 -8.45363557e-01 -1.38623059e+00 -5.19067645e-01 5.32625198e-01 -2.65806764e-01 -8.44605565e-01 -8.18442941e-01 2.57723361e-01 -4.31119591e-01 5.53729117e-01 9.14555550e-01 3.88190567e-01 -9.41395462e-01 6.39136612e-01 -3.78954172e-01 2.57630467e-01 5.72019041e-01 1.20562993e-01 -1.87750816e-01 -1.11224902e+00 -2.09498525e-01 3.51649135e-01 -3.29641759e-01 7.42772222e-01 9.61058736e-01 1.43550372e+00 1.59054309e-01 -2.09002525e-01 7.81205475e-01 7.87330329e-01 7.39392698e-01 1.09657800e+00 -1.42008722e-01 5.39942145e-01 -1.37898996e-01 -9.93873086e-03 2.74844736e-01 9.77248535e-04 -1.29749954e-01 -3.05804938e-01 -6.49712861e-01 2.65263349e-01 3.10987473e-01 1.89796254e-01 1.40074289e+00 -4.85154420e-01 3.12613279e-01 -1.36259925e+00 4.41362709e-01 -1.62507224e+00 -1.03715980e+00 -5.12790740e-01 1.84584713e+00 8.35239828e-01 2.86366045e-01 3.12825412e-01 8.36601257e-01 6.36382222e-01 -8.67917612e-02 -7.02542484e-01 -9.18408930e-01 -4.18855131e-01 6.05276704e-01 2.68544834e-02 -3.90215255e-02 -1.06829798e+00 3.64313453e-01 6.79215670e+00 2.38263816e-01 -1.85298824e+00 4.18373555e-01 9.39196169e-01 -1.12245560e-01 4.75071102e-01 -9.24150169e-01 -1.98331669e-01 3.16567957e-01 1.87226927e+00 -1.42452627e-01 -2.53669500e-01 3.18340510e-01 4.15364057e-01 5.55280387e-01 -7.96751559e-01 1.37746131e+00 7.45352134e-02 -1.08599246e+00 -7.07595766e-01 -4.51741874e-01 3.58957380e-01 7.35973656e-01 -1.02337979e-01 5.20598114e-01 -5.93129456e-01 -1.33896279e+00 3.84627014e-01 9.15063024e-01 1.47053063e+00 -8.45877349e-01 1.21394980e+00 1.42668620e-01 -1.19534600e+00 -2.22306713e-01 -5.95658086e-02 -3.13318558e-02 -4.84512970e-02 6.10699356e-01 -6.15291774e-01 4.80607569e-01 1.07505274e+00 1.29623067e+00 -4.94903415e-01 1.07840264e+00 -1.48941487e-01 1.37046850e+00 -1.75938740e-01 6.36348054e-02 -1.56314567e-01 3.56725425e-01 3.78323764e-01 1.73366332e+00 1.05963416e-01 4.14033085e-01 -3.07896771e-02 5.88310242e-01 -1.84092492e-01 -1.73191071e-01 -4.86367732e-01 1.73770890e-01 2.05143079e-01 9.49335039e-01 -4.69869614e-01 -4.77540761e-01 1.81137487e-01 6.91667497e-01 -5.93638383e-02 7.46379733e-01 -1.05116165e+00 -1.20399725e+00 -3.94526497e-02 -3.78487974e-01 -1.37725040e-01 3.90504122e-01 -9.46732402e-01 -8.42994392e-01 1.33250564e-01 -1.13602650e+00 5.87002933e-01 -8.16188395e-01 -1.15386581e+00 1.29779088e+00 -6.59520626e-01 -1.05478764e+00 -5.47743499e-01 -2.77820081e-01 -8.18291128e-01 1.15990770e+00 -9.74230289e-01 -5.63741565e-01 -8.83281231e-03 6.72651768e-01 2.20442146e-01 -1.38650155e-02 1.36723387e+00 7.45242953e-01 -7.69123077e-01 8.65576267e-01 -3.62698466e-01 7.19673991e-01 4.32722777e-01 -1.12298155e+00 4.69094783e-01 5.68323731e-01 -1.52688757e-01 6.94249988e-01 3.22671920e-01 -3.05830032e-01 -8.42864513e-01 -1.09725201e+00 1.17979944e+00 -1.76011428e-01 1.97507590e-01 -4.78931144e-02 -1.08298528e+00 6.08075321e-01 1.54971808e-01 1.48487076e-01 8.43050897e-01 3.91694844e-01 1.54477596e-01 -4.46090966e-01 -8.24556470e-01 2.20282581e-02 6.41701162e-01 -9.58254755e-01 -7.35414982e-01 1.12206005e-01 1.20698735e-01 -5.87738097e-01 -1.49548590e+00 1.02832949e+00 1.06321704e+00 -6.13239288e-01 8.37640345e-01 -9.06749845e-01 1.70779571e-01 1.84373885e-01 3.57713014e-01 -1.25125182e+00 -3.78700048e-01 -1.00060904e+00 -2.81459600e-01 5.33195794e-01 9.63524938e-01 -6.92160964e-01 5.40173173e-01 2.30738707e-02 -4.05557692e-01 -1.22638452e+00 -7.84563363e-01 -5.02162755e-01 -1.41197339e-01 -6.42949104e-01 -1.19560193e-02 8.59959483e-01 3.23223442e-01 4.61435318e-01 -7.29030728e-01 -1.44847795e-01 2.12719917e-01 -1.11567371e-01 -2.56625533e-01 -1.24655318e+00 -4.12440859e-02 -2.88688391e-01 -4.18240577e-01 -4.20559555e-01 2.29218341e-02 -9.78724062e-01 -1.86525397e-02 -1.55426240e+00 -1.96125627e-01 -1.42542839e-01 -1.39212978e+00 8.36945236e-01 -1.78230312e-02 6.56880915e-01 -1.97964594e-01 -1.59278750e-01 -5.56033969e-01 -2.17567589e-02 5.64609945e-01 -1.30786136e-01 -6.43995643e-01 3.78808528e-01 -4.02541101e-01 6.18602216e-01 1.40568733e+00 -7.88348734e-01 -1.64026633e-01 -6.19552471e-02 2.56237686e-01 6.85617328e-01 1.37300864e-01 -1.36292171e+00 1.31028056e-01 7.10234523e-01 1.13823199e+00 -7.73342490e-01 2.59300470e-01 -1.62579924e-01 3.94161850e-01 1.08557022e+00 -7.36056983e-01 5.58528244e-01 4.82802361e-01 -9.09059644e-02 -3.23372841e-01 1.34758815e-01 5.20046353e-01 2.38768216e-02 6.86800922e-04 1.17175683e-01 -8.72281969e-01 3.28768969e-01 3.91874611e-01 -1.98488981e-01 2.30619106e-02 -1.53427407e-01 -1.42291915e+00 -5.01360409e-02 -8.66349101e-01 3.40291172e-01 1.05266130e+00 -1.27892876e+00 -9.70653951e-01 2.78584093e-01 -8.35561827e-02 -7.44313776e-01 4.51537430e-01 1.42491996e+00 -5.29203117e-01 3.11372548e-01 -5.15932441e-01 -8.86546969e-01 -1.32894635e+00 -2.76929010e-02 1.26867902e+00 -2.32523575e-01 -1.05595005e+00 9.03875351e-01 -6.68783545e-01 8.14080611e-02 5.07255673e-01 -5.52376390e-01 -5.13739347e-01 1.13541819e-01 5.77471614e-01 4.65541631e-01 5.27211964e-01 -2.81794399e-01 -6.15942836e-01 4.08547550e-01 2.04886571e-01 2.83403963e-01 1.42666686e+00 2.65193284e-01 -2.54656911e-01 1.06293368e+00 1.24721777e+00 -7.18257904e-01 -5.68800569e-02 8.65901336e-02 3.37181568e-01 4.48953420e-01 4.81722392e-02 -1.26512551e+00 -1.39467704e+00 1.20640349e+00 1.19371140e+00 3.01019251e-01 1.45728612e+00 -5.88480413e-01 1.18006241e+00 6.38425052e-01 9.37191956e-03 -9.62261975e-01 -2.55103916e-01 5.61421394e-01 1.01791120e+00 -6.91300571e-01 -3.67040843e-01 5.20959675e-01 -7.86009371e-01 1.42170393e+00 -1.88031923e-02 -5.14719784e-01 1.00532615e+00 3.70466799e-01 6.04219437e-01 -3.70534629e-01 -7.72874653e-01 2.67417908e-01 3.22981834e-01 4.38802242e-01 1.03803003e+00 2.32764035e-01 -4.62048799e-01 1.06799376e+00 -9.36868191e-02 6.82037413e-01 4.73979920e-01 7.21156240e-01 -5.53954300e-03 -6.35839403e-01 2.74335761e-02 9.40460861e-01 -1.25389314e+00 -1.20081879e-01 -3.47150087e-01 2.06877679e-01 -1.09431267e-01 9.38316822e-01 -2.73692459e-01 -7.04776347e-01 5.41020334e-01 5.60278952e-01 3.43523890e-01 -2.88972795e-01 -1.43569374e+00 2.36288950e-01 3.38277370e-01 -4.47388560e-01 -2.43873507e-01 -5.00288963e-01 -1.31160533e+00 3.23864043e-01 -4.71338332e-02 3.05652440e-01 4.56843346e-01 6.63853288e-01 6.01984918e-01 1.32393646e+00 5.22979200e-01 -4.37301904e-01 -3.65493774e-01 -1.56063712e+00 -7.59009838e-01 1.08369276e-01 8.02290797e-01 6.42286167e-02 6.75844913e-03 2.40381405e-01]	[14.341415405273438, 3.3117568492889404]
841eae4f-75e8-40fd-82f7-35e821a86fac	deep-learning-for-automated-medical-image	1903.04711	null	http://arxiv.org/abs/1903.04711v1	http://arxiv.org/pdf/1903.04711v1.pdf	Deep Learning for Automated Medical Image Analysis	Medical imaging is an essential tool in many areas of medical applications, used for both diagnosis and treatment. However, reading medical images and making diagnosis or treatment recommendations require specially trained medical specialists. The current practice of reading medical images is labor-intensive, time-consuming, costly, and error-prone. It would be more desirable to have a computer-aided system that can automatically make diagnosis and treatment recommendations. Recent advances in deep learning enable us to rethink the ways of clinician diagnosis based on medical images. In this thesis, we will introduce 1) mammograms for detecting breast cancers, the most frequently diagnosed solid cancer for U.S. women, 2) lung CT images for detecting lung cancers, the most frequently diagnosed malignant cancer, and 3) head and neck CT images for automated delineation of organs at risk in radiotherapy. First, we will show how to employ the adversarial concept to generate the hard examples improving mammogram mass segmentation. Second, we will demonstrate how to use the weakly labeled data for the mammogram breast cancer diagnosis by efficiently design deep learning for multi-instance learning. Third, the thesis will walk through DeepLung system which combines deep 3D ConvNets and GBM for automated lung nodule detection and classification. Fourth, we will show how to use weakly labeled data to improve existing lung nodule detection system by integrating deep learning with a probabilistic graphic model. Lastly, we will demonstrate the AnatomyNet which is thousands of times faster and more accurate than previous methods on automated anatomy segmentation.	['Wentao Zhu']	2019-03-12	null	null	null	null	['lung-nodule-detection']	['medical']	[ 4.38319266e-01 6.50713861e-01 -1.89690530e-01 -3.36503386e-01 -9.41148162e-01 -3.90743345e-01 2.23843623e-02 -5.93555346e-02 -2.21722350e-01 6.72311008e-01 1.09733166e-02 -1.10910881e+00 -2.21968368e-02 -1.09433448e+00 -5.94013453e-01 -7.79854059e-01 3.55495252e-02 9.30884838e-01 3.34777534e-01 -1.23224622e-02 -2.66644448e-01 6.94559097e-01 -7.96366632e-01 3.68832678e-01 5.65739632e-01 9.64071929e-01 2.61420399e-01 1.05368578e+00 -1.14158066e-02 9.54388678e-01 -2.76037902e-01 -4.20325369e-01 3.07974130e-01 -7.20392704e-01 -1.05706584e+00 1.47132501e-01 3.36449057e-01 -5.73881745e-01 -2.27597371e-01 9.39315081e-01 6.78215802e-01 -4.79577184e-01 9.92465317e-01 -8.43512475e-01 -3.85137141e-01 8.53971481e-01 -6.16252124e-01 2.03523174e-01 -3.05111241e-02 7.34399557e-02 2.96137840e-01 -4.86064672e-01 2.30746686e-01 8.31146121e-01 8.39696765e-01 1.13396716e+00 -4.94147182e-01 -6.22562051e-01 -4.58998621e-01 -1.38952956e-01 -8.94099176e-01 1.59558937e-01 1.01082735e-01 -4.39989001e-01 5.35800278e-01 5.90854645e-01 7.31920958e-01 1.04372501e+00 6.01291776e-01 6.98161066e-01 9.66592550e-01 -3.34243417e-01 8.73703859e-04 1.62525147e-01 1.08145177e-01 1.35218048e+00 3.42644036e-01 2.50929207e-01 4.35855985e-01 -2.05298185e-01 9.84099507e-01 2.57780164e-01 -2.25952610e-01 -1.31083250e-01 -1.15109885e+00 9.57574308e-01 6.70306861e-01 5.75787961e-01 -1.02788836e-01 3.34453344e-01 2.31377408e-01 -1.91915140e-01 1.81005687e-01 2.25734428e-01 -1.38551399e-01 3.14939409e-01 -1.11453450e+00 -9.48518962e-02 8.80096555e-01 6.53236628e-01 -9.93527547e-02 3.58599275e-02 -3.56321037e-01 5.22331238e-01 3.98352146e-01 5.95132947e-01 9.15698230e-01 -7.23810136e-01 7.65453875e-02 3.61133486e-01 -1.81631193e-01 -5.67095101e-01 -7.09350288e-01 -7.06558704e-01 -1.32550848e+00 1.84381440e-01 4.02447253e-01 -3.18681821e-02 -1.61295283e+00 1.17455602e+00 3.02736372e-01 2.49483779e-01 6.48523271e-02 7.14184582e-01 1.18040538e+00 1.78963870e-01 2.06789613e-01 8.14090073e-02 1.61110580e+00 -9.43896890e-01 -4.70754713e-01 -2.77425814e-02 8.70167911e-01 -5.30556917e-01 7.69247591e-01 1.96483567e-01 -1.38888609e+00 -4.18033421e-01 -8.66473377e-01 8.77412129e-03 -1.91250011e-01 2.75707006e-01 7.43077040e-01 9.61626470e-01 -9.41166580e-01 5.76507390e-01 -1.37417662e+00 -3.89485657e-01 9.02179837e-01 7.79035687e-01 -2.30169788e-01 -6.15046844e-02 -1.03797662e+00 1.05108845e+00 4.44509745e-01 -1.48904666e-01 -1.09795356e+00 -9.38235641e-01 -8.39038491e-01 -5.92750981e-02 2.54144162e-01 -1.11965811e+00 1.72236216e+00 -7.98055649e-01 -1.21799994e+00 1.30728579e+00 1.76429629e-01 -8.41179073e-01 8.36792827e-01 2.28178307e-01 -2.15673640e-01 4.34327930e-01 6.80361316e-02 8.94339144e-01 7.51770437e-01 -8.99703681e-01 -8.23446929e-01 -3.05067986e-01 -2.93302983e-01 1.21921770e-01 -3.41869928e-02 -3.81113380e-01 -3.94225210e-01 -6.32356465e-01 2.26139277e-01 -1.02485895e+00 -5.67094386e-01 1.68521345e-01 -7.03160226e-01 -2.73494665e-02 9.88182545e-01 -8.93942118e-01 8.09832573e-01 -1.89482391e+00 -2.80763417e-01 3.10082555e-01 5.50240815e-01 2.68985569e-01 3.38413805e-01 -5.61995566e-01 -1.53632998e-01 5.19846916e-01 -5.03508925e-01 6.45488650e-02 -2.39238858e-01 2.84534782e-01 1.51815951e-01 3.98144543e-01 9.79719609e-02 1.47065961e+00 -8.25324357e-01 -9.21495140e-01 4.79366571e-01 2.70124406e-01 -1.68049291e-01 1.60908327e-01 3.62221785e-02 5.48621655e-01 -5.96913695e-01 9.69623506e-01 5.13676345e-01 -6.97283804e-01 -2.18716353e-01 -2.23869205e-01 6.26313269e-01 -9.12256017e-02 -5.95136404e-01 1.20968902e+00 -4.86672670e-01 1.30811021e-01 1.38026401e-01 -1.13299847e+00 4.68125701e-01 6.19432390e-01 6.85108840e-01 -3.59228164e-01 5.72485685e-01 4.10236984e-01 1.88192487e-01 -7.23352253e-01 -1.26119480e-01 -6.66038036e-01 9.68654826e-02 5.67854106e-01 -2.82676220e-01 -7.31061339e-01 -1.95564777e-01 1.94966272e-01 1.41818559e+00 -1.72793105e-01 4.62441951e-01 -7.36887828e-02 4.01848584e-01 3.04339379e-01 9.02793705e-02 7.73271918e-01 -3.55728894e-01 9.06917036e-01 1.16081074e-01 -5.25967240e-01 -9.23129857e-01 -1.13308942e+00 -4.66780096e-01 5.29560924e-01 -1.18408971e-01 6.47747397e-01 -8.56441975e-01 -1.13744152e+00 -2.73610614e-02 6.15678966e-01 -8.43401730e-01 -1.93542227e-01 -5.31336010e-01 -8.93186152e-01 7.01140940e-01 9.17846262e-01 7.97991335e-01 -1.13259184e+00 -5.90996444e-01 1.38945624e-01 -2.09214002e-01 -8.41006160e-01 -2.73254991e-01 5.03321171e-01 -1.13506591e+00 -1.24227178e+00 -1.32961512e+00 -1.08064044e+00 9.94516134e-01 1.57515891e-03 1.30492353e+00 4.29451138e-01 -8.46133769e-01 5.55510700e-01 2.24942877e-03 -6.36350155e-01 -1.05746317e+00 1.15585305e-01 -2.34250486e-01 -7.34879732e-01 2.67028242e-01 -1.61958382e-01 -7.54404724e-01 3.54542077e-01 -9.97794449e-01 2.60327160e-01 8.77692938e-01 9.65589523e-01 8.40389669e-01 1.48372278e-01 2.86589205e-01 -1.47160113e+00 2.28007987e-01 -5.97853184e-01 -1.46364570e-01 2.92924076e-01 -1.00641184e-01 -1.89569116e-01 3.61881226e-01 -1.74262881e-01 -1.00721550e+00 3.99461240e-01 -5.82054794e-01 -1.08111881e-01 -3.85288924e-01 2.60258377e-01 3.20560217e-01 -2.41376534e-01 1.02121449e+00 -1.93237364e-01 -2.20728785e-01 1.08216055e-01 -2.46748142e-02 6.99065447e-01 8.86283100e-01 -1.10536538e-01 8.52294922e-01 7.02488661e-01 3.79547805e-01 -4.87405896e-01 -1.06472254e+00 -4.39033568e-01 -5.43592215e-01 1.36579443e-02 1.35184348e+00 -5.92553735e-01 -3.24297041e-01 2.41111249e-01 -6.71227157e-01 -4.13056046e-01 -4.78017688e-01 5.76373994e-01 -4.85376418e-01 3.31352800e-01 -8.75044346e-01 -2.97320545e-01 -6.76309943e-01 -1.25282323e+00 1.21105123e+00 2.32779846e-01 -2.27252483e-01 -1.27759957e+00 -2.93413609e-01 7.21677244e-01 4.54227716e-01 4.97976750e-01 1.00404024e+00 -7.54469156e-01 -5.33181012e-01 -5.68281710e-01 -2.92094857e-01 3.18613052e-01 3.34081769e-01 -1.89788282e-01 -9.46672857e-01 -2.20989600e-01 9.87055227e-02 -4.78269666e-01 7.98805535e-01 9.86931622e-01 1.81222391e+00 -1.68205500e-02 -8.12317491e-01 8.52436721e-01 1.11284995e+00 4.53981161e-01 6.20435953e-01 1.01010196e-01 8.47026110e-01 4.00984973e-01 3.37694496e-01 -2.25129381e-01 9.26429927e-02 -3.59096825e-02 6.01864159e-01 -6.66779637e-01 -3.05694997e-01 1.97418407e-02 -2.62927830e-01 4.69478071e-01 -9.79818683e-03 -4.11102325e-01 -1.10898709e+00 4.20305461e-01 -1.42506778e+00 -7.53674865e-01 -2.45923758e-01 1.79657686e+00 8.56401384e-01 1.62664533e-01 -2.83772767e-01 1.30318627e-01 6.08170867e-01 -5.74467540e-01 -6.37911558e-01 -1.59992114e-01 4.41938430e-01 6.14792407e-01 7.89234400e-01 2.92518318e-01 -1.42916703e+00 4.97728407e-01 6.46547461e+00 7.19241381e-01 -1.13206208e+00 2.08126530e-01 1.35326028e+00 2.07582638e-01 -1.62260264e-01 -5.92212617e-01 -5.94745636e-01 1.37965426e-01 8.25451791e-01 2.60804474e-01 -3.38347673e-01 9.95678604e-01 -2.18442827e-02 -2.94470191e-01 -1.29340899e+00 8.13955367e-01 -9.68930051e-02 -1.31802952e+00 -1.38676748e-01 3.14192213e-02 9.96785522e-01 -1.67453513e-01 2.43698731e-01 2.76868194e-01 5.49907565e-01 -1.57280505e+00 4.15491126e-02 3.41661304e-01 9.47945714e-01 -5.16384721e-01 1.14294052e+00 5.37245870e-01 -8.02135468e-01 1.89937428e-01 -2.73789316e-01 6.15441442e-01 -1.08908154e-01 4.53908563e-01 -1.52377617e+00 3.00375730e-01 7.28609085e-01 2.48542890e-01 -4.48503315e-01 9.97535527e-01 -5.59974313e-02 7.91878879e-01 -3.92384499e-01 4.44420613e-02 3.88034195e-01 3.75864983e-01 1.19150177e-01 1.19502354e+00 6.91832006e-01 3.89885962e-01 2.33489528e-01 7.31595874e-01 -2.27713630e-01 -1.87495306e-01 -5.91405451e-01 1.10117599e-01 -1.80175707e-01 1.48422480e+00 -1.28922069e+00 -5.24006248e-01 -1.57113820e-01 9.79151666e-01 -3.88348550e-01 -1.04419038e-01 -1.00162458e+00 7.83733353e-02 -3.83324772e-01 3.32271099e-01 2.57967301e-02 4.67056781e-01 -5.03247082e-01 -6.96136236e-01 -3.98935646e-01 -8.08369040e-01 4.95796859e-01 -7.83670902e-01 -1.21867061e+00 6.75036728e-01 -3.74201648e-02 -1.08197570e+00 -5.12708426e-01 -7.06538320e-01 -7.18054354e-01 7.66118705e-01 -1.38247490e+00 -1.28762114e+00 -6.80708468e-01 5.47098756e-01 5.18868268e-01 -1.31658778e-01 8.37132037e-01 1.05453387e-01 -1.66013375e-01 5.96722424e-01 -1.83676988e-01 3.51380974e-01 5.61738908e-01 -1.64453351e+00 1.72277361e-01 3.94422263e-01 -2.00056598e-01 5.78595363e-02 3.17122370e-01 -6.90176427e-01 -8.04863691e-01 -1.52462077e+00 3.74885917e-01 -4.60949719e-01 3.04037482e-01 2.09682181e-01 -8.47194374e-01 6.08246028e-01 -1.88079346e-02 3.77588540e-01 8.12700987e-01 -7.83002138e-01 4.11159873e-01 1.57623753e-01 -1.79355729e+00 5.25059164e-01 6.01695240e-01 -1.56643376e-01 -4.62872237e-01 9.61815417e-01 5.80452919e-01 -9.64272380e-01 -7.12683201e-01 7.53152668e-01 4.06577528e-01 -7.96644509e-01 1.01289082e+00 -4.32291359e-01 6.35942519e-01 -1.30728045e-02 2.31545910e-01 -1.02319539e+00 -3.66277039e-01 -1.36898071e-01 7.69170523e-02 4.20977622e-01 6.47640228e-01 -3.93757403e-01 1.38929915e+00 7.16134369e-01 -2.23277867e-01 -1.00654638e+00 -6.80694282e-01 -2.76339620e-01 3.53044331e-01 -3.94820780e-01 3.14276278e-01 8.25273216e-01 -5.24673104e-01 -2.97169089e-01 7.65722543e-02 2.15090543e-01 5.89689255e-01 -1.68896437e-01 2.83301890e-01 -9.38959360e-01 -6.34748816e-01 -4.50460762e-01 -4.40817118e-01 -6.91829145e-01 -3.09234262e-01 -1.03678036e+00 1.59191296e-01 -2.05928969e+00 4.61350650e-01 -3.65409106e-01 -2.48770297e-01 6.16695046e-01 -2.20782951e-01 7.24560618e-01 -2.51764476e-01 7.26287514e-02 -9.43343714e-02 -2.98967987e-01 1.82041073e+00 -4.86763954e-01 1.63980424e-01 7.15858996e-01 -7.10558414e-01 1.12509131e+00 9.82650638e-01 -5.23972034e-01 -3.28480601e-01 -1.46621421e-01 -5.16165271e-02 3.58606547e-01 5.80122054e-01 -1.08416629e+00 1.47343814e-01 3.17631215e-02 9.53750730e-01 -9.60187912e-01 -9.68786702e-03 -1.10045385e+00 -9.38669369e-02 1.15352845e+00 -3.48239124e-01 -2.71981120e-01 3.00105155e-01 4.66138482e-01 -5.34767322e-02 -6.42990351e-01 1.19890511e+00 -8.39169979e-01 -3.03753078e-01 5.81898868e-01 -5.07871091e-01 -1.73134804e-01 1.50251675e+00 -4.26767826e-01 8.11841413e-02 -4.09879863e-01 -1.20404494e+00 2.81015247e-01 -5.36101684e-03 -3.64704318e-02 6.85208738e-01 -1.08025825e+00 -8.00627172e-01 -1.09850436e-01 -2.80055195e-01 5.38382888e-01 2.38094226e-01 8.67040575e-01 -1.10136473e+00 3.16785812e-01 5.26044369e-02 -7.21538723e-01 -1.52170730e+00 2.84025371e-01 8.67390394e-01 -7.59382486e-01 -4.78717685e-01 1.25190485e+00 3.52781624e-01 -3.84797662e-01 5.30141354e-01 -7.71039963e-01 -7.74478093e-02 -5.89342058e-01 4.01883066e-01 1.18415430e-01 2.78186262e-01 -1.67769700e-01 -2.04214722e-01 4.46757853e-01 -3.19583565e-01 1.73606247e-01 1.08988822e+00 2.33121961e-01 -3.66969518e-02 2.20576208e-02 9.90265906e-01 -3.20194781e-01 -7.41626441e-01 1.66143194e-01 -2.66857386e-01 1.20611303e-01 2.61282235e-01 -1.22775984e+00 -1.41291308e+00 1.07626188e+00 1.08285022e+00 3.01860899e-01 1.15770161e+00 1.67335033e-01 1.03159130e+00 5.17829537e-01 7.65118748e-02 -6.71436548e-01 6.48072436e-02 -2.02302262e-02 6.19948745e-01 -1.64192331e+00 2.65964657e-01 -4.63746637e-01 -6.51484489e-01 1.45618963e+00 7.44429827e-01 -8.04872587e-02 9.02127802e-01 6.59190476e-01 3.40160906e-01 -4.48046058e-01 -1.64703786e-01 -2.25348398e-02 5.10716617e-01 6.26527131e-01 5.09149432e-01 3.51783842e-01 -4.28791121e-02 6.22958422e-01 -2.56286830e-01 9.05359387e-02 3.13995123e-01 1.04553795e+00 -5.13857901e-01 -1.04417264e+00 -7.07739651e-01 1.02990258e+00 -9.39360738e-01 -9.18525010e-02 -5.15319943e-01 8.96244466e-01 3.11628014e-01 5.05276382e-01 -1.46445826e-01 -2.79942229e-02 -6.65579084e-03 4.22591120e-02 5.38857758e-01 -8.69986236e-01 -6.68719530e-01 -1.08487559e-02 -1.98319346e-01 -1.54192939e-01 -3.17785084e-01 -1.03923932e-01 -1.54622400e+00 -1.06914595e-01 -3.99559766e-01 -2.20914602e-01 6.51454985e-01 8.93596947e-01 -2.25720152e-01 9.74271476e-01 2.65841454e-01 -6.13389790e-01 -7.32347131e-01 -8.96888793e-01 -5.35180867e-01 2.36540794e-01 2.58136541e-01 -2.28464663e-01 -1.78747669e-01 2.43743718e-01]	[15.234183311462402, -2.1379740238189697]
e890a60f-84c8-4b5b-8b37-d52aedad828e	around-the-world-in-60-words-a-generative	2302.01614	null	https://arxiv.org/abs/2302.01614v1	https://arxiv.org/pdf/2302.01614v1.pdf	Around the world in 60 words: A generative vocabulary test for online research	Conducting experiments with diverse participants in their native languages can uncover insights into culture, cognition, and language that may not be revealed otherwise. However, conducting these experiments online makes it difficult to validate self-reported language proficiency. Furthermore, existing proficiency tests are small and cover only a few languages. We present an automated pipeline to generate vocabulary tests using text from Wikipedia. Our pipeline samples rare nouns and creates pseudowords with the same low-level statistics. Six behavioral experiments (N=236) in six countries and eight languages show that (a) our test can distinguish between native speakers of closely related languages, (b) the test is reliable ($r=0.82$), and (c) performance strongly correlates with existing tests (LexTale) and self-reports. We further show that test accuracy is negatively correlated with the linguistic distance between the tested and the native language. Our test, available in eight languages, can easily be extended to other languages.	['Nori Jacoby', 'Elisabeth André', 'Francesca Lanzarini', 'Ilia Sucholutsky', 'Raja Marjieh', 'Harin Lee', 'Yue Sun', 'Pol van Rijn']	2023-02-03	null	null	null	null	['culture']	['speech']	[-5.30868292e-01 -2.57010847e-01 -3.22818637e-01 -2.93261766e-01 -8.17141652e-01 -1.28790152e+00 6.91039562e-01 4.18368489e-01 -8.46344769e-01 9.97865021e-01 5.05665004e-01 -5.12789190e-01 7.33362436e-02 -8.32015038e-01 -6.79996252e-01 1.31937131e-01 1.72843382e-01 4.65468109e-01 1.01457469e-01 -1.90784276e-01 2.73030549e-01 -7.99493343e-02 -1.48090601e+00 5.27860317e-03 1.56161952e+00 2.61753760e-02 2.26492003e-01 5.23993015e-01 -2.16951534e-01 5.68407476e-01 -7.84219265e-01 -7.99842179e-01 1.77310213e-01 -5.72292387e-01 -6.99215174e-01 -4.12879288e-01 9.27089036e-01 -4.72097903e-01 -8.50857701e-03 1.22963870e+00 1.62672564e-01 5.55173531e-02 5.51265597e-01 -5.69248557e-01 -1.22312760e+00 9.38659489e-01 -4.10901792e-02 1.94983512e-01 7.52657056e-01 4.55719382e-01 9.24376786e-01 -7.17774093e-01 9.23690736e-01 1.08058238e+00 6.56147659e-01 4.11347836e-01 -1.25358307e+00 -1.07751429e+00 2.59356350e-01 -3.69354755e-01 -1.45214844e+00 -4.15153027e-01 -1.04337744e-01 -7.98793852e-01 1.03803158e+00 -1.21127985e-01 1.12123775e+00 1.45015645e+00 -6.52657747e-02 8.49994048e-02 1.71947551e+00 -2.96950638e-01 2.63131354e-02 5.78517675e-01 3.03075194e-01 7.75682449e-01 9.45431292e-01 3.15545723e-02 -9.15508270e-01 2.09657952e-01 5.93036890e-01 -6.50732219e-01 -1.35847300e-01 6.20725304e-02 -1.68016839e+00 6.53246939e-01 -6.07335605e-02 7.10479677e-01 1.71473064e-02 -4.43627805e-01 1.91980571e-01 7.87555397e-01 2.12046936e-01 9.42120373e-01 -4.38294590e-01 -6.13950789e-01 -8.28096449e-01 2.42097870e-01 1.00996757e+00 9.36871827e-01 3.93682867e-01 2.70622581e-01 -1.81725081e-02 9.40763175e-01 1.08669564e-01 1.04693770e+00 8.87315869e-01 -6.45616651e-01 5.56448758e-01 4.00809824e-01 2.23433971e-01 -5.31688035e-01 -3.25610936e-01 -6.41547501e-01 -1.85849369e-01 1.70429543e-01 9.66427863e-01 -3.20324510e-01 -4.35853630e-01 2.07351303e+00 -1.55007675e-01 -3.15887451e-01 -8.08591023e-02 6.51691675e-01 7.62109697e-01 2.13849068e-01 3.45515937e-01 -2.35479221e-01 1.17509699e+00 -3.77447426e-01 -3.66853714e-01 -6.39181733e-01 9.41627681e-01 -7.14460075e-01 1.57576382e+00 4.74980205e-01 -1.27559245e+00 -7.55157888e-01 -1.23459589e+00 -1.20111518e-01 -6.18975639e-01 -1.53442428e-01 7.78126955e-01 1.36356878e+00 -1.50635064e+00 1.79983914e-01 -4.79182482e-01 -5.78017175e-01 -9.29810852e-02 3.69820148e-01 -7.19625771e-01 -1.45525590e-01 -1.13919103e+00 8.14716458e-01 4.20060813e-01 -5.59600651e-01 -5.70712268e-01 -5.20601332e-01 -9.04093325e-01 -2.27028891e-01 -5.94655499e-02 -2.70397335e-01 1.00671530e+00 -1.00781858e+00 -1.56081295e+00 1.19853401e+00 -8.22298229e-02 1.13601133e-01 3.54146779e-01 -1.69002950e-01 -7.40674853e-01 -2.22294465e-01 7.15107918e-01 4.64276671e-01 1.29809350e-01 -3.48200411e-01 -6.78544521e-01 -2.93731034e-01 -1.26143992e-01 3.12970489e-01 -7.15097904e-01 6.13797158e-02 -2.03409046e-01 -5.96266389e-01 -4.84024845e-02 -5.14552593e-01 4.29575592e-01 -4.17180002e-01 -9.22435299e-02 -1.17294587e-01 -3.15456539e-01 -1.17847693e+00 1.14261258e+00 -2.11560941e+00 -1.44480258e-01 5.70591331e-01 2.61664480e-01 2.60651316e-02 -2.05673158e-01 1.22833729e-01 4.80284929e-01 7.46121526e-01 2.34524012e-01 1.43590003e-01 3.75940681e-01 -4.17571783e-01 2.45665032e-02 5.24734795e-01 2.34142896e-02 1.07566953e+00 -1.05157685e+00 -2.97742784e-01 -1.37239560e-01 7.68887326e-02 -9.13269520e-01 -7.81914741e-02 -6.47710562e-02 5.47786415e-01 1.49248645e-01 5.42351365e-01 3.48874658e-01 1.62458584e-01 3.55763286e-01 8.56177211e-01 -6.74451292e-01 8.18160832e-01 -9.18681622e-01 1.53888750e+00 -4.42038447e-01 8.97087812e-01 -2.48741172e-02 -2.02631637e-01 9.21306014e-01 -1.54992193e-01 -3.41202885e-01 -1.07387638e+00 -5.14716581e-02 6.76081359e-01 6.90246284e-01 -4.40181196e-01 3.00634146e-01 5.00497073e-02 -1.57551616e-01 5.84028363e-01 1.53867751e-01 -2.81535119e-01 7.60085642e-01 -1.22596957e-01 1.01156652e+00 1.53561503e-01 4.65119928e-01 -7.74788082e-01 2.27120072e-01 1.37817532e-01 6.13362253e-01 1.06429470e+00 -3.46038342e-01 -3.30933444e-02 4.78496879e-01 1.91352531e-01 -1.07882810e+00 -1.51281238e+00 -2.14659125e-01 1.27531767e+00 -3.06478888e-01 -4.61006492e-01 -6.17070854e-01 -2.23880813e-01 1.31451577e-01 9.63010192e-01 -3.62886876e-01 -6.53928742e-02 -2.72745103e-01 -5.13446657e-03 6.94105864e-01 4.99449581e-01 5.06325126e-01 -6.54681504e-01 1.68175533e-01 -7.92353526e-02 -1.74136311e-01 -1.16832483e+00 -5.99004686e-01 -5.14009655e-01 -5.62531531e-01 -7.96710670e-01 -6.64248526e-01 -1.15227711e+00 3.37545514e-01 -1.63303525e-03 1.32110059e+00 2.15617299e-01 2.49870032e-01 4.15327519e-01 -2.15505943e-01 -3.81896973e-01 -4.68487471e-01 3.11149687e-01 5.14452159e-01 -6.85921550e-01 1.11204314e+00 -2.31306851e-01 -9.29477215e-02 1.19277582e-01 -1.77818358e-01 -2.49829337e-01 6.05030537e-01 4.57095742e-01 1.27613053e-01 -3.16845238e-01 5.12202978e-01 -7.73964107e-01 6.87856674e-01 -5.98911762e-01 -7.96178162e-01 2.87265122e-01 -5.72944701e-01 -1.11381099e-01 6.14889085e-01 -7.84699559e-01 -6.37337565e-01 -5.41975200e-01 -5.63016944e-02 6.62645519e-01 -3.77982438e-01 5.93876362e-01 2.18774691e-01 7.61104608e-03 8.00598323e-01 1.82648689e-01 -1.83408603e-01 -1.20263420e-01 7.51086324e-02 7.66403437e-01 4.79707390e-01 -7.84258008e-01 9.03127432e-01 -1.12981692e-01 -7.12626934e-01 -1.25535214e+00 -3.85064781e-01 -7.80017823e-02 -7.27291882e-01 -2.26468131e-01 7.76751697e-01 -1.40842378e+00 -9.94598866e-01 3.21209997e-01 -5.00047803e-01 -8.09974015e-01 3.57156061e-02 1.47864556e+00 -8.25060308e-02 -1.79958325e-02 -5.31646371e-01 -6.26474142e-01 2.27478915e-03 -1.02078652e+00 6.21925414e-01 2.76214331e-01 -8.78393114e-01 -1.08588290e+00 2.41731301e-01 2.58923739e-01 3.02130491e-01 -2.54775494e-01 1.00740075e+00 -9.94203389e-01 -3.66860390e-01 -1.16633333e-01 -3.59831601e-02 8.83736536e-02 -2.41322294e-01 8.27213675e-02 -5.15375733e-01 -2.81037122e-01 -3.71493936e-01 -8.05196822e-01 4.47576195e-01 2.90753841e-01 3.37425739e-01 4.14462835e-02 1.44875482e-01 5.84236741e-01 1.10377979e+00 -1.46100849e-01 3.07116687e-01 4.17557150e-01 3.14888269e-01 5.90595245e-01 1.14644572e-01 -3.16951610e-02 5.44377983e-01 3.18799525e-01 -8.37676823e-01 5.36211908e-01 -1.52370155e-01 -5.22024393e-01 1.15384042e+00 1.31695688e+00 1.05211236e-01 5.25423437e-02 -1.17216516e+00 6.45940244e-01 -8.82585585e-01 -9.67459023e-01 -3.33043575e-01 2.64224768e+00 1.08170366e+00 4.90024835e-01 5.04950166e-01 -2.51263291e-01 5.54638028e-01 -2.93710560e-01 -4.81764883e-01 -5.70652425e-01 -6.43814266e-01 6.24718070e-01 4.06061769e-01 9.46110368e-01 -5.35221994e-01 1.31377363e+00 7.40046835e+00 3.62050384e-01 -1.07445407e+00 1.27713829e-02 2.16870680e-01 -3.90023172e-01 -6.63533449e-01 -2.08529070e-01 -8.91017020e-01 2.48824120e-01 1.14928544e+00 -6.62743330e-01 8.13775480e-01 4.22189593e-01 -9.71047804e-02 -1.81764066e-01 -1.07331967e+00 7.09103107e-01 2.16614291e-01 -6.29815519e-01 -1.30622223e-01 -3.16838324e-02 7.59783328e-01 3.18709821e-01 2.70793885e-01 8.97387207e-01 7.51849771e-01 -1.11690569e+00 1.05813062e+00 4.15113330e-01 1.38670731e+00 -4.95670974e-01 3.18120331e-01 2.92623430e-01 -7.53615022e-01 -2.46850718e-02 -3.78725439e-01 -6.97547793e-01 -3.76389712e-01 3.63455266e-01 -7.44356692e-01 -2.46257097e-01 6.25446737e-01 4.01262015e-01 -1.00272274e+00 8.38576078e-01 -6.99602544e-01 1.00554752e+00 -3.27758580e-01 -6.43665075e-01 -7.43595734e-02 -2.52826720e-01 2.55035371e-01 1.09352636e+00 5.53843319e-01 -8.42537731e-02 4.07706648e-02 9.27872777e-01 -2.20004231e-01 8.69919538e-01 -8.59904468e-01 -7.13822901e-01 6.40925050e-01 7.55684495e-01 -6.45317256e-01 -2.12513030e-01 -1.05355632e+00 7.91416109e-01 7.39708245e-01 5.28323233e-01 -2.97340363e-01 -4.45306718e-01 7.38196671e-01 2.78123409e-01 -8.99712294e-02 -7.60487020e-01 -5.68272054e-01 -1.41654682e+00 1.46936402e-01 -1.10248303e+00 1.01586007e-01 -6.20554626e-01 -1.18621385e+00 1.10235497e-01 -2.98314422e-01 -7.15362370e-01 -3.15600872e-01 -1.03883052e+00 -5.57743162e-02 1.35129201e+00 -7.95844316e-01 -5.14802873e-01 -2.43487045e-01 4.29350317e-01 2.33042046e-01 -5.52024722e-01 9.24000978e-01 1.87975690e-01 -4.82216001e-01 8.72010231e-01 1.56798571e-01 5.80094218e-01 1.16960621e+00 -1.23320901e+00 5.34281790e-01 8.81383836e-01 3.94746006e-01 1.15205491e+00 3.12652588e-01 -1.18464839e+00 -9.33242559e-01 -6.19353175e-01 1.22436285e+00 -6.81574404e-01 1.07991970e+00 -7.90682375e-01 -5.87776840e-01 8.46752644e-01 2.99788952e-01 -8.05684745e-01 1.09157217e+00 8.50852013e-01 -7.27652192e-01 1.24340080e-01 -8.74239147e-01 8.92351329e-01 1.44756997e+00 -1.02211118e+00 -5.18113613e-01 4.19735372e-01 6.43635035e-01 -1.83301479e-01 -9.89454508e-01 -1.86225474e-01 9.80174363e-01 -8.98731112e-01 4.95997280e-01 -6.22301042e-01 4.32002217e-01 1.29833475e-01 -1.29206534e-02 -1.55102789e+00 -5.73227584e-01 -3.81533653e-01 5.89655101e-01 1.19472313e+00 8.26099873e-01 -9.50417161e-01 1.02975897e-01 7.53598273e-01 8.87123793e-02 1.08486332e-01 -7.14045346e-01 -9.38136160e-01 7.53982782e-01 -4.22049493e-01 4.91732776e-01 1.28177571e+00 3.74664456e-01 2.14421377e-01 3.19899023e-01 1.39690980e-01 2.99215853e-01 -3.15517843e-01 7.02778339e-01 -1.55692267e+00 -2.95373976e-01 -8.95086527e-01 -3.98408473e-01 -5.87652385e-01 6.83289886e-01 -1.32947981e+00 -3.40495646e-01 -1.19549596e+00 2.87027687e-01 -1.89020634e-01 1.17654987e-01 2.19328001e-01 -3.54570061e-01 2.12939456e-01 -1.18543066e-01 -1.73504204e-01 -3.25356007e-01 -4.79547121e-02 9.53040659e-01 6.97309971e-02 -3.76902699e-01 -3.74594510e-01 -1.13216019e+00 5.24975300e-01 1.03095174e+00 -1.30349159e-01 -3.05184990e-01 -7.82395422e-01 6.53969586e-01 -5.17226398e-01 1.31256714e-01 -1.31530011e+00 1.99084088e-01 -2.18249664e-01 6.66483343e-01 1.80058286e-01 -1.94193900e-01 -5.08966744e-02 -2.26939723e-01 4.46750909e-01 -2.76341915e-01 3.92419010e-01 4.21339929e-01 -2.33962774e-01 3.99608873e-02 -2.79772103e-01 4.25637513e-01 -2.33591214e-01 -6.75809801e-01 -9.60972905e-02 -5.41933715e-01 7.24161446e-01 6.59129322e-01 -1.33501127e-01 -6.22326851e-01 -4.87885356e-01 -1.58783972e-01 1.03877246e-01 8.35585594e-01 5.14201760e-01 4.49315868e-02 -1.09219277e+00 -1.11731184e+00 4.62018758e-01 3.09107691e-01 -9.77372050e-01 -2.56523401e-01 6.46815240e-01 -7.29338527e-01 6.66046083e-01 -2.52795637e-01 -1.44043490e-01 -7.38291264e-01 4.00433004e-01 1.03692852e-01 2.83010602e-01 -1.74622193e-01 8.13930571e-01 -1.14661358e-01 -6.42947853e-01 1.03598358e-02 -1.52948126e-01 -2.71821231e-01 7.51957819e-02 6.92909300e-01 2.30081961e-01 -2.63498425e-01 -6.80532515e-01 -1.27971381e-01 5.32414317e-01 -1.53825492e-01 -8.51649046e-01 1.03726065e+00 9.97073799e-02 -2.77670592e-01 1.00501502e+00 7.40909517e-01 1.08655167e+00 -4.57428515e-01 -3.99505794e-02 -1.64485767e-01 -4.03843105e-01 -2.97875881e-01 -9.06907797e-01 -2.88464993e-01 4.51996207e-01 2.18023688e-01 -1.27144372e-02 5.59934616e-01 -3.03033441e-01 1.97526321e-01 5.17812848e-01 5.29024780e-01 -1.60575259e+00 -2.87221938e-01 9.70278680e-01 5.13626397e-01 -1.08553660e+00 -8.50176960e-02 -1.96088493e-01 -4.31880623e-01 6.47446990e-01 9.22038078e-01 1.64047793e-01 4.80894357e-01 4.51934487e-02 7.51544237e-02 1.75447553e-01 -5.31588674e-01 -4.40172702e-01 3.05234551e-01 8.81936789e-01 1.17258525e+00 3.55085909e-01 -9.75194752e-01 7.86598265e-01 -1.29854453e+00 -1.88628599e-01 5.32644629e-01 5.44460058e-01 -4.75815564e-01 -1.04721284e+00 -4.74006563e-01 7.60585666e-01 -3.75560045e-01 -5.93531847e-01 -8.83069456e-01 1.24369752e+00 1.73995599e-01 1.14789987e+00 3.26597095e-01 -4.25044835e-01 2.35346481e-01 5.19175053e-01 6.37448013e-01 -8.78105760e-01 -5.81640780e-01 -3.51151288e-01 1.90070242e-01 -3.59532923e-01 -5.55905420e-03 -1.07453728e+00 -8.80275786e-01 -6.76797867e-01 1.13918871e-01 1.15801901e-01 3.61063302e-01 8.11308384e-01 -1.63016468e-02 5.51569313e-02 -6.90893978e-02 8.62956718e-02 -3.32006902e-01 -1.23206925e+00 -8.08790147e-01 3.98237258e-01 2.45536175e-02 -4.01118904e-01 -3.39238405e-01 -3.14934611e-01]	[10.823559761047363, 10.026910781860352]
c04360a0-2a51-452b-b0c9-049082b8d93a	distributed-optimization-for-reactive-power	2302.09241	null	https://arxiv.org/abs/2302.09241v2	https://arxiv.org/pdf/2302.09241v2.pdf	Distributed Optimization for Reactive Power Sharing and Stability of Inverter-Based Resources Under Voltage Limits	Reactive power sharing and containment of voltages within limits for inverter-based resources (IBRs) are two important, yet coupled objectives in ac networks. In this article, we propose a distributed control technique to simultaneously achieve these objectives. Our controller consists of two components: a purely local nonlinear integral controller which adjusts the IBR voltage setpoint, and a distributed primal-dual optimizer that coordinates reactive power sharing between the IBRs. The controller prioritizes the voltage containment objective over reactive power sharing at all points in time; excluding the IBRs with saturated voltages, it provides reactive power sharing among all the IBRs. Considering the voltage saturation and the coupling between voltage and angle dynamics, a formal closed-loop stability analysis based on singular perturbation theory is provided, yielding practical tuning guidance for the overall control system. To validate the effectiveness of the proposed controller for different case studies, we apply it to a low-voltage microgrid and a microgrid adapted from the CIGRE medium-voltage network benchmark, both simulated in the MATLAB/Simulink environment.	['Gilbert Bergna-Diaz', 'John W. Simpson-Porco', 'Babak Abdolmaleki']	2023-02-18	null	null	null	null	['distributed-optimization']	['methodology']	[-0.25744477 0.14964168 -0.2829394 0.41682202 -0.3621028 -1.0959083 0.12021749 0.334474 0.2453997 1.2202216 -0.3710196 -0.1913228 -0.7440739 -0.7278242 -0.333259 -1.2754232 -0.35326436 -0.02269368 -0.1902037 -0.5852471 0.02223462 0.6491808 -0.95982563 -0.6689218 1.1515671 1.0767632 -0.27027282 0.19809121 0.8176027 0.4111179 -0.94433403 0.29457048 0.2677869 -0.2376037 -0.10467619 0.13659458 -0.3186033 -0.22321542 -0.10329296 1.3800166 0.72347784 0.3839796 0.59488803 -1.9724905 -0.3595362 0.36266944 -0.78267664 0.33071074 0.23356166 0.18343829 0.9653659 -0.413383 0.24016768 0.74968696 0.37159815 0.10254549 -1.6277325 -0.54402655 0.47637624 0.07558884 -1.730238 -0.02235244 0.76521677 -0.4222338 0.8815151 0.6144346 0.9214581 0.01468748 0.39677075 0.1585541 0.6565845 -0.16822523 0.4152976 0.05043891 -0.08889976 0.07798231 0.6337873 -0.26066864 0.1286518 -0.34507746 0.55273545 -0.32083964 -1.0517795 -0.7953113 -0.79935986 0.8230094 0.49964225 0.3539696 -0.48089477 -0.24663645 0.13757132 0.3674286 0.48530987 0.05699317 -0.35203144 0.5006965 -0.4839045 0.23665361 0.85656893 0.95300853 0.31254566 0.8432419 -0.07018133 0.35990703 0.34979454 0.71199834 -0.02461576 -0.71879137 0.25245023 0.6407072 0.47910145 -0.9924916 -0.29743916 -0.6920993 -0.7553708 0.46940932 0.19997579 -0.71602625 0.04877704 1.8041052 0.53309596 -0.38017488 -0.07635015 1.033321 -0.21039507 0.9749878 -0.3611639 -1.1157804 1.2817401 -0.31164992 -1.1879653 0.3249463 0.34725377 -0.44789335 0.4012819 0.1850834 -1.348916 0.03765152 -1.5244786 0.40997547 -0.38508666 0.23322104 -0.41813836 0.3617302 -1.3065745 0.4243587 -0.3389194 -0.08395435 -0.32730407 0.40087956 0.09818547 0.7269105 -1.500119 1.061284 0.4495503 0.41967222 -0.22988129 -1.0304552 -0.7116663 0.13958395 0.7084659 -0.70184267 0.9182276 -0.84996074 -1.6957477 0.035153 0.13771738 -0.40620947 0.41680062 0.3464856 -0.30444166 0.20504546 -0.07725286 -0.3672649 0.7054012 -1.1063006 -0.5981507 -0.23815082 0.023644 0.5347285 -0.24355243 -0.33051443 0.30386332 -0.6350964 -0.2619918 -0.5099394 -0.24634276 -0.19135307 -0.42246598 -0.3568389 1.01468 -0.64580065 1.217377 -1.9019052 0.7660947 0.6516474 -0.06200875 0.07230026 0.43819085 0.8254647 -0.21527748 -0.14939573 0.12701593 0.27830207 0.15592979 0.11792221 -0.13763365 1.0647814 0.22599466 0.37397105 -0.8341291 0.13469832 0.3160603 0.55528533 -0.32657066 -0.0828372 -0.05021854 0.3008618 -0.5157808 0.2419442 0.70661217 0.00954177 0.6830858 -0.6479059 -0.64452296 -0.42704058 -1.686561 0.97255266 -0.393517 0.31817728 1.1658762 -1.4444019 0.6669614 0.41225702 0.6954674 -0.5545903 0.34539318 0.07102302 -0.14866407 0.00753318 -0.10690037 0.02676049 0.07917101 0.252158 0.19325964 -0.4491643 0.6269604 0.24697687 0.33073536 -0.34626135 0.60757065 -1.0987536 1.244932 -0.21281202 1.0169463 -0.3872143 -0.18197103 -0.3931904 0.86963415 0.3413099 -0.7949515 -0.86723125 -0.17092445 0.53563195 0.42362943 0.0706704 -0.37620312 -0.0776096 0.2713952 1.0948796 -0.2967431 -0.39566076 -0.34202835 -0.73812026 -0.38445064 0.13496128 -0.03932076 0.15337461 -0.5044931 0.25579783 0.16445349 -0.8589955 -0.657094 0.36041957 -0.31225485 -1.1907635 -0.6205773 -0.71828204 0.90666705 -0.20800936 0.76201373 0.0257296 -0.14378002 0.47448418 0.27521735 -0.28217322 -0.13282132 -0.31661683 0.47969687 0.1225987 -0.54962426 -0.5616958 -0.4699219 0.54557806 -0.6100352 -0.11865872 -0.18144268 0.664809 0.57828915 0.62733054 1.2007763 -0.02031492 0.7483176 -0.6352337 -1.6756563 0.22709405 -0.6294961 -0.37291786 1.134014 -0.25431338 -0.74433166 -0.16513358 0.27737796 -0.26717547 0.5215916 0.24395949 -0.7074188 -0.4735151 -0.17752524 0.17491528 0.3235037 -0.23774546 0.25536066 0.38860953 0.34695268 -0.4213445 1.1885327 0.17691556 0.4379934 -0.7246183 -0.14333111 -0.05741346 -0.32041174 -0.37889907 0.36787677 -1.0092694 -1.4862511 0.33321434 -0.86686313 -0.12187921 -0.28975663 0.18069836 -0.5974675 0.06483042 -0.3674185 -0.91940606 -0.50526524 -1.2712411 0.35611504 0.6977733 0.2409278 -1.3545114 -0.05304681 -0.23649491 0.42025626 0.8845775 0.9433319 -0.20846443 -0.41396698 0.11676824 0.24341843 0.49075398 0.09969439 0.31168112 -0.18406925 -1.1061232 0.22842099 0.20967193 -0.42766416 0.41664383 0.37867656 -0.67003524 -0.2746624 0.5326962 2.3761785 0.6379034 -0.06742323 0.23590761 0.29922545 0.5353332 0.41378248 0.56470114 0.5920057 0.71188617 0.8125392 -0.18823633 0.72637737 0.5210117 0.30338484 0.59909207 0.22250576 -0.34333405 -0.40864277 0.63331014 -1.7972491 -0.37954 -0.09897378 2.2700577 0.6505815 -0.21937713 0.21472502 0.5176468 1.0023105 -0.15908849 -0.6480946 -0.6592125 -0.27721336 -0.25918415 0.8012679 0.5002389 -0.7114076 -0.37613997 4.8996263 0.671282 -1.3867592 -0.0159867 0.5241786 -0.39592278 0.01689825 -0.21021086 -0.45647892 0.6202214 1.001116 -1.3316032 0.7508113 0.65962744 0.8951515 -0.24483126 -0.69546497 0.34478632 -0.22837815 -1.036142 -0.31976977 -0.05662203 1.0620428 -0.31358615 -0.16969794 -0.06827476 -0.07952158 -0.28468966 0.8871295 0.25800973 0.2983667 -1.4251875 0.66886 0.36256373 -1.2677718 -0.48383385 0.15026414 -0.01512139 0.63071376 0.58004177 -0.01262598 1.302548 0.26388898 0.53227437 -0.08934094 0.8202949 -0.36240298 0.0487482 -0.3760232 -0.06942008 -0.01381996 -0.8461575 0.81658614 0.38273868 -0.01636923 0.29371387 0.4381902 0.9056049 0.28544256 0.08718931 -0.2865623 0.33050168 0.59177196 1.4482317 -0.32535684 -0.15609907 -0.34787154 0.34007972 -0.24471581 0.8049774 -0.9991341 -0.76157606 1.3040786 -0.18440513 0.03121491 -0.08496703 -0.266606 -0.869451 0.18847947 -0.7417719 0.38132367 -0.6139006 -1.0823061 0.07851407 0.23895688 -1.3475246 -0.6519325 -0.20481531 -0.82063687 1.4179536 -1.7241702 -0.57929474 0.26532453 0.6585236 0.19300312 0.1693468 0.41805312 0.5936766 -1.3068943 0.23023188 0.6642151 -0.21317467 -0.08185305 -1.291309 -0.5793116 1.0937856 -1.0210023 0.44154853 0.9751022 0.05879619 -2.0489764 -0.89974177 0.32341528 0.76942706 1.1339346 -0.3026489 -0.8957351 0.52237415 0.83213717 0.05890688 0.100678 -0.96870434 0.5765358 -0.4681491 -1.521198 0.6646635 0.08852627 -0.20435746 -0.25816303 0.28757718 0.507596 -0.47813997 -1.3121314 0.46156564 -0.03259984 -0.23918866 0.9091293 0.04425441 -0.45921564 -0.9425806 -0.10732273 -2.1075602 -0.09103753 -1.1843193 -0.41401568 1.4394084 0.05281942 -1.2866291 0.10568023 0.1393118 0.21498744 -0.8075015 -1.2437096 -0.77126354 0.22870171 0.49788743 0.45238447 0.9161325 0.64749664 0.4856655 0.29977098 1.1108454 1.0340458 -0.01380745 0.20130795 -0.79500645 0.0310775 -0.5526753 -0.21430914 -0.18723032 0.15932119 -0.4023265 -0.3074028 -1.5617415 -0.6519919 0.08923569 -0.10268177 0.28594396 -0.08882686 0.08315147 0.20097366 -0.20098816 0.06669363 0.5448139 1.0112722 -0.33930823 -0.25458136 0.04366915 -0.28244904 0.52005726 0.8948938 0.32596108 -0.9477322 -0.04677535 -0.03627619 0.6364066 -0.03561993 -1.0009756 0.18029593 -0.30274194 -0.10185585 -0.7140433 0.01762656 -1.3655878 0.49390048 0.89324784 0.16987385 0.18454558 0.13319951 0.3098226 -0.19162917 0.16618992 1.1368346 0.5136093 -0.35522062 -0.05380238 -0.7647688 -0.05101278 1.7237927 0.1375177 -0.46361884 -0.5288657 -0.7430642 1.2622933 0.45138222 0.23515265 -0.08063943 -1.1253726 -0.515975 0.223199 -0.40692037 -0.2431232 0.19862837 1.1660643 -0.27559915 0.6653528 -0.04192271 -0.72628546 -0.99160135 0.58115786 0.9401514 0.0483527 -0.05708639 0.1333202 -0.31894624 0.0661663 0.05245534 -0.48145497 -0.3006426 0.6477761 0.35518217 0.8051705 0.2718983 -0.64527935 -0.2661584 0.5977142 0.7056793 0.25347498 1.2305018 -0.7240408 -0.38076878 0.36069983 1.2618394 -0.22006701 -1.0971644 0.31424603 -0.3458308 -0.10553493 0.24586032 -0.50213957 -1.6270924 0.03111678 0.27701673 1.0340257 1.4092511 -0.9073054 0.11694983 -0.28712773 0.7388578 -1.1247274 -0.5808524 0.39953336 0.88350934 -0.4360316 0.1606626 -0.61386746 -0.30022848 0.9830872 0.4123096 -0.50320476 0.5938315 0.5092802 -0.28468278 0.3888604 -0.8707425 0.27185965 0.04652108 0.42228213 0.27316594 -0.00739174 -1.1540204 0.35067022 0.48995447 -0.10022473 0.96038824 0.9263347 -0.16434313 -0.6173484 -0.73917586 0.12221627 -0.591919 0.52571875 0.3721481 1.0352281 0.09389269 0.9938743 0.36693344 0.6086186 0.91756374 -0.11890515 0.10652771 -0.15949547 -0.6424231 0.26655805 0.01806458 -0.42362013 -0.01473271 -0.48356658 -1.3638602 -0.44374278 -0.7418506 0.7039837 0.6792085 0.7646992 0.05258255 0.81810385 1.1506324 -0.8761736 -0.98234135 -0.4650784 -1.0622722 -0.27668768 0.48629367 -0.6267593 -0.80093724 -0.40480554]	[5.649630069732666, 2.580583095550537]
5ae2b46a-00c5-4be3-8ad3-4232c2dc4813	using-k-way-co-occurrences-for-learning-word	1709.01199	null	http://arxiv.org/abs/1709.01199v1	http://arxiv.org/pdf/1709.01199v1.pdf	Using $k$-way Co-occurrences for Learning Word Embeddings	Co-occurrences between two words provide useful insights into the semantics of those words. Consequently, numerous prior work on word embedding learning have used co-occurrences between two words as the training signal for learning word embeddings. However, in natural language texts it is common for multiple words to be related and co-occurring in the same context. We extend the notion of co-occurrences to cover $k(\geq\!\!2)$-way co-occurrences among a set of $k$-words. Specifically, we prove a theoretical relationship between the joint probability of $k(\geq\!\!2)$ words, and the sum of $\ell_2$ norms of their embeddings. Next, we propose a learning objective motivated by our theoretical result that utilises $k$-way co-occurrences for learning word embeddings. Our experimental results show that the derived theoretical relationship does indeed hold empirically, and despite data sparsity, for some smaller $k$ values, $k$-way embeddings perform comparably or better than $2$-way embeddings in a range of tasks.	['Ken-ichi Kawarabayashi', 'Yuichi Yoshida', 'Danushka Bollegala']	2017-09-05	null	null	null	null	['learning-word-embeddings']	['methodology']	[-2.60803878e-01 -2.58251578e-02 -4.93519783e-01 -3.12355548e-01 -5.91046929e-01 -3.17696899e-01 3.46033543e-01 7.90912092e-01 -8.73628199e-01 2.86302477e-01 2.35785559e-01 -4.50395793e-01 -3.55279863e-01 -1.02083969e+00 -5.10118186e-01 -6.93084121e-01 -7.25878596e-01 -2.95011327e-02 -9.32512805e-02 -2.09137350e-01 2.77804077e-01 8.56416151e-02 -1.51153946e+00 -2.70252347e-01 4.29456562e-01 9.26714540e-01 1.80530399e-01 4.22925830e-01 -4.62280899e-01 3.58363916e-03 -3.80420357e-01 -6.11285210e-01 3.39343846e-01 -3.62977326e-01 -6.98911428e-01 -1.62368238e-01 2.91399747e-01 2.79625118e-01 -3.19497943e-01 1.20561886e+00 1.70110717e-01 5.86182296e-01 8.55818450e-01 -1.01026881e+00 -1.00034130e+00 5.59543431e-01 -5.18688917e-01 4.08404380e-01 1.52540550e-01 -5.74697070e-02 1.82645667e+00 -1.03184390e+00 9.48077962e-02 9.08159554e-01 6.69940293e-01 3.07841957e-01 -1.07116902e+00 -6.88709438e-01 2.58188903e-01 1.39500469e-01 -1.69906831e+00 4.38739592e-03 7.60656357e-01 -2.99743384e-01 1.30202639e+00 6.03501983e-02 6.39026105e-01 4.72609520e-01 1.19015835e-01 3.92823070e-01 6.89967275e-01 -8.43171000e-01 5.97436950e-02 3.26778650e-01 4.11075294e-01 6.70981765e-01 4.90318120e-01 -1.47405922e-01 -6.06161118e-01 -1.23687305e-01 3.48670125e-01 1.08785167e-01 -1.84933230e-01 -2.22590297e-01 -6.66699767e-01 1.29686213e+00 4.47494030e-01 8.34094644e-01 -1.13672718e-01 6.66788757e-01 4.59524482e-01 2.23564416e-01 5.63046873e-01 5.32858014e-01 -4.21103299e-01 -2.24672154e-01 -5.50874293e-01 1.66836962e-01 4.07823503e-01 8.57036173e-01 1.10587943e+00 -1.93065956e-01 3.42996091e-01 9.06519413e-01 4.65933442e-01 3.12711060e-01 6.91969752e-01 -2.51437634e-01 3.16079915e-01 5.17778456e-01 -5.11578731e-02 -1.33763385e+00 -3.45005915e-02 -2.02858418e-01 -5.68677068e-01 -3.83847296e-01 3.23855758e-01 1.87552035e-01 -5.31105936e-01 2.19001174e+00 7.46747665e-03 9.58739817e-02 -2.13836022e-02 5.04309237e-01 3.67240280e-01 7.21050203e-01 2.26486683e-01 -3.03031564e-01 1.55486035e+00 -5.87981045e-01 -7.90118337e-01 -5.04161835e-01 1.04238868e+00 -7.03683436e-01 1.52160084e+00 -1.59373999e-01 -8.00854027e-01 -4.10575002e-01 -9.28936124e-01 9.57328081e-03 -9.34877753e-01 -3.52901429e-01 7.27811992e-01 9.41198945e-01 -7.23210394e-01 2.82989979e-01 -3.07865739e-01 -3.58513027e-01 2.96036035e-01 2.31345922e-01 -3.80409837e-01 -2.62616515e-01 -1.36583018e+00 8.87427926e-01 4.48180825e-01 -3.79847974e-01 -2.56963342e-01 -6.60728037e-01 -1.24550951e+00 8.79496410e-02 1.88093886e-01 -8.48315060e-02 7.54289329e-01 -2.90513515e-01 -6.95998430e-01 9.76668417e-01 -5.33825755e-01 -3.20941269e-01 -4.32793677e-01 -2.25768507e-01 -6.45205438e-01 -1.45330265e-01 6.70645535e-02 2.94648707e-01 7.55813658e-01 -1.00868797e+00 -6.63637698e-01 -3.07624578e-01 2.27722868e-01 4.59760129e-02 -1.01731050e+00 -2.13720560e-01 -1.17474459e-01 -7.34975159e-01 3.87597345e-02 -5.56763828e-01 -7.43006840e-02 6.54562889e-03 -7.56921023e-02 -7.58268356e-01 5.02385497e-01 -1.86355621e-01 1.77922130e+00 -2.55038762e+00 -1.53397247e-01 4.30588484e-01 2.77875721e-01 3.15672785e-01 -2.84284770e-01 7.69366026e-01 -2.41738379e-01 3.55840266e-01 -2.76125431e-01 -3.20012033e-01 3.35702211e-01 4.51901138e-01 -4.04712617e-01 5.08596063e-01 8.40278119e-02 7.37277865e-01 -1.02529764e+00 -1.48096666e-01 1.61877707e-01 3.47818941e-01 -5.45041084e-01 1.48780197e-01 -4.66227857e-03 -5.32712877e-01 -2.36414939e-01 8.84581078e-03 4.56784904e-01 -1.28656521e-01 2.61471361e-01 -1.06287971e-01 1.76163658e-01 6.17015421e-01 -1.10185337e+00 1.48290575e+00 -7.98466206e-01 5.74640393e-01 -2.68528700e-01 -1.51670277e+00 1.06321919e+00 2.39403158e-01 4.47209507e-01 -6.03378594e-01 7.50498474e-02 4.00479317e-01 4.22687419e-02 -5.37028849e-01 6.73489094e-01 -9.11118150e-01 -4.39325809e-01 7.69691229e-01 1.80274218e-01 -1.73791796e-01 3.48873943e-01 5.93285374e-02 1.03469908e+00 -7.03969538e-01 5.08100390e-01 -5.38858593e-01 4.28133190e-01 -4.46546793e-01 2.83110499e-01 5.13829648e-01 -3.54364634e-01 7.15239570e-02 5.56901813e-01 -3.59771937e-01 -1.02077234e+00 -1.19311905e+00 -4.38973188e-01 1.33133471e+00 2.98105359e-01 -8.41214061e-01 -4.10879880e-01 -6.49530411e-01 2.28544593e-01 9.33466434e-01 -9.09768283e-01 -4.01281446e-01 -4.24390495e-01 -8.29409182e-01 5.76705575e-01 7.96646237e-01 -8.51473212e-02 -7.70347893e-01 -5.42587399e-01 1.63005874e-01 2.04540610e-01 -8.57643485e-01 -8.23865533e-01 6.25773013e-01 -5.36500096e-01 -1.05667651e+00 -5.46085060e-01 -1.09380615e+00 5.98111987e-01 4.59013075e-01 1.17191923e+00 9.31097493e-02 -5.47193706e-01 5.09452522e-01 -6.53860748e-01 -1.89250320e-01 7.63806328e-02 -1.91733271e-01 3.64008695e-01 -1.73612103e-01 1.14589369e+00 -6.90769196e-01 -5.87386787e-01 2.75943819e-02 -1.34635007e+00 -7.11212814e-01 2.54869521e-01 9.42945600e-01 6.16042674e-01 3.89983058e-01 5.70252955e-01 -5.25863826e-01 8.67874861e-01 -5.45033455e-01 -2.28539668e-02 4.85658459e-02 -7.09116936e-01 2.62446761e-01 7.58522391e-01 -5.71095467e-01 -5.57743385e-02 -6.55123651e-01 -2.12275565e-01 -2.60512412e-01 9.80782975e-03 7.85703957e-01 5.64068463e-03 3.66598397e-01 4.95868117e-01 3.69171500e-01 -1.84852779e-01 -2.37841353e-01 9.39966977e-01 5.91429591e-01 -1.14880204e-01 -6.02693260e-01 6.99453652e-01 2.74077624e-01 -3.28403801e-01 -1.19144917e+00 -6.84412897e-01 -7.89661825e-01 -2.05698386e-01 4.14628804e-01 7.69604385e-01 -6.36863649e-01 -4.38180804e-01 -1.47587374e-01 -7.91709960e-01 8.52513090e-02 -5.58321118e-01 5.85154533e-01 -3.46068472e-01 5.10379016e-01 -1.22707695e-01 -9.07215655e-01 -1.58647120e-01 -8.87416899e-01 7.02101111e-01 -2.78976802e-02 -6.34779453e-01 -1.35648024e+00 2.61864692e-01 -1.89316809e-01 3.70095193e-01 -2.37360284e-01 1.61565650e+00 -8.00355911e-01 2.01699093e-01 -5.83980739e-01 -2.65789360e-01 6.14319146e-01 5.61675072e-01 -4.79599535e-01 -5.81142783e-01 -3.55534494e-01 -4.29329611e-02 -2.77962238e-01 8.31138194e-01 1.41811654e-01 1.09528077e+00 -3.49559486e-01 -3.35433215e-01 5.39320223e-02 1.63820183e+00 1.34437129e-01 5.35698712e-01 3.66615430e-02 5.08354664e-01 5.80855429e-01 2.45079830e-01 4.63807821e-01 2.73659497e-01 5.60443819e-01 3.25625718e-01 3.86078030e-01 2.37717673e-01 -3.44326198e-01 2.60963500e-01 1.32205522e+00 4.43565130e-01 -2.72195071e-01 -8.82473588e-01 1.35744011e+00 -1.20634472e+00 -6.76623225e-01 1.18241467e-01 2.30704188e+00 1.00810206e+00 2.31554881e-01 8.07033554e-02 2.57318020e-01 5.94382405e-01 7.29223907e-01 -3.03613842e-01 -7.80629277e-01 9.80006382e-02 1.13077819e+00 4.37832326e-01 5.39736390e-01 -7.79420853e-01 8.27952385e-01 6.02342224e+00 1.27155292e+00 -9.61774707e-01 2.61804402e-01 3.51980597e-01 -6.35567158e-02 -7.64356256e-01 -9.42260101e-02 -6.72269821e-01 6.17233157e-01 8.95547271e-01 -4.68810081e-01 1.06413268e-01 6.76853120e-01 -1.56994447e-01 -1.10498637e-01 -1.17700577e+00 1.12547576e+00 2.56358117e-01 -1.03932881e+00 5.34238517e-02 2.25099787e-01 5.36025524e-01 -3.39622021e-01 4.51542109e-01 2.67188013e-01 2.33808041e-01 -1.41192746e+00 2.93197542e-01 -2.30053186e-01 9.91550803e-01 -1.07529283e+00 6.03295982e-01 2.73683310e-01 -1.66189170e+00 -4.16678973e-02 -6.27788365e-01 -2.05110192e-01 1.42498106e-01 9.79996741e-01 -5.35696507e-01 2.77164876e-01 5.28158486e-01 5.89416921e-01 -9.67740119e-02 3.94032121e-01 -2.41063088e-01 4.60607529e-01 -3.51685464e-01 -4.09471154e-01 5.39609551e-01 -1.36551157e-01 1.61978543e-01 1.29728580e+00 3.67772907e-01 1.20034002e-01 -5.04337810e-02 7.05183148e-01 -1.94415256e-01 4.08059269e-01 -8.16307187e-01 -6.36428356e-01 6.20752633e-01 7.79096484e-01 -4.78881985e-01 -1.30384848e-01 -6.02991164e-01 6.96274459e-01 4.90631312e-01 9.12292376e-02 -7.24612534e-01 -8.88798535e-01 1.58862817e+00 2.37275407e-01 6.84633553e-01 -7.29511440e-01 -1.93119809e-01 -8.99287403e-01 2.56395221e-01 -3.86434078e-01 3.38802844e-01 -2.36986130e-01 -1.79493964e+00 4.67465252e-01 -3.93414609e-02 -9.45519388e-01 -5.26689850e-02 -9.23488021e-01 -4.83764172e-01 1.06782353e+00 -1.48842347e+00 -5.66174209e-01 4.24432158e-01 3.26870680e-01 4.18017417e-01 2.20376812e-02 1.26220071e+00 2.21176073e-01 -3.20710331e-01 1.01360369e+00 3.37285370e-01 3.42691094e-01 4.09293264e-01 -1.15726173e+00 2.56819934e-01 6.42122626e-01 7.01552451e-01 1.08072698e+00 6.71528995e-01 -2.90154904e-01 -1.27722490e+00 -9.32726085e-01 1.30637681e+00 -2.79572129e-01 8.95932794e-01 -4.54816818e-01 -9.22214985e-01 5.38946867e-01 1.74770936e-01 3.62791628e-01 1.32650399e+00 4.45407957e-01 -9.63016748e-01 -8.89747888e-02 -1.08621931e+00 5.78209937e-01 9.73693967e-01 -1.01384211e+00 -7.81201720e-01 2.81069726e-02 8.47119153e-01 2.45700493e-01 -9.36938167e-01 1.83047250e-01 4.43063200e-01 -6.61695957e-01 9.69743073e-01 -7.63204515e-01 4.65277463e-01 -3.51036452e-02 -8.23365331e-01 -1.38610005e+00 -1.08294517e-01 -1.89898029e-01 -1.43677324e-01 7.71118641e-01 4.47443962e-01 -7.68110752e-01 5.35836637e-01 3.55761588e-01 1.75827280e-01 -1.00186515e+00 -1.28398299e+00 -1.09514344e+00 7.48325288e-01 -9.23497558e-01 3.81088793e-01 1.03520930e+00 5.82308710e-01 7.71940276e-02 -2.47189924e-02 4.59937118e-02 3.96908224e-01 -1.15348741e-01 3.22483420e-01 -6.97919786e-01 -3.41783911e-01 -7.15803385e-01 -7.83526719e-01 -1.41389525e+00 2.92612493e-01 -9.74024951e-01 1.49217278e-01 -1.06956494e+00 4.57382128e-02 -8.58753622e-01 -5.74782848e-01 2.71428108e-01 -2.84854800e-01 3.16498876e-01 -2.74737389e-03 -1.92146376e-01 -3.17940801e-01 7.63357699e-01 7.26242661e-01 -5.64228147e-02 1.60238713e-01 -5.60990810e-01 -8.59422386e-01 5.13134658e-01 5.67995071e-01 -4.56437826e-01 -4.31321144e-01 -5.53548217e-01 6.65125489e-01 -6.15488052e-01 -6.63492084e-02 -6.64165735e-01 -1.90248638e-01 -1.76738873e-01 -3.54879111e-01 -1.09320320e-01 4.19030070e-01 -8.82331729e-01 -4.38103169e-01 2.98214614e-01 -5.40298462e-01 1.98673934e-01 1.57086328e-01 7.84741223e-01 -3.34943146e-01 -5.55041254e-01 7.20017076e-01 1.84729453e-02 -7.20940888e-01 1.72044337e-01 -2.65593767e-01 4.47687000e-01 1.14878118e+00 -3.55182022e-01 2.02023581e-01 -2.74064958e-01 -5.24935067e-01 -4.45233583e-02 1.95612192e-01 4.91097897e-01 7.04590738e-01 -1.44530797e+00 -2.96357006e-01 4.08252686e-01 6.09377682e-01 -8.06619897e-02 3.90660539e-02 4.27387476e-01 -2.83869773e-01 4.20855284e-01 3.37415248e-01 -2.68834293e-01 -8.58018935e-01 6.25106454e-01 4.69795354e-02 -1.93087474e-01 -3.80921036e-01 1.41402972e+00 1.02429047e-01 -2.76098430e-01 2.45022774e-01 -3.91934365e-01 8.87206644e-02 1.84347048e-01 6.79557621e-01 6.09981082e-02 -2.46324250e-03 -7.16639578e-01 -5.10263145e-01 6.84215128e-01 -2.63311565e-01 -3.21617752e-01 1.32672393e+00 -8.79850164e-02 -2.04964086e-01 8.53278458e-01 1.84214914e+00 -9.19090509e-02 -5.69251537e-01 -6.21477962e-01 -1.32752433e-02 -8.18250775e-01 -2.07071140e-01 -1.15680546e-01 -9.20619905e-01 1.03145695e+00 5.08654416e-01 5.91005206e-01 8.01647782e-01 4.98708248e-01 1.08127642e+00 1.13754123e-01 4.36411858e-01 -1.01165676e+00 3.58661413e-01 5.84320784e-01 2.70045012e-01 -9.25207198e-01 -1.94616005e-01 -3.98253053e-01 -2.02136502e-01 9.23384726e-01 3.20468664e-01 -4.35339302e-01 1.17101943e+00 -1.43643975e-01 -1.38406813e-01 -3.96158963e-01 -4.54418391e-01 -3.37297142e-01 1.36274099e-01 5.33222973e-01 7.29230702e-01 2.78748512e-01 -7.70745933e-01 6.02581143e-01 -3.93604040e-01 -6.25276268e-01 1.91391334e-01 9.47986901e-01 -6.60661936e-01 -1.40122378e+00 -1.03725269e-02 6.20445728e-01 -3.64976525e-01 -4.43344265e-01 1.63383529e-01 6.29003286e-01 3.43139917e-01 9.16888356e-01 4.65237409e-01 -3.77303660e-01 1.42699406e-01 4.40551251e-01 4.82943743e-01 -8.82714629e-01 -2.15301007e-01 -4.36618894e-01 -3.21695060e-01 -1.15683533e-01 -3.16954434e-01 -3.16589981e-01 -1.37553000e+00 -4.82261121e-01 -4.44035977e-01 3.98726702e-01 4.92092967e-01 1.21167314e+00 1.87856928e-01 2.37982333e-01 6.76609933e-01 -2.49072939e-01 -4.74270999e-01 -9.91564691e-01 -1.05214751e+00 7.83021927e-01 1.51327386e-01 -7.94722021e-01 -6.43553495e-01 -1.22631378e-01]	[10.441325187683105, 8.772536277770996]
98be4784-b7ef-4d6b-b2b4-d6e2d6d5e6a9	openp5-benchmarking-foundation-models-for	2306.11134	null	https://arxiv.org/abs/2306.11134v1	https://arxiv.org/pdf/2306.11134v1.pdf	OpenP5: Benchmarking Foundation Models for Recommendation	This paper presents OpenP5, an open-source library for benchmarking foundation models for recommendation under the Pre-train, Personalized Prompt and Predict Paradigm (P5). We consider the implementation of P5 on three dimensions: 1) downstream task, 2) recommendation dataset, and 3) item indexing method. For 1), we provide implementation over two downstream tasks: sequential recommendation and straightforward recommendation. For 2), we surveyed frequently used datasets in recommender system research in recent years and provide implementation on ten datasets. In particular, we provide both single-dataset implementation and the corresponding checkpoints (P5) and another Super P5 (SP5) implementation that is pre-trained on all of the datasets, which supports recommendation across various domains with one model. For 3), we provide implementation of three item indexing methods to create item IDs: random indexing, sequential indexing, and collaborative indexing. We also provide comprehensive evaluation results of the library over the two downstream tasks, the ten datasets, and the three item indexing methods to facilitate reproducibility and future research. We open-source the code and the pre-trained checkpoints of the OpenP5 library at https://github.com/agiresearch/OpenP5.	['Yongfeng Zhang', 'Wenyue Hua', 'Shuyuan Xu']	2023-06-19	null	null	null	null	['sequential-recommendation', 'benchmarking', 'benchmarking']	['miscellaneous', 'miscellaneous', 'robots']	[-2.44413957e-01 -3.80631626e-01 -6.26338005e-01 -5.00081718e-01 -9.82532561e-01 -8.08798075e-01 5.83961785e-01 -1.85174957e-01 3.42704728e-02 4.16700155e-01 6.56736612e-01 -4.38951761e-01 -7.81437337e-01 -5.41648507e-01 -6.00486577e-01 -3.95509928e-01 -2.00145021e-02 8.01997185e-01 3.64808887e-01 -3.64904970e-01 6.66798890e-01 2.65185624e-01 -1.93009460e+00 8.38477492e-01 6.47817194e-01 9.13383722e-01 4.00854498e-01 7.11297512e-01 -4.29219678e-02 5.25917292e-01 -3.59465718e-01 -2.10854992e-01 3.67639184e-01 2.61111110e-01 -9.55842495e-01 -5.01684785e-01 3.91069412e-01 -3.50234330e-01 -1.06729828e-01 2.85208553e-01 7.42852986e-01 5.60914516e-01 7.16839254e-01 -1.39186573e+00 -9.11253512e-01 1.00376546e+00 -1.48545980e-01 1.02385521e-01 6.00510120e-01 -6.37179464e-02 1.03694856e+00 -1.21289420e+00 3.73256475e-01 9.56174254e-01 8.71100187e-01 2.64598966e-01 -9.84713972e-01 -5.50959945e-01 1.94147229e-01 1.74910761e-02 -1.29547048e+00 -6.50106132e-01 4.08256240e-02 -4.91782576e-01 1.26703990e+00 5.58921039e-01 1.57297716e-01 1.24900723e+00 -7.67828524e-02 5.19152164e-01 1.01434159e+00 -1.45453364e-01 1.28796071e-01 3.17885697e-01 1.06752348e+00 3.02345455e-01 7.62454560e-03 3.80666763e-01 -7.45729923e-01 -5.81342936e-01 4.47491795e-01 3.73587698e-01 -1.76191598e-01 2.29947671e-01 -1.23002541e+00 6.45141065e-01 -6.53114244e-02 2.41716698e-04 -6.12780988e-01 -9.71341953e-02 3.21255565e-01 6.82154596e-01 5.44224918e-01 4.16998714e-01 -9.19186294e-01 -2.43169844e-01 -8.88408422e-01 5.52406311e-01 1.36119759e+00 1.26890123e+00 4.24173504e-01 -4.42362905e-01 -6.85628951e-01 1.21144152e+00 3.75577986e-01 5.48758924e-01 6.20846450e-01 -1.16825831e+00 2.99580395e-01 2.33520627e-01 4.90869135e-01 -7.43757069e-01 -5.96095204e-01 -5.25555253e-01 -3.38576287e-01 -4.89577025e-01 9.09901187e-02 -1.63776889e-01 -4.77505982e-01 1.32506573e+00 3.16466391e-01 3.36325765e-01 1.13951094e-01 9.16876733e-01 1.46793115e+00 7.11788714e-01 -2.23502025e-01 -1.90723330e-01 1.62606001e+00 -1.69464886e+00 -2.99207449e-01 1.25846222e-01 7.86466897e-01 -1.21754718e+00 1.28276086e+00 5.80707252e-01 -1.01621819e+00 -7.90482044e-01 -7.25634813e-01 -1.80387497e-01 -5.96971571e-01 3.24554384e-01 6.94686353e-01 3.04791003e-01 -1.29628396e+00 7.11747289e-01 -5.82754910e-01 -7.05311477e-01 -3.12035441e-01 4.16925341e-01 -5.90052679e-02 -2.64703810e-01 -1.18707597e+00 6.19384170e-01 2.20074952e-02 -5.50660193e-01 -7.74449170e-01 -9.73163962e-01 -4.57515232e-02 1.31198227e-01 2.20066890e-01 -8.49193156e-01 1.57549894e+00 -3.36304933e-01 -1.58862257e+00 5.31455100e-01 8.66778195e-02 -4.13101017e-02 -7.28605762e-02 -7.06438661e-01 -6.93617582e-01 -3.82737100e-01 -1.02739215e-01 3.41048092e-01 5.50788701e-01 -9.43603635e-01 -7.48257697e-01 -1.82803571e-01 2.68247932e-01 3.55317622e-01 -2.24066123e-01 5.26031613e-01 -9.10197020e-01 -7.84406364e-01 -3.99282902e-01 -1.12770188e+00 -7.38087744e-02 -5.39646387e-01 -2.68852711e-01 -4.55302954e-01 2.54195660e-01 -5.57671726e-01 1.56797373e+00 -2.12929273e+00 -2.35852742e-04 1.03980981e-01 -2.23738983e-01 2.12964132e-01 -6.58533633e-01 1.00668144e+00 1.38078094e-01 -1.50339112e-01 6.21116877e-01 -5.06423175e-01 3.24836910e-01 4.00558203e-01 -6.97130024e-01 2.55355854e-02 -5.09797692e-01 4.56150323e-01 -7.13176429e-01 -1.24130793e-01 -5.82899004e-02 4.60251302e-01 -9.12170708e-01 3.62218350e-01 -3.67163241e-01 1.85372517e-01 -4.88150388e-01 4.78321135e-01 6.34086847e-01 -5.83330393e-01 8.24720487e-02 -4.48705733e-01 -3.49271864e-01 9.63259161e-01 -1.46377230e+00 1.73494351e+00 -5.30910015e-01 -3.18990976e-01 -1.50812909e-01 -4.77816075e-01 8.09273422e-01 3.39472741e-01 5.66925347e-01 -4.77341622e-01 -3.78890544e-01 2.20689744e-01 -5.86718261e-01 -3.73668820e-01 1.06047273e+00 7.06971645e-01 -9.15778894e-03 1.19301033e+00 1.50603428e-01 3.18548024e-01 3.53811413e-01 3.73737574e-01 1.28461349e+00 2.00212419e-01 5.90588944e-03 -3.84682298e-01 3.30065370e-01 -7.24084452e-02 1.96922615e-01 1.23447025e+00 5.84007919e-01 5.19112468e-01 -2.21451931e-02 -4.32420909e-01 -4.36806142e-01 -7.46759236e-01 -2.27008685e-01 2.35465026e+00 -1.22549653e-01 -9.67900932e-01 -1.60914227e-01 -7.97830760e-01 4.34978694e-01 9.23363090e-01 -3.12282354e-01 1.81567729e-01 -3.00135255e-01 -6.11833096e-01 2.38629267e-01 4.55594003e-01 1.62536036e-02 -1.15370679e+00 6.79527819e-02 2.01323256e-01 -2.56349802e-01 -6.19504869e-01 -8.90778720e-01 2.55819559e-01 -6.81717634e-01 -1.04932189e+00 -3.61579806e-01 -5.53084731e-01 2.01519251e-01 1.03914821e+00 1.52787423e+00 3.90074015e-01 3.41850668e-01 7.86552787e-01 -9.32576597e-01 -1.95071511e-02 -1.02086075e-01 3.65510255e-01 3.20103347e-01 -5.17261565e-01 4.19796705e-01 -5.90021610e-01 -9.27298248e-01 1.15219867e+00 -5.45802891e-01 1.44489437e-01 3.03248107e-01 5.07804334e-01 7.47564375e-01 -3.55159402e-01 6.00384891e-01 -1.25800335e+00 8.46657634e-01 -1.16121686e+00 -4.96382773e-01 3.47326666e-01 -1.15471148e+00 -1.97470486e-01 5.29204965e-01 -2.86656976e-01 -9.29342091e-01 -4.11825746e-01 -4.50418085e-01 -8.74594599e-02 -2.13257715e-01 6.93038225e-01 4.11308944e-01 2.67393321e-01 9.14811552e-01 1.18620671e-01 -1.99851647e-01 -1.20997429e+00 7.45115817e-01 1.23702264e+00 4.41351026e-01 -1.11314440e+00 2.45818332e-01 1.17428228e-02 -6.56188250e-01 -1.65662184e-01 -1.31927204e+00 -8.88474345e-01 -1.66475013e-01 6.07174337e-02 1.81022093e-01 -1.03238356e+00 -5.88333368e-01 6.93372041e-02 -8.69977534e-01 -6.76890910e-01 -3.03675443e-01 5.15269279e-01 -4.87733394e-01 3.58241377e-03 -8.61150920e-01 -2.76054621e-01 -1.08503556e+00 -1.13089681e+00 1.21244407e+00 5.28402766e-03 -1.13211028e-01 -7.48676956e-01 6.02980137e-01 5.69871604e-01 8.07652354e-01 -6.60803914e-01 5.14545679e-01 -1.22948158e+00 -1.23208128e-01 -5.94771393e-02 -3.37479681e-01 2.07545191e-01 -1.99950427e-01 2.29622293e-02 -8.21288347e-01 -5.41709244e-01 -4.21254486e-01 -4.01267290e-01 6.50689960e-01 1.98212221e-01 1.24004197e+00 -3.78903657e-01 -4.83836651e-01 7.20788240e-01 1.08556545e+00 -2.29596756e-02 3.54694903e-01 3.40589672e-01 3.98361295e-01 3.65337223e-01 1.12115967e+00 7.05057502e-01 7.36066699e-01 1.12904477e+00 -4.03358415e-02 2.61613041e-01 -2.61538327e-01 -1.08165525e-01 6.92777276e-01 1.39359128e+00 -3.82870644e-01 -5.06085873e-01 -6.21497273e-01 1.48589537e-01 -2.30039668e+00 -8.75363767e-01 -3.57874930e-01 2.41822386e+00 8.60111654e-01 -3.47610384e-01 4.81269836e-01 -3.03742915e-01 3.41255784e-01 -1.09236300e-01 -4.78531301e-01 -3.95985782e-01 2.89176524e-01 2.68421918e-01 3.12908739e-01 2.81560630e-01 -9.26664293e-01 7.06505656e-01 6.53375769e+00 7.33982444e-01 -8.62650812e-01 4.05805737e-01 5.50206341e-02 -2.73929685e-01 -2.17710301e-01 7.69406557e-03 -1.55972660e+00 9.11814034e-01 1.57963300e+00 -2.97060788e-01 7.93029070e-01 1.05540502e+00 1.36210695e-01 2.98779249e-01 -1.34190547e+00 6.51218593e-01 -1.69171561e-02 -1.37612188e+00 5.50770760e-02 1.37936771e-01 7.05892742e-01 8.33477199e-01 2.09301084e-01 9.73975182e-01 7.78622091e-01 -5.29438913e-01 3.29076409e-01 8.57348621e-01 7.51275420e-01 -4.11122948e-01 7.39948928e-01 4.02399689e-01 -1.05115867e+00 -2.45971233e-01 -6.31136656e-01 7.66534060e-02 2.13519763e-02 5.23107648e-01 -2.84136266e-01 7.63944507e-01 9.25835729e-01 1.04949105e+00 -3.82754117e-01 1.01430964e+00 7.14523345e-02 9.09859002e-01 -3.81735772e-01 3.59269261e-01 -2.90570736e-01 -3.65128696e-01 2.80963689e-01 1.41621673e+00 5.32340765e-01 1.92718863e-01 4.95829493e-01 5.35618607e-03 -1.40885105e-02 3.16712588e-01 -2.37593219e-01 3.51651579e-01 1.09350777e+00 1.44252038e+00 -2.10170984e-01 -4.14829731e-01 -7.22396910e-01 5.43749988e-01 4.30007339e-01 2.70219296e-01 -8.90755355e-01 -1.25103161e-01 6.75160229e-01 1.91061929e-01 3.87452066e-01 1.53531507e-01 2.67037481e-01 -1.06479645e+00 -4.75671351e-01 -1.37657583e+00 8.53778720e-01 -7.82796621e-01 -1.73127973e+00 3.34157825e-01 1.48475051e-01 -1.08390129e+00 -1.60979196e-01 -3.84119898e-01 -4.88063633e-01 7.46711433e-01 -1.36974335e+00 -9.14303064e-01 -3.95221651e-01 8.60378683e-01 6.99422002e-01 -2.97313929e-01 1.13346219e+00 9.27791417e-01 -7.18315363e-01 7.87804723e-01 6.07544422e-01 -6.62180841e-01 1.42023945e+00 -1.14572287e+00 4.75425959e-01 1.22268617e-01 5.40617779e-02 1.22377324e+00 4.17724609e-01 -4.26716149e-01 -1.82933187e+00 -1.31634820e+00 6.65959299e-01 -6.69141531e-01 6.60090804e-01 -2.25769612e-03 -7.53053606e-01 9.43657815e-01 1.14471175e-01 -1.82469472e-01 1.15015471e+00 7.69819498e-01 -3.94891232e-01 -3.70401233e-01 -9.05584276e-01 1.04729339e-01 1.24807131e+00 -1.83313891e-01 -5.15065491e-01 1.11760139e+00 9.25722539e-01 -5.11312842e-01 -1.49893343e+00 1.58888325e-01 9.27176595e-01 -9.02643085e-01 1.20270550e+00 -6.57432556e-01 2.67267257e-01 -1.51303098e-01 -5.78370690e-01 -1.21892393e+00 -9.02715862e-01 -6.75345123e-01 -5.98749459e-01 1.32062936e+00 6.29679024e-01 -8.18191171e-01 4.21376377e-01 4.86622393e-01 -5.44320345e-01 -9.10122454e-01 -1.23714477e-01 -7.60200143e-01 -2.75693834e-01 -3.71840030e-01 9.88742769e-01 9.36381340e-01 -9.56312642e-02 4.62769449e-01 -7.06329942e-01 1.40688896e-01 2.74241924e-01 6.65091574e-01 1.17957890e+00 -1.17555821e+00 -1.04207265e+00 -1.30867332e-01 6.27643168e-01 -1.55383253e+00 -2.67572433e-01 -1.22204983e+00 -1.79713696e-01 -1.46774924e+00 4.33060795e-01 -9.57314968e-01 -8.53697777e-01 8.58268499e-01 7.39362389e-02 2.11807624e-01 -1.58923548e-02 9.23829138e-01 -1.03106654e+00 9.28963423e-02 8.10352862e-01 3.94859940e-01 -3.02144259e-01 5.17181516e-01 -1.02137733e+00 4.15502936e-01 7.51024306e-01 -6.21516407e-01 -5.59118211e-01 -5.07564127e-01 4.19984192e-01 1.36539221e-01 -1.21630460e-01 -7.14451909e-01 2.87873745e-01 -1.00879759e-01 -5.32594360e-02 -8.01277876e-01 1.55949578e-01 -6.00259423e-01 4.01733756e-01 6.31043548e-03 -5.39454281e-01 3.61535698e-01 -3.66801955e-02 3.51770788e-01 3.34667146e-01 -1.34022325e-01 3.19814175e-01 1.81665331e-01 -4.22451377e-01 3.33480746e-01 -3.88794579e-02 1.68027114e-02 5.53633451e-01 4.56027925e-01 -1.08539534e+00 -6.20058216e-02 -8.61574709e-01 4.85856920e-01 3.87025386e-01 6.41375840e-01 3.97737175e-01 -1.14712465e+00 -6.93510771e-01 1.62594244e-02 2.47952923e-01 -7.08495498e-01 1.39481142e-01 1.13396907e+00 -2.07404256e-01 5.80889821e-01 -4.74387370e-02 -2.68544525e-01 -1.25883973e+00 6.69176877e-01 -4.23218645e-02 -5.17758787e-01 -5.42523921e-01 6.82982087e-01 -1.89131603e-01 -9.44398522e-01 4.25529897e-01 -8.88678953e-02 -5.37674546e-01 -1.78652983e-02 8.53484035e-01 6.45856917e-01 3.11923206e-01 -2.34375879e-01 -3.18547010e-01 3.55522156e-01 -6.67297125e-01 2.74323046e-01 1.59189653e+00 -7.63935074e-02 -1.30429447e-01 3.04856241e-01 8.20011616e-01 1.44432649e-01 -6.82376027e-01 -6.06409490e-01 -1.24392934e-01 -3.85024726e-01 1.91775113e-01 -1.25918460e+00 -9.37212586e-01 3.26433172e-03 4.59225595e-01 3.39269102e-01 9.51556623e-01 -6.18002266e-02 7.74013579e-01 5.58880746e-01 6.40067697e-01 -8.53556573e-01 -2.81042486e-01 5.89368522e-01 1.08978856e+00 -9.66331363e-01 3.99894655e-01 -2.55699567e-02 -5.65519392e-01 8.01404893e-01 5.22297978e-01 -3.12188901e-02 1.13314199e+00 2.43979126e-01 -6.54089674e-02 -4.14103001e-01 -1.49534106e+00 7.84323812e-02 6.00873649e-01 2.63346791e-01 8.88778687e-01 1.84705362e-01 -6.17152452e-01 1.20259714e+00 -1.00972757e-01 3.64565909e-01 6.30635992e-02 9.05170619e-01 -3.58769685e-01 -1.46797657e+00 -1.67386308e-01 9.34139788e-01 -2.95356631e-01 -3.75374585e-01 -1.21195123e-01 4.05703902e-01 -3.20104919e-02 1.18341863e+00 -2.15676218e-01 -6.40348196e-01 5.28755963e-01 -6.13972992e-02 2.24422216e-01 -8.13838959e-01 -1.20560002e+00 -1.75966606e-01 6.44418955e-01 -9.27591860e-01 -1.82733461e-01 -5.67074239e-01 -8.08934689e-01 -6.86491430e-01 -4.42458004e-01 6.53574109e-01 8.51086676e-01 6.17760777e-01 1.22303355e+00 2.25492984e-01 7.51827538e-01 -1.04621387e+00 -7.96448827e-01 -1.17341733e+00 -6.11646414e-01 2.51122773e-01 -3.06195199e-01 -8.38199556e-01 -2.78656632e-01 -1.60225302e-01]	[10.172045707702637, 5.733040809631348]
27a95049-1376-474c-8a66-adef8298efaf	error-correction-for-dense-semantic-image	1712.03812	null	http://arxiv.org/abs/1712.03812v1	http://arxiv.org/pdf/1712.03812v1.pdf	Error Correction for Dense Semantic Image Labeling	Pixelwise semantic image labeling is an important, yet challenging, task with many applications. Typical approaches to tackle this problem involve either the training of deep networks on vast amounts of images to directly infer the labels or the use of probabilistic graphical models to jointly model the dependencies of the input (i.e. images) and output (i.e. labels). Yet, the former approaches do not capture the structure of the output labels, which is crucial for the performance of dense labeling, and the latter rely on carefully hand-designed priors that require costly parameter tuning via optimization techniques, which in turn leads to long inference times. To alleviate these restrictions, we explore how to arrive at dense semantic pixel labels given both the input image and an initial estimate of the output labels. We propose a parallel architecture that: 1) exploits the context information through a LabelPropagation network to propagate correct labels from nearby pixels to improve the object boundaries, 2) uses a LabelReplacement network to directly replace possibly erroneous, initial labels with new ones, and 3) combines the different intermediate results via a Fusion network to obtain the final per-pixel label. We experimentally validate our approach on two different datasets for the semantic segmentation and face parsing tasks respectively, where we show improvements over the state-of-the-art. We also provide both a quantitative and qualitative analysis of the generated results.	['Luc van Gool', 'Tinne Tuytelaars', 'Yu-Hui Huang', 'Xu Jia', 'Stamatios Georgoulis']	2017-12-11	null	null	null	null	['face-parsing']	['computer-vision']	[ 6.90667808e-01 4.05879080e-01 1.08393587e-01 -7.90825307e-01 -7.61616886e-01 -5.54653823e-01 5.26010573e-01 2.12441266e-01 -4.76294041e-01 5.91106355e-01 -1.73411548e-01 -1.02499865e-01 4.55474220e-02 -7.43231356e-01 -9.59628582e-01 -7.73050725e-01 3.81812602e-01 6.81971014e-01 5.46036661e-01 2.83593953e-01 1.71253145e-01 3.25344026e-01 -1.65305877e+00 2.92243361e-01 7.71834314e-01 1.22397697e+00 3.09114307e-01 4.74262953e-01 -4.06846255e-01 5.61536372e-01 -2.96625495e-01 -5.00931680e-01 1.39037997e-01 -4.79701936e-01 -8.11934292e-01 4.82663691e-01 2.65953571e-01 -2.01263905e-01 3.33925664e-01 1.39442086e+00 8.32854956e-02 -5.62504455e-02 6.19613051e-01 -9.52735126e-01 -2.95232028e-01 6.25356615e-01 -7.78971493e-01 -2.46736825e-01 8.31322186e-03 9.93561894e-02 9.17975783e-01 -7.19324946e-01 5.44321895e-01 1.32472920e+00 6.42113149e-01 3.57285380e-01 -1.24892223e+00 -5.14109313e-01 4.46797460e-01 2.45411955e-02 -1.23266673e+00 -5.65426588e-01 8.24130177e-01 -4.37631696e-01 3.89947176e-01 -1.59601849e-02 3.52467000e-01 8.21523011e-01 -3.48345131e-01 5.81697404e-01 1.14241230e+00 -5.75962543e-01 4.09227371e-01 3.34757119e-01 1.94909915e-01 9.11461115e-01 -1.87226422e-02 -1.25161275e-01 -2.35893413e-01 5.85159920e-02 6.52455032e-01 -1.24765905e-02 -1.17012978e-01 -2.64938235e-01 -7.19041049e-01 7.74612129e-01 5.27509928e-01 2.41028443e-01 -4.47175950e-01 2.46944726e-01 4.46270443e-02 -1.40033722e-01 4.32345510e-01 3.48795094e-02 -5.06716728e-01 3.66788298e-01 -1.11777198e+00 -1.92524940e-02 5.98220229e-01 6.45393729e-01 1.16763806e+00 -5.22487938e-01 -1.52424902e-01 1.01156569e+00 6.80777729e-01 2.64310032e-01 5.77844717e-02 -1.05660594e+00 4.45970982e-01 5.46590924e-01 2.22536072e-01 -9.40988302e-01 -3.41635495e-01 -3.77032280e-01 -4.69337583e-01 2.33780131e-01 8.08067858e-01 -1.43974423e-01 -1.49070275e+00 1.88282073e+00 5.66345572e-01 3.28571469e-01 -9.10774842e-02 9.38865304e-01 5.87732852e-01 5.67572653e-01 4.34711039e-01 2.60778647e-02 1.49029815e+00 -1.04837930e+00 -5.19903362e-01 -5.53032577e-01 5.12889147e-01 -7.45699942e-01 8.77574146e-01 1.54571384e-01 -1.03542185e+00 -5.02782941e-01 -8.03937137e-01 -1.07168630e-01 -3.19147408e-01 5.08199811e-01 4.66344029e-01 6.33276641e-01 -1.09602880e+00 6.31710649e-01 -9.78305459e-01 -2.78162986e-01 7.65151143e-01 4.45298404e-01 -2.00364023e-01 -2.56387442e-01 -9.17847991e-01 6.28408730e-01 4.80595022e-01 4.80914295e-01 -7.98475087e-01 -3.70170593e-01 -8.11770737e-01 1.21812582e-01 6.38805509e-01 -5.42759538e-01 1.16707253e+00 -1.29766190e+00 -1.56752121e+00 9.33043778e-01 -2.79781133e-01 -1.97355717e-01 4.95290279e-01 -1.09784968e-01 1.45943761e-01 1.96635127e-01 9.71556231e-02 1.14522278e+00 7.90029824e-01 -1.58240032e+00 -8.85439694e-01 -4.49717909e-01 1.99066028e-01 2.11247295e-01 -3.74609837e-03 -4.16685827e-02 -9.20334995e-01 -4.20627087e-01 4.19060260e-01 -9.53510463e-01 -2.75514454e-01 1.56267464e-01 -7.54594922e-01 -1.91918805e-01 6.39815092e-01 -8.37928355e-01 6.71454489e-01 -2.13508940e+00 2.03739300e-01 3.72073352e-01 -1.05367519e-01 1.68958023e-01 -8.59356746e-02 -4.99100834e-02 5.85122742e-02 1.18318282e-01 -6.79843962e-01 -8.02734852e-01 -7.54868016e-02 3.35864693e-01 -1.52254194e-01 3.07695627e-01 4.02013481e-01 7.45038271e-01 -7.48050094e-01 -5.95431745e-01 2.89454550e-01 5.37556946e-01 -5.22707701e-01 2.53477246e-01 -6.44193590e-01 7.64294803e-01 -5.08323252e-01 5.16269267e-01 6.46458328e-01 -3.96109760e-01 2.67753452e-01 -2.93693930e-01 4.07091379e-02 1.97894499e-01 -1.34346509e+00 1.73869717e+00 -3.17566872e-01 1.46069303e-01 2.51217723e-01 -1.08147264e+00 6.64892197e-01 1.51879743e-01 2.50917494e-01 -5.05168736e-01 3.84975106e-01 1.71496451e-01 -3.49473029e-01 -3.99657726e-01 4.16300632e-02 -2.96189010e-01 6.38399422e-02 4.99883801e-01 1.36608019e-01 -8.41950849e-02 1.83753908e-01 -3.03808507e-02 7.13942766e-01 4.45983142e-01 -4.77831475e-02 -7.23323959e-04 6.79897130e-01 -1.14368953e-01 8.47979784e-01 5.38781166e-01 2.13193540e-02 7.54184663e-01 8.44177544e-01 -3.22276622e-01 -7.94243336e-01 -8.61420274e-01 -5.64708486e-02 1.07007098e+00 2.57904351e-01 4.22473811e-02 -1.17304158e+00 -1.03039360e+00 -1.35054305e-01 7.28169024e-01 -6.11553013e-01 1.30430132e-01 -4.40039933e-01 -7.88459420e-01 1.36466399e-01 5.45592427e-01 4.84311044e-01 -1.14162517e+00 -4.83069271e-01 8.76598358e-02 -7.38667399e-02 -1.29854751e+00 -2.06941932e-01 3.31535459e-01 -8.13397765e-01 -9.89049256e-01 -5.25612235e-01 -8.07348907e-01 1.07797325e+00 -1.92765161e-01 9.42081273e-01 2.47683257e-01 -3.40318903e-02 -5.64638488e-02 -2.06962273e-01 -1.41155690e-01 -3.48371327e-01 -5.60534820e-02 -4.15556282e-01 3.67686570e-01 9.98585224e-02 -5.69640100e-01 -5.24856210e-01 3.09827387e-01 -1.04141152e+00 3.08072448e-01 8.33967566e-01 7.20512629e-01 8.90703976e-01 2.60447323e-01 4.33634162e-01 -1.45118976e+00 1.63508784e-02 -2.81680077e-01 -9.62042749e-01 4.97689337e-01 -3.77711982e-01 4.49001372e-01 4.03443038e-01 -3.22592020e-01 -1.33436823e+00 5.96695185e-01 -4.05315578e-01 -3.03589642e-01 -4.40513730e-01 3.57789516e-01 -3.64328116e-01 9.99031439e-02 1.94643304e-01 -1.16935231e-01 -2.20194966e-01 -5.96601188e-01 7.05415905e-01 5.68076313e-01 6.51314676e-01 -6.26059473e-01 4.89960611e-01 5.82411408e-01 -9.83739123e-02 -4.06563103e-01 -1.18350005e+00 -2.16413245e-01 -7.92489469e-01 -1.47388428e-01 1.14343023e+00 -7.24440515e-01 -2.99653262e-01 5.69895089e-01 -1.35075641e+00 -4.34395999e-01 -1.52390167e-01 2.21659735e-01 -4.33942050e-01 2.02259943e-01 -6.53394818e-01 -7.21719384e-01 -3.46275084e-02 -1.40980422e+00 1.30174458e+00 5.40604055e-01 -6.60158768e-02 -9.22102213e-01 -3.04037154e-01 5.95488191e-01 5.42945862e-02 1.55964583e-01 1.15805960e+00 -5.24653435e-01 -7.67416000e-01 -1.38734564e-01 -6.67488217e-01 4.04301226e-01 -3.91855054e-02 -7.53067732e-02 -1.18513429e+00 1.62685007e-01 -5.90240769e-02 -2.94717968e-01 9.43012834e-01 4.86121923e-01 1.19934237e+00 -5.22334268e-03 -4.65581208e-01 4.05066878e-01 1.37330425e+00 1.20622972e-02 5.51044405e-01 -1.18091196e-01 8.46055269e-01 1.01594186e+00 4.91463274e-01 2.12431386e-01 4.45374131e-01 6.30497932e-01 4.62234944e-01 -1.98381603e-01 -3.58094752e-01 -2.94569880e-01 1.49240494e-02 4.33414847e-01 1.69589221e-01 -2.92035550e-01 -7.70880997e-01 5.07451475e-01 -1.87797034e+00 -3.95781904e-01 -2.85514332e-02 2.12354541e+00 1.01444447e+00 3.04499239e-01 -2.21099526e-01 2.03679278e-02 1.01341128e+00 7.23636197e-03 -6.12903416e-01 -1.02766939e-01 2.43885204e-01 3.19376200e-01 5.33194363e-01 6.23806536e-01 -1.13780630e+00 1.23632860e+00 5.80633545e+00 6.96266532e-01 -1.02346694e+00 3.09820056e-01 1.05314946e+00 2.38474920e-01 -2.28354499e-01 2.37443790e-01 -7.90776432e-01 5.81263661e-01 6.80075526e-01 6.92372084e-01 3.81233811e-01 6.44973159e-01 9.21706483e-02 -5.21260917e-01 -1.18220210e+00 7.82815456e-01 -1.46470964e-01 -1.00587404e+00 -3.57899331e-02 -1.07232921e-01 7.75663972e-01 -1.88300282e-01 -1.28125727e-01 -1.41980397e-02 5.03789425e-01 -9.23237503e-01 1.02745497e+00 6.33649886e-01 5.37329197e-01 -5.87791085e-01 8.48356605e-01 3.43401611e-01 -8.50443423e-01 -1.26897888e-02 -2.35028282e-01 9.14763063e-02 3.25533867e-01 1.03641891e+00 -6.07233822e-01 3.15035105e-01 5.74834824e-01 4.72391307e-01 -5.83255827e-01 8.17584515e-01 -7.87562728e-01 6.12231910e-01 -5.45152903e-01 3.31101835e-01 2.01486841e-01 -3.19915265e-01 3.73596922e-02 9.08821762e-01 7.67964721e-02 1.37659591e-02 2.12757275e-01 1.16525555e+00 -2.67040431e-01 -1.27156138e-01 -4.42155749e-02 1.21125374e-02 3.70345473e-01 1.44182265e+00 -1.34644198e+00 -4.01332200e-01 -3.25072408e-01 9.34581697e-01 5.10382116e-01 5.24315894e-01 -7.77292252e-01 -9.97253433e-02 3.48389089e-01 4.43128832e-02 5.02384841e-01 9.87883890e-04 -5.47281682e-01 -8.10400903e-01 1.24689311e-01 -4.00114447e-01 2.46771052e-01 -7.74493694e-01 -1.01104951e+00 5.07942438e-01 -1.13762759e-01 -5.06048322e-01 -2.46460482e-01 -4.34602976e-01 -5.45926273e-01 9.16300595e-01 -1.78403175e+00 -1.16032183e+00 -2.93499708e-01 2.02745706e-01 4.40809548e-01 4.84079272e-01 5.82387626e-01 4.46711540e-01 -8.07572305e-01 2.99761295e-01 -3.13713431e-01 1.41460076e-01 4.80984688e-01 -1.17599773e+00 1.64325997e-01 8.53726327e-01 1.93929464e-01 1.30067319e-01 5.96473396e-01 -6.20790243e-01 -7.61640370e-01 -1.10236526e+00 8.50890756e-01 -3.16216886e-01 2.85912335e-01 -4.01450723e-01 -9.16743636e-01 7.01922536e-01 -2.16327365e-02 1.80147395e-01 2.99449027e-01 -3.75951938e-02 -1.92186654e-01 -8.97708535e-02 -1.21187115e+00 3.53889376e-01 8.87307525e-01 -2.77527422e-01 -3.98500681e-01 2.83340275e-01 5.95456421e-01 -2.70638734e-01 -4.05109406e-01 3.90746802e-01 4.25561696e-01 -1.17515492e+00 7.48279989e-01 -1.61074296e-01 3.48397434e-01 -5.33151388e-01 -4.44636010e-02 -1.01988399e+00 -1.77770462e-02 -2.82793641e-02 3.62199426e-01 1.57423162e+00 6.55728519e-01 -5.30457795e-01 9.19341743e-01 8.98329675e-01 5.20249307e-02 -8.55250239e-01 -7.03253865e-01 -1.62730336e-01 -1.78253368e-01 -4.63952392e-01 5.54809153e-01 6.44939065e-01 -6.27353430e-01 2.73975939e-01 -1.48214683e-01 4.43321377e-01 6.57657444e-01 2.08993733e-01 3.96073759e-01 -1.24199545e+00 -2.47565970e-01 -4.01382148e-01 -2.67016202e-01 -9.85328555e-01 2.94399023e-01 -7.65458286e-01 5.14091253e-01 -1.62328219e+00 1.45774290e-01 -8.91834736e-01 -2.14173615e-01 7.72596359e-01 -2.95863390e-01 4.13933307e-01 -3.34651582e-02 4.02963869e-02 -7.15150356e-01 3.34555745e-01 9.09727275e-01 2.92615965e-02 -1.89005837e-01 3.37657258e-02 -7.15472341e-01 1.00801301e+00 5.37575841e-01 -7.40778744e-01 -4.38817769e-01 -6.45086229e-01 1.57687798e-01 6.01024739e-03 4.31442916e-01 -8.85484695e-01 1.90175682e-01 -6.49899021e-02 4.79434401e-01 -3.81179839e-01 2.51023442e-01 -8.89153719e-01 4.96849902e-02 1.56082407e-01 -3.73254448e-01 -3.74467522e-01 -8.63971189e-02 6.90711319e-01 -1.71531573e-01 -5.95346868e-01 1.04722059e+00 -1.74375504e-01 -6.61449194e-01 1.86513111e-01 1.53706009e-02 -1.72980234e-01 1.03493690e+00 -7.38726184e-02 1.44086955e-02 -1.03391968e-01 -9.81098890e-01 3.41299742e-01 5.94280183e-01 1.41017124e-01 2.68598318e-01 -9.64615524e-01 -3.80286902e-01 2.78235018e-01 -1.65844321e-01 4.00936753e-01 1.65890992e-01 8.02758157e-01 -4.51116502e-01 7.58828744e-02 1.16289802e-01 -7.47102618e-01 -1.07705462e+00 4.17396814e-01 3.56223494e-01 -3.24445188e-01 -4.31966603e-01 1.07023335e+00 4.03877437e-01 -4.11442280e-01 2.79801518e-01 -3.61001670e-01 -3.14023048e-01 5.20055071e-02 2.77675390e-01 1.44124776e-02 8.12800303e-02 -6.77524686e-01 -2.90512413e-01 8.18045139e-01 -8.35299268e-02 -3.40748042e-01 1.31694853e+00 -3.20151716e-01 -2.54538208e-01 2.74177104e-01 1.12274480e+00 -2.97774047e-01 -1.67700708e+00 -3.99460852e-01 2.26480961e-01 -3.44521821e-01 1.58409595e-01 -9.17283118e-01 -1.41897595e+00 9.68131244e-01 5.06582618e-01 -9.79014635e-02 1.27917254e+00 2.41358161e-01 7.71365106e-01 -4.81810085e-02 1.71649396e-01 -1.05607891e+00 -1.52123943e-01 1.40049055e-01 3.93216550e-01 -1.23833013e+00 -1.63152546e-01 -9.14271951e-01 -5.23750544e-01 9.79162931e-01 2.89194286e-01 1.65431853e-02 6.88593328e-01 2.26892903e-01 2.22017646e-01 -1.36859983e-01 -4.69929963e-01 -3.68476868e-01 2.69324332e-01 2.46632427e-01 1.66115105e-01 -7.33809024e-02 -1.13121830e-01 5.79314172e-01 1.76042318e-01 8.63069743e-02 3.14690806e-02 8.25908601e-01 -4.39967602e-01 -1.44454122e+00 -3.58315080e-01 3.35032463e-01 -4.82719690e-01 1.40034342e-02 -3.34380656e-01 3.16845328e-01 5.58203876e-01 1.01599336e+00 2.35994458e-02 -9.34736505e-02 1.29350036e-01 3.83191317e-01 4.90010738e-01 -7.39633203e-01 -2.64728934e-01 2.23914430e-01 -1.06717832e-01 -4.25802708e-01 -5.61964393e-01 -6.42054915e-01 -1.59752488e+00 1.65723369e-01 -3.88634384e-01 3.39968242e-02 1.07012713e+00 1.28284466e+00 3.36007029e-01 5.28496921e-01 3.91111344e-01 -9.62283134e-01 -2.50917852e-01 -8.86547804e-01 -4.70573872e-01 6.04151607e-01 4.33947779e-02 -7.87630916e-01 -2.51501054e-01 3.89677286e-01]	[9.548188209533691, 0.5539113879203796]
bf24e74f-351d-44f1-b375-104f5980c8c0	cutting-through-the-noise-an-empirical	2211.01704	null	https://arxiv.org/abs/2211.01704v2	https://arxiv.org/pdf/2211.01704v2.pdf	Cutting Through the Noise: An Empirical Comparison of Psychoacoustic and Envelope-based Features for Machinery Fault Detection	Acoustic-based fault detection has a high potential to monitor the health condition of mechanical parts. However, the background noise of an industrial environment may negatively influence the performance of fault detection. Limited attention has been paid to improving the robustness of fault detection against industrial environmental noise. Therefore, we present the Lenze production background-noise (LPBN) real-world dataset and an automated and noise-robust auditory inspection (ARAI) system for the end-of-line inspection of geared motors. An acoustic array is used to acquire data from motors with a minor fault, major fault, or which are healthy. A benchmark is provided to compare the psychoacoustic features with different types of envelope features based on expert knowledge of the gearbox. To the best of our knowledge, we are the first to apply time-varying psychoacoustic features for fault detection. We train a state-of-the-art one-class-classifier, on samples from healthy motors and separate the faulty ones for fault detection using a threshold. The best-performing approaches achieve an area under curve of 0.87 (logarithm envelope), 0.86 (time-varying psychoacoustics), and 0.91 (combination of both).	['Gregory Palmer', 'Zhao Ren', 'David Pelkmann', 'Yvonne Richter', 'Peter Wißbrock']	2022-11-03	null	null	null	null	['one-class-classifier', 'fault-detection']	['methodology', 'miscellaneous']	[ 2.45257050e-01 -1.47440061e-01 8.02564383e-01 -3.97937596e-02 -8.43640327e-01 -2.62222350e-01 9.40332636e-02 4.78943624e-02 -1.18775643e-01 2.52454370e-01 -5.63461304e-01 -1.43811852e-01 -6.03611887e-01 -5.52925348e-01 -5.16847730e-01 -9.47588861e-01 -1.92315474e-01 5.75622581e-02 7.48256505e-01 -2.12980911e-01 2.24816456e-01 4.22714531e-01 -2.17002082e+00 2.70583302e-01 5.69641709e-01 1.19302762e+00 6.05917454e-01 8.16034555e-01 5.83505869e-01 3.00971478e-01 -1.46854079e+00 1.90258712e-01 8.81748088e-03 -3.93638849e-01 -6.26111507e-01 4.50555831e-02 4.30797637e-02 -7.11719021e-02 -2.73492098e-01 9.81590688e-01 5.67431390e-01 5.08769900e-02 6.12821579e-01 -1.25735676e+00 -1.80363551e-01 3.80412638e-01 -3.94631140e-02 2.56351471e-01 4.26761121e-01 2.66781062e-01 5.85654974e-01 -5.58040738e-01 2.04880838e-03 1.06921971e+00 6.12841547e-01 1.76390454e-01 -8.52676272e-01 -5.06164849e-01 -3.44515681e-01 6.01971149e-01 -1.09406388e+00 -1.67982444e-01 8.61983657e-01 -4.57080752e-01 8.97941172e-01 2.22802475e-01 3.34267855e-01 9.92907524e-01 7.14939237e-01 2.66407639e-01 1.08850670e+00 -5.75408936e-01 2.74248272e-01 -1.87367037e-01 -5.38255600e-03 4.03410554e-01 1.32305786e-01 1.50390312e-01 -5.10629952e-01 -2.34851718e-01 3.60646307e-01 -4.40863580e-01 -4.61670280e-01 3.01665634e-01 -8.51006985e-01 4.31644589e-01 6.22866005e-02 6.49498403e-01 -5.24089754e-01 7.25930035e-02 5.22011280e-01 8.84603858e-01 2.81177521e-01 7.03896165e-01 -5.58791399e-01 -3.76474410e-01 -4.91947949e-01 6.42400384e-02 8.66392672e-01 2.45562673e-01 3.90320420e-01 5.42157948e-01 9.25176442e-02 1.07005930e+00 3.25976789e-01 5.78081012e-01 5.80607653e-01 -7.46874452e-01 -1.37932956e-01 4.61453162e-02 4.93546836e-02 -6.93020940e-01 -5.12645721e-01 -4.00395453e-01 -3.06956291e-01 5.14029920e-01 3.15190226e-01 -1.56442821e-01 -7.51909316e-01 1.06072271e+00 3.60974632e-02 2.90576726e-01 2.46816397e-01 7.86112845e-01 5.37536144e-01 4.93972659e-01 -4.81566370e-01 -2.08125681e-01 1.36595619e+00 -4.22767729e-01 -1.05978203e+00 -2.33403996e-01 3.49823028e-01 -1.11750710e+00 1.05797553e+00 1.27293336e+00 -6.56080067e-01 -8.31182361e-01 -1.47706676e+00 8.41875911e-01 -2.23959550e-01 3.01878154e-01 2.96301208e-02 9.63110685e-01 -6.63678825e-01 8.20306659e-01 -9.37402725e-01 -5.37603199e-02 -2.38824427e-01 1.75759599e-01 -4.68792409e-01 -1.00179590e-01 -1.29646373e+00 1.01439583e+00 -1.15826800e-01 4.23988968e-01 -1.15993536e+00 -4.93763834e-01 -7.65116990e-01 -4.18081343e-01 5.00373542e-01 7.61793107e-02 1.44240797e+00 -4.21140581e-01 -1.56903100e+00 3.02705407e-01 4.01819497e-01 -4.38457340e-01 3.12336266e-01 -5.76802671e-01 -9.35632825e-01 4.41644996e-01 2.04407964e-02 -3.50662470e-01 1.06807876e+00 -1.33841145e+00 -6.32336497e-01 -2.00283423e-01 -3.04336488e-01 -4.37584221e-01 -8.19180980e-02 3.25514436e-01 1.68981627e-01 -5.40050805e-01 1.24422662e-01 -8.00031602e-01 1.79023907e-01 -3.68189245e-01 -5.36279380e-01 -3.58424366e-01 1.12232900e+00 -9.12741423e-01 9.68592465e-01 -2.56277800e+00 -5.29977024e-01 2.54829288e-01 -4.43501800e-01 2.52418756e-01 3.09274755e-02 4.92303669e-01 -9.64953750e-02 -4.11625832e-01 -2.65473932e-01 2.67760158e-02 -3.16583028e-05 3.50944698e-01 8.88155848e-02 7.73299932e-01 6.41081631e-01 -5.52260764e-02 -7.98858225e-01 7.71612227e-02 2.34612688e-01 2.99642414e-01 -1.06535301e-01 4.40364510e-01 2.97814161e-01 7.33991265e-02 -1.60508975e-01 7.02163637e-01 4.20487493e-01 6.63857758e-01 -3.81699532e-01 -6.14914894e-01 -8.68337229e-02 1.56297997e-01 -1.48304880e+00 1.15917671e+00 -5.40686667e-01 7.44594276e-01 5.70648253e-01 -1.03939927e+00 1.36233687e+00 7.16774166e-01 4.81840104e-01 -3.82038206e-01 1.41352311e-01 6.36844397e-01 4.72264558e-01 -9.92658556e-01 2.71335572e-01 -6.20210096e-02 -6.69283569e-02 -9.94069353e-02 3.35701734e-01 -5.18216431e-01 2.28129759e-01 -4.34042782e-01 1.44628918e+00 -1.92423537e-01 -2.53462404e-01 -3.12814295e-01 3.77636909e-01 -3.37099046e-01 3.99773151e-01 5.59063375e-01 -1.62267774e-01 6.22340322e-01 4.66702670e-01 2.60910928e-01 -4.99437928e-01 -9.30777550e-01 -2.45453164e-01 7.63272703e-01 -1.12049259e-01 1.37580276e-01 -8.16798031e-01 -2.16153190e-01 2.16331735e-01 7.89646268e-01 -2.92572200e-01 -7.51629114e-01 -4.99720573e-01 -6.65995240e-01 8.35465193e-01 3.87409687e-01 5.98360114e-02 -1.05451477e+00 -7.67956018e-01 4.10543591e-01 1.16946623e-02 -1.15738523e+00 -1.67181483e-03 8.96208465e-01 -3.14068258e-01 -1.38176179e+00 -2.28785679e-01 -8.38083386e-01 2.57019758e-01 -6.62905425e-02 7.70276606e-01 2.70474050e-03 -7.66461551e-01 6.06753349e-01 -5.62439859e-01 -7.20299125e-01 -9.89974976e-01 -6.66713536e-01 3.49001437e-01 9.97985341e-03 -6.53333403e-03 -3.79996449e-01 -1.74997851e-01 8.33557725e-01 -9.31335866e-01 -1.12052095e+00 7.51812220e-01 8.90214443e-01 3.00087750e-01 1.04689384e+00 8.01867902e-01 -3.49007905e-01 7.58075535e-01 -3.90383095e-01 -4.19317931e-01 -1.05492234e-01 -3.41061920e-01 -2.79284745e-01 5.79431355e-01 -6.75041854e-01 -8.55544209e-01 -1.53754801e-01 -4.30713981e-01 -4.19690013e-01 -6.95191145e-01 4.91012484e-01 -4.63780522e-01 4.16939240e-03 6.90411210e-01 -1.60319939e-01 2.79840410e-01 -7.65880942e-01 -2.56134212e-01 1.07962835e+00 1.18775439e+00 -5.58060825e-01 7.28802979e-01 1.88158322e-02 5.92730269e-02 -1.27147710e+00 -3.20503801e-01 -6.27497494e-01 -2.87263125e-01 -5.38392961e-01 5.77770054e-01 -6.07464790e-01 -4.78285939e-01 1.15681648e+00 -9.33634341e-01 -3.25771868e-01 -2.72942066e-01 7.79602289e-01 -4.89826977e-01 6.60434484e-01 -7.44839430e-01 -1.20682132e+00 -2.92108566e-01 -1.22002208e+00 1.00011742e+00 -1.46790147e-01 9.83676128e-03 -3.93935919e-01 -2.64201790e-01 2.04437137e-01 2.16267660e-01 4.88011748e-01 6.31621838e-01 -7.49917567e-01 3.57780337e-01 -5.84141850e-01 5.45711100e-01 1.14673030e+00 4.30331469e-01 2.33016163e-01 -1.30770099e+00 -2.33630016e-01 5.59430599e-01 -2.94804335e-01 6.83695376e-01 2.70831198e-01 7.97666013e-01 3.17755550e-01 -1.19411005e-02 -1.96636945e-01 1.10068393e+00 5.15511751e-01 5.74850500e-01 3.00725341e-01 1.77502364e-01 6.90662503e-01 1.23946106e+00 4.93677467e-01 -3.96413833e-01 5.64190984e-01 7.35490859e-01 2.07862891e-02 -4.56591956e-02 3.29047978e-01 7.62290001e-01 8.39523196e-01 1.49526328e-01 -2.79133201e-01 -7.26749182e-01 6.41453922e-01 -1.11653471e+00 -4.79659587e-01 -6.03781343e-01 2.08866930e+00 7.63676882e-01 6.41101241e-01 2.44998317e-02 1.33057308e+00 7.81525075e-01 -5.09136498e-01 -1.96320385e-01 -4.17498529e-01 4.61846180e-02 3.13935518e-01 2.28083849e-01 1.41590372e-01 -1.18961692e+00 6.02238104e-02 5.67292404e+00 9.66513515e-01 -8.87179732e-01 -9.32758301e-02 5.55380667e-03 3.30307126e-01 4.13075507e-01 -4.53735709e-01 -3.96326274e-01 4.49442506e-01 1.28085017e+00 5.42069614e-01 1.91863015e-01 7.45330691e-01 2.25573391e-01 -4.21626538e-01 -9.15718675e-01 5.85619569e-01 6.28747186e-03 -2.41385490e-01 -7.53011525e-01 -1.64209232e-01 2.15289742e-01 -1.85486048e-01 -1.20572843e-01 -6.12488911e-02 -2.50000238e-01 -6.60529852e-01 9.48614836e-01 7.28711128e-01 5.04494548e-01 -8.98973167e-01 1.45431530e+00 1.56216532e-01 -1.15953445e+00 -3.48491877e-01 -2.76700765e-01 8.32516979e-03 2.14259669e-01 1.00893033e+00 -7.33622015e-01 8.58667791e-01 1.04231405e+00 1.84152842e-01 -3.77798885e-01 1.23238194e+00 -1.14720173e-01 1.07014954e+00 -3.91324401e-01 4.77864183e-02 -5.63344657e-02 2.53996998e-01 5.44162273e-01 1.17761111e+00 5.81580818e-01 -3.66307795e-01 2.04306260e-01 5.41431129e-01 7.49534726e-01 -1.77647755e-01 -4.16878819e-01 6.27004579e-02 5.17010808e-01 1.03217113e+00 -5.93326092e-01 2.06512287e-01 -3.80173057e-01 5.85060000e-01 -6.53736532e-01 3.32777277e-02 -6.03197157e-01 -8.63049746e-01 6.56080544e-01 4.81938850e-03 3.79972279e-01 -1.15618467e-01 1.57132953e-01 -4.76279594e-02 3.68351750e-02 -9.71503019e-01 1.39456481e-01 -8.43892157e-01 -1.58469129e+00 6.33913577e-01 1.62401706e-01 -1.42094564e+00 -1.56687930e-01 -1.11471808e+00 -8.60464811e-01 7.54793644e-01 -1.15957570e+00 -6.19908154e-01 -3.45816940e-01 2.34648451e-01 6.69121861e-01 -3.55692536e-01 7.68825710e-01 2.89994597e-01 -4.30759013e-01 4.36407954e-01 1.77210588e-02 4.21849824e-02 7.25236535e-01 -1.31593955e+00 1.22945063e-01 7.52207756e-01 -2.54907846e-01 1.06783941e-01 1.26698244e+00 -5.92839122e-01 -1.35111594e+00 -9.35419738e-01 1.93526849e-01 1.13267526e-02 9.85580087e-01 -3.82479653e-02 -1.28310633e+00 -2.07100138e-01 1.41842127e-01 3.42815071e-02 5.17000198e-01 -4.65363443e-01 1.83686301e-01 -2.00434551e-01 -1.31993270e+00 -2.21553668e-02 4.06268537e-01 -5.79207897e-01 -7.67741740e-01 2.03122765e-01 5.47019124e-01 -3.33849162e-01 -1.39674616e+00 6.97783172e-01 3.04461092e-01 -7.80162632e-01 8.34363639e-01 1.06066771e-01 6.73093125e-02 -5.58758855e-01 -3.62415873e-02 -1.55853474e+00 -2.00473726e-01 -5.33733726e-01 1.53484672e-01 1.33081400e+00 2.34810397e-01 -9.73120153e-01 4.68079112e-02 -2.79129237e-01 -8.57571840e-01 -6.38510704e-01 -1.09091008e+00 -1.16675413e+00 -2.97808528e-01 -8.31384420e-01 2.79890597e-01 4.18891728e-01 5.22457026e-02 -1.84325695e-01 7.76174990e-03 7.24416912e-01 2.77801454e-01 -3.72879267e-01 4.08499211e-01 -1.38682473e+00 -4.26359147e-01 -1.91993475e-01 -9.09634650e-01 -2.22624674e-01 -6.64665401e-02 -1.18359387e-01 8.29324245e-01 -1.20804369e+00 -5.82111835e-01 2.02476978e-04 -5.18158138e-01 4.68147397e-01 -2.63668269e-01 2.83258379e-01 -2.66041845e-01 -2.35666826e-01 1.23144992e-01 2.66255140e-01 9.29258108e-01 -1.10402271e-01 2.73957342e-01 4.16052908e-01 -8.93047359e-03 6.06758773e-01 9.98460591e-01 -5.24207354e-01 -2.84219325e-01 1.41696125e-01 -6.75669238e-02 7.18838274e-02 5.56619763e-01 -1.48748159e+00 -6.08575642e-02 3.50221664e-01 -1.73411723e-02 -7.41690934e-01 2.07235932e-01 -9.37320769e-01 1.28852606e-01 8.67714107e-01 1.02249393e-02 1.43213868e-01 3.10241878e-01 6.47580802e-01 -5.28816581e-01 -8.68671060e-01 6.36572242e-01 1.00910872e-01 -5.57471693e-01 -5.68982005e-01 -1.03680646e+00 -3.22694480e-01 6.42746985e-01 -1.58451632e-01 -2.88385808e-01 -5.51085509e-02 -5.44261873e-01 -2.37813249e-01 1.63990438e-01 5.01793623e-01 6.61991060e-01 -8.70027423e-01 -7.79041946e-01 5.27968884e-01 8.38401467e-02 -1.90613806e-01 1.74783021e-01 8.79637659e-01 -2.78479546e-01 9.61763114e-02 -1.30399778e-01 -7.40839601e-01 -1.49672592e+00 2.19564974e-01 5.04252732e-01 2.13457927e-01 -2.92783111e-01 8.44206870e-01 -3.37893546e-01 -1.04054235e-01 1.31083146e-01 -8.05665731e-01 -2.56762683e-01 -8.85845721e-02 2.99374104e-01 8.15897286e-01 8.90318692e-01 -4.54739958e-01 -3.14032465e-01 3.88748735e-01 4.93550569e-01 7.74917305e-02 1.10346663e+00 3.14872146e-01 -1.48508772e-01 8.84991646e-01 9.03976262e-01 -4.63643037e-02 -9.50626612e-01 1.81583032e-01 8.95504802e-02 -2.56615698e-01 2.13012099e-01 -8.57066274e-01 -9.64758694e-01 6.49928391e-01 1.07297993e+00 1.01750278e+00 1.37724423e+00 7.61496648e-03 5.20979524e-01 3.60819817e-01 3.63347799e-01 -1.20298553e+00 3.91069233e-01 2.69660622e-01 1.00658667e+00 -8.35055649e-01 -2.34524608e-01 -5.35083175e-01 -4.37517613e-01 1.29580832e+00 4.40618664e-01 -1.94235504e-01 8.14275086e-01 8.22876453e-01 2.95198888e-01 -1.64589360e-01 -6.43762112e-01 -1.90197140e-01 3.67861331e-01 9.49109972e-01 1.93015397e-01 -4.12866399e-02 -6.19367398e-02 8.08178782e-01 -3.80969763e-01 -4.46574330e-01 4.74832356e-01 1.18798554e+00 -7.78563380e-01 -7.29707420e-01 -9.19377267e-01 5.80630064e-01 -6.40559494e-01 4.06882018e-01 -1.06992945e-01 6.76086307e-01 2.65739679e-01 1.75610447e+00 3.13390829e-02 -5.84950030e-01 9.56055403e-01 1.28679529e-01 3.02910358e-01 -4.19632226e-01 -6.89834416e-01 4.26332355e-01 4.43937182e-01 -3.93967420e-01 -2.67748147e-01 -8.22113156e-01 -1.26475441e+00 3.79606515e-01 -1.07461917e+00 3.45926166e-01 8.86910796e-01 8.78016174e-01 -2.60919034e-01 1.20685351e+00 9.23875451e-01 -7.50921547e-01 -8.85909379e-01 -1.48595071e+00 -1.11096299e+00 2.63802171e-01 4.29147303e-01 -9.87533092e-01 -9.12340701e-01 -7.97217116e-02]	[6.756795883178711, 2.3490638732910156]
06d157f4-f5d2-4c6c-8e1f-696aeb7c4a50	dialogps-dialogue-path-sampling-in-continuous	2306.16770	null	https://arxiv.org/abs/2306.16770v1	https://arxiv.org/pdf/2306.16770v1.pdf	DialoGPS: Dialogue Path Sampling in Continuous Semantic Space for Data Augmentation in Multi-Turn Conversations	In open-domain dialogue generation tasks, contexts and responses in most datasets are one-to-one mapped, violating an important many-to-many characteristic: a context leads to various responses, and a response answers multiple contexts. Without such patterns, models poorly generalize and prefer responding safely. Many attempts have been made in either multi-turn settings from a one-to-many perspective or in a many-to-many perspective but limited to single-turn settings. The major challenge to many-to-many augment multi-turn dialogues is that discretely replacing each turn with semantic similarity breaks fragile context coherence. In this paper, we propose DialoGue Path Sampling (DialoGPS) method in continuous semantic space, the first many-to-many augmentation method for multi-turn dialogues. Specifically, we map a dialogue to our extended Brownian Bridge, a special Gaussian process. We sample latent variables to form coherent dialogue paths in the continuous space. A dialogue path corresponds to a new multi-turn dialogue and is used as augmented training data. We show the effect of DialoGPS with both automatic and human evaluation.	['Rui Yan', 'Ji Zhang', 'Xing Gao', 'Yuhan Chen', 'Jinpeng Li', 'Ang Lv']	2023-06-29	null	null	null	null	['dialogue-generation', 'semantic-textual-similarity', 'semantic-similarity', 'dialogue-generation']	['natural-language-processing', 'natural-language-processing', 'natural-language-processing', 'speech']	[ 3.40086073e-01 4.41314161e-01 1.89581718e-02 -7.57068336e-01 -1.11826563e+00 -7.97068655e-01 9.62708712e-01 -1.34454876e-01 -2.72580773e-01 1.11328161e+00 6.22727096e-01 -1.66502655e-01 7.88713153e-03 -8.55552077e-01 -1.82393059e-01 -5.84034204e-01 4.31868076e-01 1.13711143e+00 2.25150943e-01 -6.99768364e-01 2.05334485e-01 -1.26156449e-01 -9.08963084e-01 5.91532052e-01 9.27900851e-01 4.16200995e-01 2.49445096e-01 1.01608300e+00 -7.51213968e-01 4.59676951e-01 -8.66886079e-01 -5.86356759e-01 1.30492434e-01 -9.71348286e-01 -1.23196030e+00 -6.27488792e-02 -5.35728671e-02 -1.40761837e-01 7.38731846e-02 9.38041151e-01 4.54798400e-01 7.48101771e-01 7.78251171e-01 -1.39069605e+00 -6.34500802e-01 7.21018314e-01 -3.10217321e-01 -3.86941075e-01 8.43058348e-01 9.33044031e-02 1.06910598e+00 -8.54338884e-01 5.85109055e-01 2.04213166e+00 4.98163581e-01 9.31791127e-01 -1.30916166e+00 -2.86725938e-01 2.02874064e-01 -2.33759388e-01 -6.55053735e-01 -1.34526983e-01 7.21913815e-01 -3.08512598e-01 7.85667837e-01 3.87991786e-01 3.56873035e-01 1.44999170e+00 5.04673347e-02 7.61090040e-01 1.37853062e+00 -4.35269922e-01 2.71102339e-01 2.93656826e-01 3.18418533e-01 2.39326715e-01 -6.11372709e-01 -4.52392370e-01 -3.65100503e-01 -7.09793091e-01 6.09888911e-01 -8.35438967e-02 -4.21432704e-02 9.02447663e-03 -1.11393952e+00 1.26721764e+00 -1.33964241e-01 8.20436254e-02 -2.30152741e-01 -2.58541793e-01 4.54600424e-01 6.52363122e-01 4.64673758e-01 6.12454176e-01 -5.90587378e-01 -5.89531898e-01 -4.38389599e-01 9.45073605e-01 1.32257867e+00 8.98383081e-01 7.98298657e-01 -3.45756531e-01 -5.47838449e-01 1.31710720e+00 1.64865509e-01 3.67151797e-01 5.61882913e-01 -1.18098712e+00 6.02291524e-01 4.48060244e-01 3.71626049e-01 -7.43606865e-01 -4.13568109e-01 3.04695457e-01 -9.23278272e-01 -1.75530970e-01 6.76734746e-01 -3.97366673e-01 -2.40246311e-01 2.07923222e+00 7.21280277e-01 -3.48356128e-01 3.40161890e-01 8.12055290e-01 5.16076028e-01 7.77046025e-01 6.04457669e-02 -3.58782917e-01 1.47666657e+00 -1.17115641e+00 -9.86206234e-01 -5.63496314e-02 5.20194650e-01 -9.85365212e-01 1.83832002e+00 4.93938982e-01 -9.59334493e-01 -5.11261582e-01 -4.41578627e-01 -7.10028112e-02 -2.35207275e-01 -3.49534780e-01 3.15583378e-01 5.78223765e-01 -8.95481229e-01 5.25089443e-01 -2.19685376e-01 -5.02174795e-01 -3.47030491e-01 2.76368018e-02 -3.28851014e-01 1.09961011e-01 -1.66302192e+00 1.02025676e+00 7.09079057e-02 -2.79527545e-01 -3.90459269e-01 -3.14965725e-01 -6.97904587e-01 -9.63037908e-02 6.29239500e-01 -7.03655124e-01 1.80184019e+00 -5.58753908e-01 -2.13264108e+00 6.20567381e-01 -2.97082454e-01 -3.89737010e-01 7.57494628e-01 -2.88316011e-01 -2.08141536e-01 -2.46573500e-02 2.77261108e-01 6.40963674e-01 7.17428565e-01 -1.13902760e+00 -4.68886912e-01 -3.90631527e-01 2.94557005e-01 5.07231951e-01 8.56130500e-04 2.88330838e-02 7.62760714e-02 -6.41360760e-01 2.40041465e-02 -1.10941505e+00 -2.95096427e-01 -5.18928111e-01 -4.99983996e-01 -6.46630824e-01 7.53066659e-01 -4.73239034e-01 1.06734717e+00 -1.73849857e+00 2.02140480e-01 -3.21849406e-01 -4.76567633e-03 -5.21983625e-03 -1.75884888e-01 8.60195458e-01 2.18972296e-01 9.92966890e-02 -3.97260070e-01 -7.99675107e-01 3.29640001e-01 3.64676833e-01 -4.30332243e-01 -5.98564669e-02 7.72504956e-02 7.17243493e-01 -1.17554450e+00 -3.85047346e-01 2.44889688e-02 -5.78390248e-02 -6.58106029e-01 4.93616283e-01 -6.06596768e-01 6.71788990e-01 -4.79454219e-01 4.30314243e-02 4.38682050e-01 1.27764558e-03 1.11053936e-01 2.82536298e-01 1.71558753e-01 4.86473560e-01 -9.70317125e-01 1.70951295e+00 -7.88035452e-01 2.70482928e-01 2.45356131e-02 -6.68061435e-01 1.23397100e+00 5.08399785e-01 1.34880975e-01 -1.59628972e-01 -1.63835570e-01 1.01072751e-01 3.35545652e-02 -4.20247704e-01 1.01828170e+00 -5.59332728e-01 -7.08949745e-01 8.18857491e-01 6.66035339e-02 -6.07471943e-01 1.26597080e-02 3.50117981e-01 8.44815135e-01 1.60140842e-02 4.36586440e-01 -1.37895107e-01 4.78134364e-01 -8.75198543e-02 4.31184322e-01 8.25167775e-01 -2.50133783e-01 6.33077383e-01 8.97272885e-01 -2.21437812e-01 -9.87612188e-01 -1.01590514e+00 1.07897431e-01 1.54331458e+00 -2.81815510e-02 -2.96025306e-01 -9.28833008e-01 -8.92461598e-01 -2.25946516e-01 1.25106430e+00 -5.28454483e-01 -6.02478608e-02 -4.18386310e-01 -3.89478594e-01 7.17755616e-01 1.86715722e-01 5.16602874e-01 -1.30261123e+00 -2.68492609e-01 5.80888212e-01 -6.94242835e-01 -9.93037105e-01 -9.18596029e-01 -1.22278500e-02 -5.60200095e-01 -9.07869995e-01 -7.35070944e-01 -5.31173527e-01 1.66319385e-01 4.84921075e-02 1.02820361e+00 -5.62027872e-01 2.20429614e-01 4.22819883e-01 -6.37688279e-01 -2.48161077e-01 -1.03192878e+00 -1.05260447e-01 2.67453432e-01 -2.38274690e-02 5.21729350e-01 -3.62340957e-01 -3.27775657e-01 4.75432128e-01 -5.99502087e-01 7.41501227e-02 1.10975534e-01 1.27534926e+00 1.60876587e-01 -4.82096881e-01 1.07606852e+00 -1.16964316e+00 1.59713030e+00 -4.86736298e-01 5.36867157e-02 3.30353796e-01 -3.28688115e-01 1.70354009e-01 7.84049809e-01 -6.29621506e-01 -1.37656021e+00 -2.10928887e-01 -2.04832628e-01 -1.43565029e-01 -5.63958347e-01 1.80444613e-01 -3.51488233e-01 6.02268815e-01 7.98132718e-01 1.40520260e-01 1.68207258e-01 -2.85313576e-01 9.66471136e-01 9.17238772e-01 3.85111660e-01 -1.07000387e+00 2.82162547e-01 9.55283418e-02 -3.72970313e-01 -9.55480576e-01 -8.13581467e-01 -4.81955945e-01 -6.41834676e-01 -1.12426363e-01 1.04790831e+00 -3.30445796e-01 -5.91516256e-01 4.07612562e-01 -1.56845999e+00 -6.65861666e-01 -3.82647097e-01 2.28363574e-01 -8.37258697e-01 4.98669088e-01 -6.89594507e-01 -1.11529863e+00 -2.83259690e-01 -9.84838009e-01 9.84749556e-01 2.79181629e-01 -9.49968874e-01 -1.27238607e+00 3.51855874e-01 3.85359526e-01 3.97536486e-01 3.15948501e-02 8.97289038e-01 -1.25285125e+00 7.35209957e-02 -6.57005683e-02 2.09639862e-01 4.03989881e-01 3.44152212e-01 -3.53177369e-01 -8.17769349e-01 -2.12502349e-02 5.75075805e-01 -6.84899509e-01 3.73837024e-01 1.14566423e-01 7.20009565e-01 -5.98562598e-01 -3.76533493e-02 -1.72162220e-01 6.06046557e-01 4.11885768e-01 3.70075464e-01 -3.43080126e-02 3.38662624e-01 1.09514749e+00 8.47164333e-01 5.15322804e-01 5.76608419e-01 6.28266335e-01 -1.13653503e-01 1.50957361e-01 2.92262614e-01 -4.77583677e-01 2.36619905e-01 8.94039571e-01 4.75619048e-01 -2.47363299e-01 -6.83012187e-01 5.08993626e-01 -1.94978523e+00 -9.48712647e-01 -2.05842286e-01 2.08478451e+00 1.29974532e+00 1.61765754e-01 3.19798797e-01 -2.72817373e-01 8.64442647e-01 1.99714929e-01 -3.49474341e-01 -7.91633010e-01 -4.13145348e-02 -1.07197814e-01 -3.92895192e-01 9.48925197e-01 -6.60004556e-01 1.20506358e+00 5.71465254e+00 8.85166228e-01 -7.40809917e-01 2.56066978e-01 6.56871319e-01 1.92576692e-01 -4.98803258e-01 2.04810962e-01 -7.81691015e-01 3.86650801e-01 6.38796389e-01 -1.17055118e-01 4.34889436e-01 8.25537503e-01 2.31990129e-01 -2.77143061e-01 -1.37788248e+00 8.01163316e-01 -7.55152926e-02 -8.58348489e-01 8.31787661e-02 -1.34010568e-01 6.98082864e-01 -6.92419529e-01 -1.17069550e-01 6.55916214e-01 7.28276074e-01 -9.05128956e-01 3.16345543e-01 4.97607350e-01 5.81906080e-01 -5.64194739e-01 5.56043029e-01 7.38641679e-01 -8.57825696e-01 3.04861367e-01 -3.31120789e-01 -1.01597995e-01 4.85694557e-01 1.44921899e-01 -1.21141791e+00 4.46526378e-01 1.68769464e-01 -1.33638591e-01 8.00114870e-02 3.40519607e-01 -1.47034973e-01 3.45630199e-01 -9.19333398e-02 -4.75905538e-01 4.47908729e-01 -5.92565060e-01 7.33018517e-01 1.30532169e+00 2.10953116e-01 3.87686878e-01 6.37502491e-01 9.11270797e-01 8.87696147e-02 1.49721980e-01 -7.05698311e-01 1.91245109e-01 8.37490797e-01 1.14072168e+00 -4.29092407e-01 -5.75977623e-01 -2.68217772e-01 1.23346353e+00 1.23090789e-01 3.98926526e-01 -5.46433210e-01 -4.82744604e-01 6.23615384e-01 -1.27880380e-01 -3.31602573e-01 -4.61999513e-03 -2.69163728e-01 -8.76001418e-01 -1.61701173e-01 -1.12448084e+00 2.87871808e-01 -6.96188450e-01 -1.82194555e+00 6.34911716e-01 1.26129940e-01 -7.36003399e-01 -7.48012781e-01 -2.00988680e-01 -7.29836643e-01 1.18832159e+00 -7.72068441e-01 -1.01381361e+00 -6.51773065e-02 6.68911815e-01 1.25787628e+00 -3.21778774e-01 1.26525557e+00 -3.38874966e-01 5.21954149e-03 4.86895412e-01 -8.14919248e-02 6.40816540e-02 1.10483384e+00 -1.59989095e+00 5.76754689e-01 1.58183083e-01 -1.09888464e-01 8.26945543e-01 8.62361610e-01 -5.73405981e-01 -8.30692232e-01 -7.54429758e-01 9.67657983e-01 -7.15797067e-01 7.07007051e-01 -4.80761915e-01 -1.19970274e+00 4.75585878e-01 5.19726157e-01 -6.90778911e-01 7.67538548e-01 3.63294154e-01 3.16126496e-02 4.25586522e-01 -1.24528289e+00 8.56705248e-01 6.76850259e-01 -7.65851855e-01 -1.11802888e+00 3.47015321e-01 1.02857530e+00 -5.15426219e-01 -5.81553102e-01 -4.26101685e-03 2.72866845e-01 -8.63175511e-01 5.75387180e-01 -9.35845733e-01 5.27775168e-01 1.02331907e-01 -3.95519793e-01 -1.86105716e+00 9.79558751e-02 -1.28758943e+00 5.99626862e-02 1.34917104e+00 3.04667950e-01 -7.38530576e-01 5.53752959e-01 7.83380032e-01 -1.77275121e-01 -8.16367745e-01 -9.56532001e-01 -6.01661026e-01 6.35645509e-01 -2.75267392e-01 7.51409054e-01 1.02516663e+00 3.21844846e-01 8.51818800e-01 -5.90338051e-01 -2.67902046e-01 2.23801717e-01 6.14627749e-02 1.23617923e+00 -1.06213450e+00 -6.07893050e-01 -4.01924968e-01 3.00867945e-01 -1.62346923e+00 2.61240065e-01 -3.41203868e-01 4.15427327e-01 -1.45130205e+00 2.08178665e-02 -4.81643081e-01 3.95615220e-01 -8.87867343e-03 -5.18624723e-01 -3.96121323e-01 1.97891906e-01 6.44514561e-02 -5.61485291e-01 9.16428030e-01 1.41527641e+00 9.53897312e-02 -5.43685973e-01 3.51252645e-01 -5.73440969e-01 7.36600220e-01 8.79346073e-01 -3.61590147e-01 -7.09264576e-01 1.25211226e-02 -2.77960319e-02 8.51167262e-01 1.87308133e-01 -4.38741982e-01 2.85327196e-01 -4.41968262e-01 -2.89308667e-01 -5.78143358e-01 7.78393388e-01 -2.46234134e-01 -2.36348107e-01 -1.08837940e-01 -1.00181985e+00 -1.71555411e-02 -1.98629975e-01 8.67564261e-01 -3.30930501e-01 -3.66581768e-01 6.33532763e-01 -4.68314201e-01 -3.09740126e-01 -8.38866606e-02 -3.80295455e-01 5.86822212e-01 9.26414311e-01 -8.76277685e-02 -2.78334498e-01 -8.74182761e-01 -1.01431215e+00 4.88176793e-01 2.56380141e-01 6.66252971e-01 4.28452522e-01 -1.27305162e+00 -7.08043873e-01 2.07757093e-02 -3.55951935e-02 1.35677531e-01 3.59599829e-01 5.21274626e-01 7.67602539e-03 3.68267834e-01 2.83335000e-02 -6.44489706e-01 -1.31960821e+00 2.03546181e-01 3.47047985e-01 -4.52840477e-01 -4.66858417e-01 9.89596486e-01 4.66889024e-01 -1.25054407e+00 1.62215009e-01 -2.62485206e-01 -1.67383343e-01 1.75402299e-01 4.15487587e-01 3.23531330e-01 -3.05099219e-01 -3.89353126e-01 1.76593855e-01 2.27971166e-01 -1.17490366e-01 -8.24480712e-01 7.51595140e-01 -4.44814235e-01 -1.45543277e-01 1.10198140e+00 9.11925614e-01 -3.64706740e-02 -1.26972651e+00 -4.83472288e-01 -3.30498000e-03 -5.71998537e-01 -8.60552490e-01 -8.33216012e-01 2.22206917e-02 1.02817595e+00 -3.00556663e-02 8.18287253e-01 4.84750330e-01 -2.80867834e-02 9.26245213e-01 6.10996902e-01 5.01601815e-01 -1.40151668e+00 6.38053060e-01 1.02338421e+00 1.22223198e+00 -1.29407930e+00 -6.70182109e-01 -2.10078895e-01 -1.46615827e+00 1.08561993e+00 8.25190425e-01 1.14379749e-01 4.43594128e-01 7.77348652e-02 3.73330593e-01 8.69143680e-02 -1.06909990e+00 2.95292348e-01 -1.04872756e-01 7.61230171e-01 6.08102500e-01 3.65175575e-01 -1.81228206e-01 8.91963542e-01 -5.83793283e-01 -4.03972119e-01 6.59767449e-01 6.07097805e-01 -4.82331425e-01 -1.46400249e+00 -5.26656866e-01 2.51953274e-01 -6.14301711e-02 3.31273153e-02 -7.53303051e-01 6.19982719e-01 -3.75907898e-01 1.40958190e+00 -4.92699631e-02 -3.42223167e-01 3.83324534e-01 8.22001100e-01 1.40661657e-01 -9.66809750e-01 -7.38514125e-01 8.39812588e-03 3.68490577e-01 -1.97355226e-01 -5.98397292e-02 -5.94871998e-01 -1.35566545e+00 -3.83511782e-01 -3.71030122e-01 5.49087703e-01 5.04293442e-01 1.04331839e+00 1.27815157e-01 4.87687379e-01 8.01535249e-01 -5.61957121e-01 -1.40257037e+00 -1.51117897e+00 -5.74207783e-01 6.78321838e-01 2.65816957e-01 -5.54689765e-01 -2.57868767e-01 -1.86745226e-01]	[12.800802230834961, 8.141160011291504]
33129f9e-707b-4ec9-9342-491f6ac0b4b1	logical-reasoning-over-natural-language-as	2303.12023	null	https://arxiv.org/abs/2303.12023v1	https://arxiv.org/pdf/2303.12023v1.pdf	Logical Reasoning over Natural Language as Knowledge Representation: A Survey	Logical reasoning is central to human cognition and intelligence. Past research of logical reasoning within AI uses formal language as knowledge representation~(and symbolic reasoners). However, reasoning with formal language has proved challenging~(e.g., brittleness and knowledge-acquisition bottleneck). This paper provides a comprehensive overview on a new paradigm of logical reasoning, which uses natural language as knowledge representation~(and pretrained language models as reasoners), including philosophical definition and categorization of logical reasoning, advantages of the new paradigm, benchmarks and methods, challenges of the new paradigm, desirable tasks & methods in the future, and relation to related NLP fields. This new paradigm is promising since it not only alleviates many challenges of formal representation but also has advantages over end-to-end neural methods.	['Erik Cambria', 'Jinjie Ni', 'Rui Mao', 'Xinya Du', 'Zonglin Yang']	2023-03-21	null	null	null	null	['logical-reasoning']	['reasoning']	[-3.42956521e-02 5.59156179e-01 -1.15210429e-01 -5.60128093e-01 -1.89164713e-01 -7.02925026e-01 5.80400825e-01 -1.62902087e-01 -3.58260363e-01 9.91606176e-01 1.04881734e-01 -7.59516537e-01 -4.76687551e-01 -1.04077315e+00 -7.26416528e-01 -1.23920932e-01 5.04580326e-02 8.32197309e-01 3.07678767e-02 -4.52453792e-01 4.85778064e-01 4.94914830e-01 -1.34399974e+00 5.84453225e-01 9.24718142e-01 1.10981107e+00 -1.64182916e-01 4.00777757e-01 -1.00560093e+00 1.92525637e+00 -6.64434254e-01 -6.52157545e-01 -2.24654540e-01 -2.86722213e-01 -1.82286501e+00 -6.13456190e-01 4.60737437e-01 -6.17646836e-02 -3.77667874e-01 1.13330150e+00 -8.53567868e-02 2.08126113e-01 6.30054593e-01 -1.32293272e+00 -1.42985988e+00 1.08844507e+00 3.77271138e-02 2.19627574e-01 9.59325910e-01 1.16549104e-01 1.16309428e+00 -3.80043775e-01 6.70278370e-01 1.75844657e+00 9.48491573e-01 9.29857612e-01 -1.10316396e+00 -6.12700582e-01 2.39514261e-01 7.34588683e-01 -1.52825963e+00 -1.73111141e-01 6.75612867e-01 -3.63911241e-01 1.60506129e+00 3.03005967e-02 7.46233165e-01 7.41060138e-01 3.03539753e-01 1.11657894e+00 1.07997191e+00 -6.65132403e-01 1.67870298e-01 1.80537581e-01 7.93065190e-01 1.09142244e+00 3.53980333e-01 -1.59010172e-01 -6.21357799e-01 7.55030289e-02 6.77021146e-01 -3.87710065e-01 -7.43829086e-02 -8.33806545e-02 -1.29346967e+00 7.48984933e-01 7.45576918e-01 4.49834943e-01 -3.52159470e-01 6.80872977e-01 6.14122629e-01 5.21505654e-01 -2.61915207e-01 9.03483272e-01 -7.22332895e-01 -1.46857187e-01 -7.76538432e-01 6.67903304e-01 1.13504589e+00 8.78171206e-01 4.79436249e-01 3.38230520e-01 -2.82042827e-02 4.76039648e-01 3.94150287e-01 6.75347567e-01 4.41974252e-01 -1.13274813e+00 4.33062226e-01 9.23444569e-01 -1.31430887e-02 -9.01914239e-01 -5.57197571e-01 1.59029171e-01 -6.51655197e-01 1.09093867e-01 2.85558999e-01 -9.86121744e-02 -5.79743207e-01 1.80952775e+00 -2.94736683e-01 -3.36046554e-02 5.96823871e-01 4.87680405e-01 1.27525723e+00 6.78197682e-01 2.81578779e-01 3.05656735e-02 1.42030668e+00 -6.47678435e-01 -9.97417331e-01 -3.69065285e-01 5.53622544e-01 -6.74944744e-02 1.09309280e+00 6.81033015e-01 -1.42538393e+00 -3.49218190e-01 -8.66465628e-01 -6.27418876e-01 -9.70461249e-01 -1.65545478e-01 1.45018256e+00 2.45370045e-01 -1.35626197e+00 2.00617164e-01 -5.12468517e-01 -3.93658817e-01 6.08108401e-01 4.44496423e-01 -2.38090634e-01 -3.32595557e-01 -1.66866422e+00 1.46149969e+00 1.13669300e+00 2.44142726e-01 -3.92695069e-01 -5.70876777e-01 -1.17111075e+00 1.94609046e-01 6.02621555e-01 -1.01568639e+00 1.47698259e+00 -8.94487619e-01 -1.58989048e+00 8.13690960e-01 -1.86909541e-01 -7.97698975e-01 2.64978617e-01 -4.77967530e-01 -4.41668272e-01 1.73752636e-01 3.32733430e-02 7.54055917e-01 2.40403354e-01 -8.89930606e-01 -2.73694605e-01 3.13813865e-01 4.71764088e-01 -2.59007905e-02 2.92234808e-01 8.92271940e-03 3.17367092e-02 -1.82543501e-01 3.22064996e-01 -5.29183090e-01 1.03627615e-01 -7.03961104e-02 -2.95713961e-01 -9.61459994e-01 1.72089309e-01 -3.74418437e-01 1.21207142e+00 -1.87511790e+00 1.86083034e-01 2.08265334e-01 3.04618776e-01 2.61417240e-01 5.25332689e-02 2.71629214e-01 -1.19130641e-01 4.03479874e-01 -3.32186595e-02 6.06143355e-01 6.20090723e-01 5.56574523e-01 -7.83904076e-01 -1.73094541e-01 4.42381769e-01 1.59609044e+00 -9.15286124e-01 -6.73743665e-01 9.22404900e-02 1.44310310e-01 -5.89794517e-01 -2.76382416e-02 -7.65264511e-01 -2.87164062e-01 -5.72924316e-01 7.72865117e-01 4.97824967e-01 -4.09605801e-01 5.10473549e-01 -1.04853466e-01 3.26340161e-02 7.38619149e-01 -1.27097082e+00 1.52437627e+00 -2.56215185e-01 6.75066471e-01 -3.17393184e-01 -1.16401291e+00 7.48295069e-01 5.15069366e-01 -1.70352250e-01 -5.78292191e-01 1.57907009e-01 1.15729153e-01 1.34712411e-02 -7.50182569e-01 1.67702883e-01 -6.99831784e-01 -2.08627373e-01 4.09718364e-01 7.89251551e-02 -5.98139763e-01 3.47832829e-01 1.47231534e-01 1.07336426e+00 4.61388946e-01 6.74587131e-01 -3.98553789e-01 7.92381167e-01 6.33988857e-01 4.05969501e-01 9.38457549e-01 -2.31167763e-01 -2.86913306e-01 6.67942822e-01 -9.69673634e-01 -3.05681854e-01 -1.22635317e+00 5.11645228e-02 9.36594486e-01 -2.89280973e-02 -2.65906274e-01 -3.73306274e-01 -4.07758445e-01 2.25086540e-01 1.22452188e+00 -4.79139715e-01 -2.31251493e-01 -5.54110467e-01 -1.76031724e-01 1.26372099e+00 7.54695535e-01 8.39450061e-01 -1.81524587e+00 -6.81977510e-01 1.51699379e-01 -2.90200561e-01 -1.18489206e+00 8.16388369e-01 3.24846894e-01 -8.03505898e-01 -1.16856802e+00 2.96006531e-01 -8.26574266e-01 3.56778920e-01 -1.21873751e-01 1.67770720e+00 3.00703317e-01 -6.67901933e-02 4.82087493e-01 -1.76831216e-01 -5.47839105e-01 -9.36213881e-02 -2.65685618e-01 1.36493608e-01 -8.53691578e-01 7.82285213e-01 -1.89988196e-01 1.23090215e-01 -3.50148320e-01 -8.39983165e-01 1.43758237e-01 6.32262707e-01 7.65089273e-01 1.30805433e-01 7.12114692e-01 4.43412930e-01 -9.49871302e-01 1.07663703e+00 -3.68697643e-01 -3.85721117e-01 7.44932532e-01 -4.43569988e-01 5.26572824e-01 7.49153018e-01 -1.90948416e-02 -1.45894480e+00 -6.88171566e-01 1.29174337e-01 -7.18402117e-03 -2.02200860e-01 1.08880317e+00 1.36636034e-01 -9.85213369e-03 8.61735582e-01 2.49991611e-01 -1.32138953e-01 -1.38835251e-01 5.99200666e-01 1.40395299e-01 4.85131353e-01 -1.39692402e+00 9.27516580e-01 6.21025898e-02 1.98489174e-01 -7.03225195e-01 -1.19403732e+00 8.15979689e-02 -6.96801305e-01 2.01027408e-01 8.46201599e-01 -7.47208297e-01 -1.36686420e+00 4.17830311e-02 -1.37988961e+00 -3.04850519e-01 -3.77592862e-01 4.22560334e-01 -6.34726942e-01 1.54300913e-01 -8.76944840e-01 -9.01342332e-01 -4.45399731e-01 -8.56462181e-01 3.44113678e-01 2.74845302e-01 -6.80666625e-01 -1.28771198e+00 -3.64424109e-01 5.69307089e-01 4.22924995e-01 1.52481169e-01 1.70321167e+00 -6.04207754e-01 -6.98666990e-01 -1.82032928e-01 -6.37589991e-01 5.44361472e-01 -4.63546097e-01 -9.61348116e-02 -6.56606913e-01 4.75954980e-01 -2.64080614e-01 -1.16774225e+00 4.55632210e-01 3.08633178e-01 1.04319179e+00 -2.64310747e-01 1.76292695e-02 2.68458039e-01 1.35891306e+00 3.90683115e-01 7.73055911e-01 4.18260306e-01 5.09337425e-01 4.98010248e-01 3.09740216e-01 -3.72997709e-02 8.59215438e-01 -1.02230422e-01 -2.90131897e-01 4.47317421e-01 4.08243947e-02 5.62458672e-02 5.89061260e-01 7.37172425e-01 -3.91375422e-01 1.41599625e-01 -1.60552263e+00 2.11513489e-01 -1.95048010e+00 -1.49153161e+00 2.08802134e-01 1.38731158e+00 1.30128360e+00 4.14728522e-01 -3.24073642e-01 1.71483457e-01 2.96310410e-02 -1.14923559e-01 -5.33721566e-01 -1.11983240e+00 -1.89297244e-01 6.81335151e-01 -2.50449687e-01 6.15604281e-01 -7.64130890e-01 1.54329765e+00 7.51222038e+00 3.36777180e-01 -8.31801593e-01 -3.80302846e-01 -9.47709288e-03 1.62114739e-01 -3.31705481e-01 -1.13477722e-01 -7.32114792e-01 -3.65938246e-01 9.14807737e-01 -1.49375051e-01 1.03704929e+00 9.26688313e-01 -4.43777114e-01 -2.18426526e-01 -1.71473241e+00 9.26020324e-01 1.88451603e-01 -1.79084229e+00 5.75319648e-01 -5.12714505e-01 2.39896908e-01 -6.26891553e-02 -4.17623162e-01 1.06135976e+00 5.22596776e-01 -1.39758015e+00 9.31627989e-01 7.93087244e-01 5.26389956e-01 -6.69942439e-01 9.65445995e-01 3.23892683e-01 -1.07125020e+00 -1.37968481e-01 -3.18276584e-01 -7.79089391e-01 -2.01348722e-01 2.00886577e-01 -7.84883976e-01 5.64603567e-01 6.53016806e-01 7.33984470e-01 -6.90181851e-01 2.32087120e-01 -5.92854679e-01 2.89267778e-01 -2.48070091e-01 -6.10918403e-01 4.20445770e-01 5.47343120e-02 -1.04676954e-01 1.28034818e+00 -2.47994199e-01 4.10437584e-01 3.03242743e-01 1.36277747e+00 1.40864015e-01 -4.46678340e-01 -6.96964800e-01 -6.08296692e-01 2.23304838e-01 4.76741940e-01 -4.14458156e-01 -6.49617612e-01 -5.62337935e-01 4.33930337e-01 7.20669329e-01 5.54420114e-01 -7.91479766e-01 -5.11526704e-01 5.51426709e-01 -3.36991876e-01 -6.34666458e-02 -9.20780748e-02 -6.36792004e-01 -1.29355896e+00 -1.06740691e-01 -1.04350674e+00 4.47164506e-01 -1.19293547e+00 -1.60630643e+00 3.88069093e-01 4.65152562e-01 -4.93127316e-01 -4.75789338e-01 -1.35253394e+00 -3.22465181e-01 9.11780655e-01 -1.73566484e+00 -1.14515483e+00 2.13519856e-02 7.03136563e-01 2.43232489e-01 -1.55153871e-01 1.22087419e+00 -1.25084728e-01 -1.81813881e-01 4.05872017e-01 -5.20887077e-01 4.24824178e-01 6.93553090e-02 -1.39674127e+00 3.28997262e-02 4.52316105e-01 -2.27865338e-01 1.39629602e+00 4.78352249e-01 -4.87812936e-01 -2.08945560e+00 -4.29771930e-01 1.17358589e+00 -8.09758484e-01 1.02088058e+00 4.23178189e-02 -9.11367834e-01 1.13069451e+00 1.32983103e-01 -2.41907001e-01 7.58290648e-01 6.92844927e-01 -8.03929925e-01 -5.82457222e-02 -1.15835536e+00 8.02206278e-01 9.16406155e-01 -9.28435028e-01 -1.67919016e+00 3.84976804e-01 6.61571681e-01 -1.81987435e-01 -7.55060792e-01 2.27355838e-01 7.75169134e-01 -6.91074729e-01 1.17059767e+00 -7.91393995e-01 6.66384101e-01 -5.70686519e-01 -1.98610514e-01 -7.62222946e-01 -4.99210209e-01 -2.75418490e-01 -3.02679658e-01 8.43965888e-01 3.56954545e-01 -6.87119067e-01 3.18181604e-01 1.03842413e+00 -5.74706495e-03 -6.23793781e-01 -8.51129055e-01 -6.15307391e-01 4.39002037e-01 -8.58768582e-01 4.90824461e-01 1.09925807e+00 7.69232154e-01 7.67304778e-01 1.30102292e-01 4.29388843e-02 3.83596331e-01 4.50385869e-01 4.75380599e-01 -1.47881627e+00 -4.36604768e-03 -8.69365335e-01 -4.48420763e-01 -8.23564112e-01 8.96037042e-01 -1.21748424e+00 6.36368468e-02 -2.01819491e+00 -1.44366920e-01 -1.09504864e-01 -5.52870035e-01 9.94108438e-01 1.39465004e-01 -3.15955728e-01 1.43154591e-01 -2.19663829e-01 -8.40224743e-01 -1.01920369e-03 1.31172955e+00 -3.55026960e-01 1.01224661e-01 -6.70690954e-01 -8.96218300e-01 1.12415564e+00 7.77443945e-01 1.62653010e-02 -7.41570532e-01 -6.02636456e-01 9.30454195e-01 3.44737843e-02 5.30837834e-01 -1.16865540e+00 2.97567248e-01 -6.74136996e-01 5.13505876e-01 -4.70504194e-01 1.48359507e-01 -8.78976882e-01 -1.14049755e-01 6.28759742e-01 -5.19424617e-01 2.31183693e-01 5.21635830e-01 8.00168812e-02 -4.02669370e-01 -4.35003221e-01 5.33184111e-01 -6.50411546e-01 -1.22002923e+00 -5.70605457e-01 -3.92111659e-01 4.40108299e-01 6.21514618e-01 -1.80243984e-01 -5.36561131e-01 9.53143537e-02 -3.90496433e-01 3.68599445e-01 -1.20986916e-01 9.03071836e-02 7.66408086e-01 -1.05856514e+00 -1.67962089e-01 -9.66238510e-03 9.57228392e-02 2.12948263e-01 -1.89526990e-01 5.76349258e-01 -1.15459061e+00 9.39652145e-01 -4.14281666e-01 1.63183920e-02 -6.57405496e-01 5.94044447e-01 2.87047118e-01 -2.44517520e-01 -7.26120174e-01 1.10015488e+00 -6.56479597e-02 -7.19676673e-01 5.20329297e-01 -1.10266900e+00 -3.13166767e-01 -2.78164715e-01 6.56165183e-01 2.24515591e-02 -3.43300253e-01 -1.98651865e-01 -8.16656590e-01 4.22495842e-01 -7.87564069e-02 -1.92313623e-02 1.21916759e+00 5.07527173e-01 -7.25555301e-01 8.85358095e-01 5.09983957e-01 -5.35286188e-01 -1.83034375e-01 -5.14935791e-01 3.51069033e-01 4.36678715e-02 -3.98468859e-02 -1.41379821e+00 -3.92559022e-01 8.55584621e-01 -9.29693058e-02 2.58277655e-01 7.16413140e-01 3.65718752e-02 5.51219523e-01 1.56908953e+00 4.84170705e-01 -1.00652397e+00 -1.93072438e-01 1.34581316e+00 9.19380724e-01 -1.04212165e+00 4.74026382e-01 -2.28354603e-01 -7.71580756e-01 1.25235307e+00 7.52245009e-01 -4.29815240e-02 6.57012463e-01 1.12699531e-01 1.23749956e-01 -7.35639036e-01 -8.93724442e-01 -1.03445062e-02 2.33611867e-01 5.06080091e-01 8.22414517e-01 2.16511145e-01 -3.60183865e-02 5.92215002e-01 -6.21648431e-01 6.75093770e-01 -4.83337343e-02 1.34052622e+00 -3.18382591e-01 -6.44039094e-01 -3.90645266e-01 4.42569107e-01 -1.32450685e-01 -4.85557824e-01 -5.10082185e-01 9.73670304e-01 3.32850218e-01 8.67565274e-01 -2.38552004e-01 -3.59415449e-02 3.71117473e-01 8.16471040e-01 7.73566723e-01 -6.33681655e-01 -5.42574525e-01 -8.37531388e-01 3.63511443e-01 -5.50706446e-01 -5.68766177e-01 -3.28838736e-01 -1.99397600e+00 -8.01858902e-01 1.76635552e-02 3.51649970e-01 1.96940318e-01 1.44556141e+00 -4.16711457e-02 8.29424083e-01 -4.08268780e-01 -4.33946550e-01 -5.65216303e-01 -6.76344693e-01 -3.47640127e-01 3.06562066e-01 1.93307877e-01 -7.32038736e-01 -1.07853018e-01 3.01424921e-01]	[9.232184410095215, 7.175995349884033]
88da34af-a446-47d2-9cc9-808a9107b897	multi-module-based-cvae-to-predict-hvcm	2304.10639	null	https://arxiv.org/abs/2304.10639v1	https://arxiv.org/pdf/2304.10639v1.pdf	Multi-module based CVAE to predict HVCM faults in the SNS accelerator	We present a multi-module framework based on Conditional Variational Autoencoder (CVAE) to detect anomalies in the power signals coming from multiple High Voltage Converter Modulators (HVCMs). We condition the model with the specific modulator type to capture different representations of the normal waveforms and to improve the sensitivity of the model to identify a specific type of fault when we have limited samples for a given module type. We studied several neural network (NN) architectures for our CVAE model and evaluated the model performance by looking at their loss landscape for stability and generalization. Our results for the Spallation Neutron Source (SNS) experimental data show that the trained model generalizes well to detecting multiple fault types for several HVCM module types. The results of this study can be used to improve the HVCM reliability and overall SNS uptime	['Sarah Cousineau', 'Pradeep Ramuhalli', 'Dan Lu', 'Majdi I. Radaideh', 'Chris Pappas', 'Lasitha Vidyaratne', 'Steven Goldenberg', 'Kishansingh Rajput', 'Malachi Schram', 'Yasir Alanazi']	2023-04-20	null	null	null	null	['type']	['speech']	[-4.67915146e-04 -7.06004277e-02 1.71461344e-01 -1.30560637e-01 -6.23622656e-01 -1.89829186e-01 5.55383265e-01 4.13832515e-02 8.34815726e-02 7.48291731e-01 -1.78269416e-01 -2.15528920e-01 -2.92915761e-01 -8.76937151e-01 -6.97734833e-01 -1.03155863e+00 -1.91676632e-01 7.43784964e-01 3.17400128e-01 -2.84396529e-01 -1.14123389e-01 6.85827017e-01 -1.63654506e+00 6.05678141e-01 6.02264583e-01 1.05797040e+00 3.19820970e-01 4.78418827e-01 4.96626794e-01 4.93407696e-01 -1.21759462e+00 1.48319915e-01 -1.45378709e-01 -5.22553504e-01 -6.56417966e-01 -2.33595952e-01 5.38050989e-03 1.78160980e-01 -3.13195169e-01 1.22921646e+00 5.11179388e-01 2.38765329e-02 1.15284824e+00 -1.22131574e+00 -7.10357800e-02 7.09686816e-01 1.52766228e-01 5.66712081e-01 1.76342219e-01 1.92859009e-01 6.84427083e-01 -4.89177465e-01 5.34660637e-01 8.69523764e-01 7.33646274e-01 4.91332352e-01 -1.23242879e+00 -5.10829926e-01 -4.88548011e-01 8.08560848e-01 -1.39380765e+00 -1.19340710e-01 7.28893161e-01 -5.43556869e-01 1.35325480e+00 3.21974516e-01 5.62309742e-01 1.36921787e+00 7.85197914e-01 2.35787168e-01 9.35587108e-01 -2.00829387e-01 5.95959365e-01 2.88220197e-01 1.01849087e-01 2.97092080e-01 2.49637470e-01 2.08093867e-01 -1.67331770e-01 -8.79621878e-02 5.30274570e-01 -5.27395368e-01 -6.67481303e-01 -7.77718276e-02 -4.36864644e-01 1.16588986e+00 4.23754394e-01 1.01411510e+00 -3.73904854e-01 -5.00509702e-02 5.13508141e-01 4.93516296e-01 4.02962297e-01 6.91304088e-01 -5.00605941e-01 2.38128174e-02 -1.07307506e+00 6.88225999e-02 9.24985409e-01 3.82209748e-01 5.22228003e-01 6.95333242e-01 -2.54894495e-01 6.28972948e-01 1.45029545e-01 4.97705400e-01 5.80663860e-01 -7.70752251e-01 7.81283379e-02 3.21998596e-01 -4.06788230e-01 -3.80428076e-01 -6.32516682e-01 -6.72143936e-01 -7.06482649e-01 5.02120435e-01 -1.89465091e-01 -3.62175941e-01 -9.08104777e-01 1.25693285e+00 -2.03390032e-01 1.71263844e-01 2.09755719e-01 8.26286197e-01 8.03367496e-01 1.03314507e+00 -1.34322450e-01 -3.78129840e-01 1.08828235e+00 -2.31050968e-01 -8.77813041e-01 -2.47858949e-02 3.69444758e-01 -1.53354645e-01 3.09104383e-01 5.48217893e-01 -1.10057652e+00 -5.64380944e-01 -1.76982903e+00 5.96240282e-01 -3.57912511e-01 -1.18833378e-01 1.12125672e-01 7.14579821e-01 -7.38477111e-01 1.48739922e+00 -8.71120512e-01 -8.54277462e-02 2.79224992e-01 4.88487154e-01 -5.76873533e-02 2.79365182e-01 -1.59750366e+00 1.32419491e+00 3.77699196e-01 6.41058013e-02 -1.38313198e+00 -7.04390287e-01 -7.80160904e-01 2.32749209e-01 -1.94573224e-01 -5.56610107e-01 1.15649092e+00 -6.82716012e-01 -1.38903630e+00 9.63736847e-02 9.74313319e-02 -8.32974911e-01 1.44886777e-01 4.02614087e-01 -9.05255616e-01 4.07008886e-01 -4.62426752e-01 -9.57052335e-02 9.87537444e-01 -1.37058961e+00 -1.97788283e-01 1.84564870e-02 -4.44783092e-01 -4.45178032e-01 -1.31519347e-01 -7.62127116e-02 5.09989083e-01 -1.89438343e-01 -8.02560672e-02 -4.28781986e-01 1.50349796e-01 -7.97748446e-01 -2.68177181e-01 -2.71903813e-01 1.02316391e+00 -8.29222918e-01 9.46528494e-01 -1.74037445e+00 6.24027431e-01 3.51938367e-01 -2.53850967e-01 1.94096386e-01 2.35229865e-01 5.92261970e-01 -3.51189315e-01 -1.72087058e-01 -7.78757930e-01 -1.64017528e-01 -1.71469286e-01 5.75635254e-01 -6.18406199e-02 7.05002964e-01 6.44913375e-01 5.24745166e-01 -3.87826353e-01 2.59713065e-02 2.44040161e-01 6.42788351e-01 -3.17934990e-01 3.61409903e-01 -3.47273916e-01 4.45438474e-01 -5.59131540e-02 4.35061395e-01 6.96995914e-01 1.51186466e-01 -3.50335501e-02 -5.00331461e-01 1.47307426e-01 3.03229764e-02 -9.91268396e-01 1.24221289e+00 -6.26447260e-01 8.96408081e-01 2.91391015e-01 -1.25914717e+00 7.17750967e-01 6.09686613e-01 4.07758176e-01 -7.01602221e-01 4.77235854e-01 9.00977105e-02 2.31690422e-01 -8.05389047e-01 6.76487684e-02 -5.23227215e-01 1.23125143e-01 -2.19143689e-01 8.88105214e-01 -2.21214712e-01 3.02380659e-02 -1.81257918e-01 1.11763394e+00 -2.90909082e-01 -2.82658517e-01 -6.80236518e-01 6.50170147e-01 -1.13897964e-01 6.10590160e-01 3.70998859e-01 2.22893938e-01 5.74711323e-01 5.17219663e-01 -1.89974345e-02 -1.26288831e+00 -1.27803338e+00 -8.55551302e-01 1.40235722e-01 -6.35144338e-02 -1.25781387e-01 -6.26305342e-01 -2.37912923e-01 1.57518927e-02 1.26792169e+00 -2.28997603e-01 -5.55628955e-01 -4.23247844e-01 -1.30128741e+00 5.48148394e-01 6.79455340e-01 1.44511744e-01 -9.01279032e-01 -6.33512497e-01 1.18956581e-01 1.39445275e-01 -1.05909991e+00 6.07967257e-01 8.69742513e-01 -7.09544539e-01 -1.19786954e+00 -2.65003055e-01 -6.06356025e-01 3.41844290e-01 -9.12595153e-01 1.10115457e+00 -1.11653663e-01 -3.37015271e-01 4.45217967e-01 -2.54077882e-01 -3.97676408e-01 -1.11247349e+00 -2.64752626e-01 5.00662565e-01 -3.06314915e-01 -5.36733866e-02 -7.42595136e-01 -2.40809545e-01 2.39395872e-02 -8.37393403e-01 -7.35994101e-01 5.03141403e-01 1.05098057e+00 1.46881685e-01 9.31741714e-01 5.87738156e-01 -6.29890263e-01 5.88158250e-01 -8.61289084e-01 -6.66278124e-01 -1.57324687e-01 -8.61984730e-01 2.25214258e-01 7.59431839e-01 -4.18034464e-01 -1.05602539e+00 -4.63698894e-01 -6.33349657e-01 -7.08202481e-01 -3.19218755e-01 2.65350938e-01 -4.37114894e-01 -2.12790847e-01 6.52690291e-01 1.20385446e-01 -4.81514707e-02 -5.40403903e-01 -2.66841978e-01 6.54104650e-01 6.64248228e-01 -3.68948638e-01 8.39080453e-01 4.10051160e-02 3.07796091e-01 -1.15378702e+00 5.33088893e-02 3.22296433e-02 -2.52597362e-01 -3.98077995e-01 1.20053923e+00 -8.53016794e-01 -8.08895230e-01 3.94156635e-01 -1.16828620e+00 -1.77416354e-01 -2.03557238e-01 4.39455658e-01 -2.76256561e-01 3.85950387e-01 -9.84928012e-01 -1.01077282e+00 -3.61071020e-01 -1.44738793e+00 9.16842699e-01 2.98396140e-01 1.19644798e-01 -1.22953379e+00 1.12517506e-01 -1.77776247e-01 6.17800534e-01 2.85094976e-01 1.44664943e+00 -7.39758849e-01 -2.98870534e-01 -1.88638821e-01 5.51769257e-01 8.76797795e-01 -5.08000314e-01 -7.41473511e-02 -1.34382248e+00 -5.92797577e-01 7.03478336e-01 -7.33766481e-02 1.03487611e+00 3.83794844e-01 1.06077087e+00 -5.99431172e-02 -4.11911577e-01 5.32949686e-01 1.83447266e+00 4.73630667e-01 6.88285172e-01 1.57496393e-01 2.93232918e-01 2.54988968e-01 -1.52396023e-01 8.67135078e-02 -3.53845716e-01 6.67670906e-01 6.83305204e-01 1.17398627e-01 1.12761326e-01 1.65827394e-01 6.05138123e-01 8.56287420e-01 3.03251706e-02 -2.32879966e-01 -6.75725162e-01 4.66625810e-01 -1.24196362e+00 -9.48739290e-01 -4.17587876e-01 1.89785397e+00 2.39056647e-01 2.34486729e-01 -1.73624113e-01 5.65835357e-01 6.95681453e-01 -3.03190481e-02 -3.08193266e-01 -7.42923260e-01 -3.79956454e-01 6.46864653e-01 5.20501494e-01 2.94060141e-01 -7.54315853e-01 -4.41019572e-02 6.93307590e+00 7.96978116e-01 -9.87538278e-01 3.79043967e-01 7.65775070e-02 -7.90252239e-02 -2.98354477e-01 -5.94867915e-02 -4.90945667e-01 6.34225190e-01 1.63693750e+00 6.45228773e-02 3.37607086e-01 4.83124971e-01 -1.33145392e-01 -2.86578387e-01 -1.19629931e+00 6.60712659e-01 9.59048271e-02 -1.28039575e+00 -2.34077230e-01 -1.01571701e-01 5.70886552e-01 1.99956551e-01 -2.44954467e-01 4.61963415e-01 -2.52856612e-01 -1.15557051e+00 4.78011221e-01 7.01903522e-01 5.13562679e-01 -9.14331615e-01 1.27866042e+00 3.36379349e-01 -7.29507446e-01 -5.50315320e-01 -5.05110323e-01 1.58631662e-03 3.87118012e-01 6.89890981e-01 -9.00565445e-01 1.08973360e+00 8.23585033e-01 2.30762705e-01 -6.84682667e-01 9.15382743e-01 1.07371442e-01 8.45943511e-01 -3.92362833e-01 4.16540168e-02 -1.43829197e-01 -1.36295676e-01 9.55037415e-01 1.01152956e+00 6.06872022e-01 -4.50603485e-01 -2.88962871e-01 1.19269276e+00 2.32906431e-01 -5.38304090e-01 -5.00375330e-01 1.61574394e-01 3.35397571e-01 1.15858841e+00 -3.63305569e-01 -1.30318686e-01 -1.85536325e-01 7.90057659e-01 -1.67685837e-01 3.52274895e-01 -8.03549707e-01 -2.61773556e-01 6.26674592e-01 8.19530934e-02 5.98528802e-01 1.36644453e-01 -1.82322577e-01 -8.99955630e-01 -1.24447681e-01 -3.19270253e-01 4.26152706e-01 -1.00921631e+00 -1.49303973e+00 9.01767015e-01 5.23855805e-01 -1.13537216e+00 -4.73487198e-01 -9.70317125e-01 -1.17672324e+00 8.79484415e-01 -1.15705431e+00 -6.52884126e-01 8.63553360e-02 3.09917718e-01 2.98279494e-01 -4.83627051e-01 8.82512927e-01 2.84878135e-01 -6.36866629e-01 3.26075047e-01 1.88418359e-01 -1.56388015e-01 5.93040064e-02 -1.38127291e+00 -7.59121403e-02 8.40800345e-01 4.91094366e-02 -1.86354760e-02 1.23704219e+00 -6.12347603e-01 -1.26426589e+00 -9.01622295e-01 1.43445045e-01 -4.51173335e-01 5.95066309e-01 -4.45532322e-01 -1.21936238e+00 4.67855066e-01 8.53596628e-01 -3.77542049e-01 2.78056651e-01 -1.07713595e-01 3.08061391e-01 5.61651960e-02 -1.47251487e+00 -6.42460436e-02 4.16298270e-01 -6.49116457e-01 -9.13760662e-01 3.33101571e-01 4.30656284e-01 -2.77302057e-01 -1.41270065e+00 8.91043246e-01 -4.48075198e-02 -8.97407413e-01 7.42423058e-01 -4.12271529e-01 4.16407526e-01 -3.11718047e-01 -7.65550882e-02 -1.81979144e+00 -4.08665568e-01 1.43650174e-01 -5.27653635e-01 1.13278615e+00 4.97149259e-01 -6.40967011e-01 3.05545777e-01 -1.10844761e-01 -5.05357325e-01 -8.38288546e-01 -1.62423193e+00 -8.31179500e-01 4.89853263e-01 -6.29783750e-01 3.92314315e-01 6.58722222e-01 7.84432366e-02 3.42085958e-01 -7.16162398e-02 8.93941879e-01 5.34776092e-01 -1.32121846e-01 -3.09861034e-01 -1.46334815e+00 -6.27036512e-01 -3.21922749e-01 -9.13138449e-01 6.10833354e-02 2.88934469e-01 -1.20665836e+00 -1.14240460e-01 -1.30085814e+00 -1.03115104e-01 1.67553067e-01 -3.61218601e-01 4.74055596e-02 2.62574166e-01 1.62236959e-01 2.25285231e-03 -1.53557271e-01 1.06078908e-01 6.93396866e-01 4.72721368e-01 -5.21858633e-01 3.42400432e-01 4.01765592e-02 5.26630692e-02 4.72236723e-01 6.37040317e-01 -5.54722548e-01 -4.65867259e-02 -9.29935500e-02 1.07508600e-02 2.84454614e-01 5.96012831e-01 -1.86196351e+00 1.66553080e-01 6.13483906e-01 1.00027382e+00 -5.99948347e-01 3.51852983e-01 -1.01535821e+00 4.90377992e-01 6.65399849e-01 2.49211788e-01 4.82568666e-02 4.18563664e-01 5.22741497e-01 -2.36830279e-01 -7.58235395e-01 9.44535971e-01 7.19333664e-02 -6.01684868e-01 -1.69286609e-01 -9.56505537e-01 -3.54554415e-01 8.80062521e-01 2.69261509e-01 -1.97077185e-01 -1.77466035e-01 -9.09148633e-01 2.73654759e-01 3.73271853e-01 4.26572233e-01 4.96465027e-01 -1.32400107e+00 -5.99179864e-01 5.82404971e-01 -1.33150980e-01 -2.52586603e-01 4.62549597e-01 7.21712768e-01 -2.47891799e-01 3.12705725e-01 -3.72804880e-01 -1.03406382e+00 -9.04441357e-01 6.41154230e-01 1.02514029e+00 -1.19174615e-01 -6.67579532e-01 5.73762715e-01 -4.91513133e-01 4.87893187e-02 -9.79095623e-02 -3.44305575e-01 -5.44035494e-01 1.29431278e-01 1.73461124e-01 4.83091384e-01 6.36738539e-01 -5.98817110e-01 -2.68373966e-01 3.94438088e-01 4.53183502e-01 2.05866564e-02 1.47060049e+00 5.10492623e-01 -1.27259970e-01 8.31214428e-01 1.26304328e+00 -4.28737462e-01 -9.89151120e-01 4.45265055e-01 -4.15467881e-02 2.95966089e-01 4.77737248e-01 -8.40655565e-01 -1.19567525e+00 7.05572307e-01 1.25895464e+00 3.20287049e-01 1.16902804e+00 1.75283909e-01 4.41149801e-01 1.36068940e-01 3.23986501e-01 -1.24326277e+00 -4.09798533e-01 3.45525712e-01 7.50742078e-01 -6.72116041e-01 8.38494822e-02 1.41632438e-01 -4.89704490e-01 1.47617805e+00 4.39039469e-01 -3.05393636e-01 7.44721174e-01 6.07805192e-01 -1.96104646e-01 -5.62290311e-01 -8.62706840e-01 9.38090906e-02 3.01663101e-01 7.50959992e-01 8.86888131e-02 -2.80354768e-02 -2.10066170e-01 9.99576628e-01 -1.11560546e-01 -6.45402312e-01 8.74659896e-01 7.23399401e-01 -4.03046310e-01 -1.04978955e+00 -5.37026465e-01 7.51278698e-01 -4.49230701e-01 3.24571252e-01 2.19915882e-02 8.05785954e-01 2.69709200e-01 9.36552465e-01 3.13183755e-01 -4.78141010e-01 6.03421628e-01 7.15973735e-01 6.48515642e-01 -2.74890482e-01 -7.60454774e-01 -1.34355113e-01 2.97981024e-01 -3.70712817e-01 -2.32179508e-01 -6.86659038e-01 -1.21050382e+00 1.16951667e-01 -5.75742126e-01 4.52749491e-01 8.28603327e-01 1.10305011e+00 -1.15543872e-01 1.33725178e+00 6.17187858e-01 -1.04037333e+00 -6.10172212e-01 -1.41584980e+00 -9.35957193e-01 4.11916792e-01 4.48555171e-01 -9.55280066e-01 -8.15093160e-01 -4.91494149e-01]	[6.72311544418335, 2.4261562824249268]
63d701d5-cda1-4208-984d-efa5e36d49d3	evaluating-coreference-resolvers-on-community	null	null	https://aclanthology.org/2022.crac-1.7	https://aclanthology.org/2022.crac-1.7.pdf	Evaluating Coreference Resolvers on Community-based Question Answering: From Rule-based to State of the Art	Coreference resolution is a key step in natural language understanding. Developments in coreference resolution are mainly focused on improving the performance on standard datasets annotated for coreference resolution. However, coreference resolution is an intermediate step for text understanding and it is not clear how these improvements translate into downstream task performance. In this paper, we perform a thorough investigation on the impact of coreference resolvers in multiple settings of community-based question answering task, i.e., answer selection with long answers. Our settings cover multiple text domains and encompass several answer selection methods. We first inspect extrinsic evaluation of coreference resolvers on answer selection by using coreference relations to decontextualize individual sentences of candidate answers, and then annotate a subset of answers with coreference information for intrinsic evaluation. The results of our extrinsic evaluation show that while there is a significant difference between the performance of the rule-based system vs. state-of-the-art neural model on coreference resolution datasets, we do not observe a considerable difference on their impact on downstream models. Our intrinsic evaluation shows that (i) resolving coreference relations on less-formal text genres is more difficult even for trained annotators, and (ii) the values of linguistic-agnostic coreference evaluation metrics do not correlate with the impact on downstream data.	['Michael Strube', 'Iryna Gurevych', 'Nafise Sadat Moosavi', 'Haixia Chai']	null	null	null	null	coling-crac-2022-10	['coreference-resolution', 'answer-selection']	['natural-language-processing', 'natural-language-processing']	[ 4.42925006e-01 4.49064583e-01 -2.16840670e-01 -3.87511164e-01 -1.08658755e+00 -9.52335238e-01 6.59620047e-01 3.75191629e-01 -7.19442725e-01 9.10295367e-01 8.15908551e-01 -3.57253999e-01 -6.20347977e-01 -6.05069101e-01 -4.82344091e-01 -2.41019890e-01 2.75192231e-01 1.19457483e+00 4.49262530e-01 -7.67543435e-01 2.85591513e-01 2.45319486e-01 -1.53211260e+00 6.77517653e-01 8.90217721e-01 3.01518828e-01 4.25631888e-02 6.70319259e-01 -3.75043482e-01 8.92062426e-01 -6.32699132e-01 -7.73245811e-01 -3.45399797e-01 -5.74705422e-01 -1.86185884e+00 -8.41123998e-01 6.13822639e-01 2.31979594e-01 -2.45560497e-01 8.96476209e-01 5.97875476e-01 2.41473496e-01 4.54651803e-01 -9.17615950e-01 -1.60071522e-01 1.33768177e+00 2.57012937e-02 6.43740654e-01 8.88039410e-01 -2.11561888e-01 1.54016936e+00 -3.58615458e-01 9.94329929e-01 1.65205216e+00 7.16384053e-01 6.63046360e-01 -1.36790121e+00 -6.40266895e-01 -6.05948968e-03 5.24729967e-01 -1.01212335e+00 -6.93438768e-01 4.87878323e-01 -1.66302100e-01 1.20468855e+00 4.78421718e-01 -1.94078922e-01 1.22732043e+00 -1.35321021e-01 5.39647520e-01 6.64940655e-01 -6.03492796e-01 -1.93398386e-01 5.70503483e-03 5.96186697e-01 1.11245364e-01 1.63621604e-01 2.00917363e-01 -6.07515633e-01 -3.52534711e-01 1.05034158e-01 -8.05233240e-01 -9.07821417e-01 -3.48416686e-01 -1.10117066e+00 1.02533102e+00 4.09560293e-01 6.58834159e-01 -9.16329548e-02 -3.31643701e-01 6.09581709e-01 6.76296711e-01 -5.82969896e-02 1.02304626e+00 -6.43605530e-01 -1.77173808e-01 -6.25700951e-01 3.87628675e-01 1.35984349e+00 6.73549354e-01 3.89580399e-01 -7.77322650e-01 -2.88772613e-01 8.99960756e-01 -5.32048754e-02 3.00758898e-01 3.61491889e-01 -1.57098722e+00 7.69524395e-01 6.56011522e-01 2.66699493e-01 -1.08193099e+00 -7.88809776e-01 -2.86874294e-01 -2.25008249e-01 -2.04777285e-01 9.06572342e-01 -1.50000662e-01 -1.87866092e-02 2.16644788e+00 8.26314911e-02 -3.08538109e-01 4.64716434e-01 8.79301429e-01 1.24035823e+00 1.74430966e-01 2.62265205e-01 -4.33443308e-01 1.59208846e+00 -7.07413971e-01 -7.01973736e-01 -2.13479698e-01 8.89015019e-01 -9.18907702e-01 8.32473040e-01 1.42170861e-01 -1.02811575e+00 -4.36725765e-01 -7.66720891e-01 -1.69677600e-01 1.56174134e-02 -3.80435854e-01 4.08692211e-01 4.03617054e-01 -8.17787230e-01 7.44257927e-01 -3.85342419e-01 -7.80214667e-01 -1.78149104e-01 4.74620789e-01 -4.96371448e-01 -2.26756185e-02 -1.64176190e+00 1.44651830e+00 5.42925239e-01 -4.08292741e-01 -2.47720107e-01 -7.49698818e-01 -7.49498248e-01 1.94924146e-01 4.62999642e-01 -7.73194849e-01 1.61786747e+00 -8.65268111e-01 -1.07425594e+00 1.17371023e+00 -3.60823214e-01 -4.35758054e-01 2.43220478e-01 -2.47946054e-01 -5.61089277e-01 2.12009072e-01 3.20791721e-01 6.96619928e-01 5.70943430e-02 -1.19057071e+00 -9.89052594e-01 -2.65635908e-01 3.24950069e-01 5.21156013e-01 1.71763748e-01 3.57615203e-01 -2.12839365e-01 -3.33940610e-02 1.69180315e-02 -7.51334608e-01 1.62304148e-01 -6.98880792e-01 -1.15993224e-01 -7.46057749e-01 6.84223473e-01 -2.98134387e-01 1.20805073e+00 -1.78894281e+00 2.21041933e-01 -1.72091663e-01 -8.52152035e-02 3.74493361e-01 -4.11237568e-01 5.49833357e-01 -4.26796019e-01 1.05047189e-01 -1.64432719e-01 7.69912302e-02 5.36483433e-03 2.12128907e-01 -5.38097501e-01 1.24372669e-01 -1.74693782e-02 7.73970306e-01 -1.18338704e+00 -5.52287102e-01 -1.28850386e-01 1.74364507e-01 -7.26059556e-01 7.98332021e-02 -3.09275359e-01 4.74229425e-01 -4.53569442e-01 8.53966549e-02 2.51583040e-01 -9.98875648e-02 5.53455412e-01 -3.92345160e-01 -1.90078706e-01 1.07119143e+00 -1.04273760e+00 1.80765676e+00 -3.28600883e-01 6.19491339e-01 2.58843184e-01 -9.33910370e-01 6.96732998e-01 6.78218663e-01 9.31850448e-02 -7.82359660e-01 1.46218881e-01 3.21754009e-01 6.18355036e-01 -5.87234437e-01 6.15845442e-01 -2.62761235e-01 -2.16078103e-01 8.21287334e-01 1.63455144e-01 1.82567965e-02 5.03315151e-01 3.64827335e-01 1.23018348e+00 1.01932444e-01 2.01964691e-01 -6.03458047e-01 9.13879752e-01 5.87193310e-01 6.03347182e-01 7.95449436e-01 -2.44248644e-01 4.97152865e-01 5.07702947e-01 -1.07944317e-01 -4.55744833e-01 -9.31446493e-01 -3.24371308e-01 1.37477350e+00 1.95274383e-01 -4.70761925e-01 -7.20790327e-01 -7.94880092e-01 -2.94634998e-01 1.16765082e+00 -3.85451347e-01 -1.44639552e-01 -1.08198583e+00 -6.59779787e-01 9.46919024e-01 3.72011721e-01 2.22878680e-01 -1.42266834e+00 -8.08676243e-01 2.06005618e-01 -1.03158045e+00 -1.05203044e+00 -3.88641357e-01 2.76968956e-01 -7.69518316e-01 -1.73909569e+00 -4.67791259e-02 -8.72330129e-01 -3.43177281e-03 6.79221153e-02 1.77855027e+00 2.98781157e-01 3.15521479e-01 6.08717501e-01 -5.19330680e-01 -5.81960455e-02 -6.85058951e-01 4.96213645e-01 -1.58589810e-01 -6.47802472e-01 1.11022484e+00 -4.65569109e-01 -3.76960278e-01 4.81762320e-01 -3.82591635e-01 -5.31849086e-01 2.69044906e-01 8.41387749e-01 6.88844919e-02 -4.10401523e-01 5.79525352e-01 -1.20926571e+00 1.05055201e+00 -3.47460389e-01 -1.36132613e-01 3.63638788e-01 -4.60293919e-01 4.55730855e-01 3.47145647e-01 -1.72750473e-01 -1.47782958e+00 -3.65020841e-01 -3.66956770e-01 1.26713574e-01 -3.72324079e-01 4.82014835e-01 -2.07108140e-01 3.45059156e-01 1.09575939e+00 -5.01043677e-01 -2.48802990e-01 -5.39162517e-01 4.47793752e-01 6.50405884e-01 1.07407749e+00 -1.03331399e+00 4.51222867e-01 1.85049787e-01 -3.73763382e-01 -4.77178037e-01 -1.22925687e+00 -7.14066267e-01 -7.65067279e-01 2.96981990e-01 8.38354766e-01 -4.90046114e-01 -9.62445378e-01 -3.88526917e-01 -1.52532279e+00 -2.31171340e-01 -2.38914356e-01 5.04576683e-01 -6.52909100e-01 4.97998923e-01 -6.71606183e-01 -2.85134405e-01 -5.67824066e-01 -1.07976913e+00 6.04861140e-01 1.98403433e-01 -1.17252409e+00 -1.11820281e+00 5.77396333e-01 8.59381855e-01 8.61321390e-02 -1.89505905e-01 1.29165494e+00 -1.27877748e+00 -3.70582975e-02 2.11468771e-01 -2.14469716e-01 -3.61100823e-01 -1.33887067e-01 -3.46280307e-01 -1.03189564e+00 -9.77748856e-02 -2.55731903e-02 -5.21766961e-01 8.22325110e-01 6.55998141e-02 1.59234703e-01 4.66668867e-02 -5.95578313e-01 2.33925566e-01 1.02461207e+00 1.40915468e-01 6.20440006e-01 6.17443204e-01 2.43972898e-01 1.22575140e+00 7.76766300e-01 -2.79723734e-01 3.73655915e-01 8.40530634e-01 2.56622583e-01 5.18024027e-01 -3.66724640e-01 -2.75741965e-01 1.08243309e-01 4.91147310e-01 7.15076029e-02 -2.52421677e-01 -1.10010242e+00 7.69035935e-01 -1.97518694e+00 -1.30883050e+00 -6.80786133e-01 2.11934543e+00 1.18503726e+00 7.00557977e-02 -1.26366690e-01 1.55182078e-01 9.76376891e-01 -1.44318953e-01 -1.12980552e-01 -5.31880677e-01 -2.52884269e-01 4.72605199e-01 -4.24606465e-02 9.94984090e-01 -9.68891144e-01 9.88609970e-01 6.29121542e+00 3.64217639e-01 -5.91278851e-01 6.26710281e-02 -6.19879924e-02 7.99473524e-02 -4.33044136e-01 2.30779141e-01 -9.20369267e-01 -1.39892045e-02 1.11830151e+00 -1.14849424e-02 1.48375288e-01 3.29487324e-01 -7.69548491e-02 5.51333837e-02 -1.51639020e+00 5.14267743e-01 1.10560358e-01 -1.19178677e+00 -7.37316813e-03 -4.56284076e-01 3.26896787e-01 2.06442982e-01 -6.16575837e-01 6.02885604e-01 6.77681029e-01 -1.03865349e+00 3.30343574e-01 2.39902049e-01 4.35787439e-01 -7.83112168e-01 1.10810542e+00 3.75014126e-01 -8.10284495e-01 -1.09707341e-01 -3.91928494e-01 -9.03206989e-02 4.28314507e-01 8.97128775e-04 -3.68572146e-01 5.37889063e-01 7.98026681e-01 2.35741585e-01 -3.49092811e-01 8.55025530e-01 -6.05821788e-01 5.34046471e-01 -2.10331693e-01 2.26116642e-01 1.05279945e-02 8.78438354e-02 8.30238640e-01 1.36885369e+00 -1.27215609e-01 4.64689761e-01 -1.63973093e-01 8.45158756e-01 -1.40765965e-01 1.28995255e-01 -3.32243562e-01 3.30986530e-01 9.47490990e-01 1.09639072e+00 -2.43937925e-01 -9.33294669e-02 -3.98666620e-01 5.81653059e-01 5.50729275e-01 1.38812751e-01 -4.12093043e-01 -4.22275752e-01 6.85949802e-01 -1.20860666e-01 2.08128616e-01 5.41238785e-01 -1.48715422e-01 -9.71415818e-01 -4.16338235e-01 -1.33393693e+00 1.23723495e+00 -6.06629252e-01 -1.26683152e+00 5.40267229e-01 4.54252772e-02 -5.85342169e-01 -5.96270919e-01 -3.99225444e-01 -8.81107748e-01 9.42132413e-01 -1.43992567e+00 -7.26038575e-01 -1.06541082e-01 6.54399455e-01 3.56674284e-01 1.05000921e-01 1.18327928e+00 2.11939961e-01 -3.26100081e-01 6.59453630e-01 -3.53532612e-01 4.10856485e-01 1.19100475e+00 -1.14740419e+00 1.29488900e-01 6.80460870e-01 1.70152649e-01 9.07989264e-01 1.16334450e+00 -4.50917244e-01 -8.64581764e-01 -6.02991521e-01 1.55398798e+00 -1.02980089e+00 6.10222638e-01 3.12328935e-01 -1.18609679e+00 8.31463695e-01 5.59888065e-01 -6.44499660e-01 7.16090858e-01 8.26521814e-01 -5.67162335e-01 2.70671815e-01 -9.99450445e-01 4.94732499e-01 1.26636803e+00 -4.97267812e-01 -1.64261663e+00 -6.05543479e-02 8.36255133e-01 -3.33104938e-01 -9.44449663e-01 7.66962886e-01 1.74212381e-01 -9.98075426e-01 9.21811223e-01 -9.48286355e-01 4.12231058e-01 -1.72744915e-01 -1.94563210e-01 -1.13246977e+00 -4.95956421e-01 -3.43038529e-01 2.11637706e-01 1.49387431e+00 7.87473083e-01 -3.00805897e-01 6.20322049e-01 7.36447334e-01 -1.30615696e-01 -2.82777399e-02 -9.62720394e-01 -2.98661649e-01 5.73766291e-01 -2.66341686e-01 3.22765976e-01 1.19323158e+00 6.67769790e-01 1.20907569e+00 3.16079378e-01 -1.64541230e-02 5.36620080e-01 5.69166720e-01 5.90980351e-01 -1.64568329e+00 -2.22535208e-01 -6.19245172e-01 2.70666420e-01 -8.77146542e-01 7.07939744e-01 -9.98457491e-01 3.90129462e-02 -1.37404394e+00 3.48994732e-01 -4.28439945e-01 -9.44758281e-02 2.58860409e-01 -2.81418592e-01 -6.05646148e-02 7.57629126e-02 3.39893013e-01 -7.03748584e-01 1.54830992e-01 9.10460472e-01 -7.29817599e-02 -2.15486795e-01 -3.73874232e-02 -1.03781104e+00 8.99741828e-01 6.56196356e-01 -5.96200764e-01 -4.95776474e-01 -4.89799887e-01 3.70275557e-01 2.51105875e-01 3.39973159e-02 -5.45257092e-01 4.64345545e-01 8.87325779e-03 -2.52713650e-01 -3.06536287e-01 6.84306771e-03 -4.32071328e-01 -2.00520813e-01 4.93042737e-01 -6.79026544e-01 -8.04327354e-02 2.75787294e-01 2.24898860e-01 -2.32405171e-01 -7.49356270e-01 6.07334375e-01 -3.10424864e-01 -6.54727578e-01 -4.15966302e-01 -2.12094128e-01 9.17869627e-01 2.44853318e-01 5.63797243e-02 -7.28215516e-01 -3.80721033e-01 -7.28948474e-01 4.09331828e-01 4.24067557e-01 7.02154815e-01 1.04588374e-01 -8.83207798e-01 -1.06003857e+00 -4.52920049e-01 2.09155232e-01 -2.29483679e-01 -3.37429047e-02 8.75922799e-01 -1.14078715e-01 9.89585698e-01 -2.63634454e-02 -4.77447808e-01 -1.49793971e+00 5.17510712e-01 6.10583425e-01 -5.52906930e-01 -4.16733921e-01 8.84892404e-01 1.37879699e-01 -8.45213830e-01 3.41384888e-01 5.47725484e-02 -8.55713785e-01 4.01208311e-01 5.68734825e-01 2.61630088e-01 1.45218760e-01 -7.15882301e-01 -5.79952776e-01 4.83684450e-01 -1.63865730e-01 -9.39435437e-02 1.09982622e+00 -1.87841684e-01 -1.94466785e-01 1.50145382e-01 7.86897361e-01 7.44223371e-02 -5.49121022e-01 -3.17420721e-01 4.99516994e-01 -4.21508588e-02 -2.44147107e-01 -8.87836277e-01 -6.01354361e-01 7.15836167e-01 1.73204482e-01 2.79914811e-02 7.09828615e-01 3.81401181e-01 6.29319429e-01 7.56157577e-01 3.29968840e-01 -9.44617927e-01 -3.68673146e-01 9.31001365e-01 9.24745500e-01 -1.23429930e+00 -1.66628748e-01 -4.43946868e-01 -4.79927897e-01 9.49131787e-01 7.40240276e-01 3.04814637e-01 1.25590414e-01 9.73664597e-02 3.40411901e-01 -4.51844662e-01 -1.06564879e+00 -3.74675423e-01 1.17821626e-01 5.76700985e-01 1.03350258e+00 -2.37697542e-01 -7.67031789e-01 6.78391874e-01 -5.95225215e-01 -3.43561441e-01 4.31130797e-01 3.77916574e-01 -5.91871023e-01 -1.37353373e+00 -5.26425838e-01 6.78711981e-02 -6.24939740e-01 -1.93303451e-01 -8.38029087e-01 9.33022380e-01 -4.85852174e-02 1.55079246e+00 1.37733772e-01 -1.14886552e-01 5.36106527e-01 2.39637181e-01 6.03269637e-01 -7.74829745e-01 -8.96412909e-01 -5.65164268e-01 9.22040224e-01 -4.45845604e-01 -7.01862395e-01 -9.00891066e-01 -1.46857679e+00 -3.05605143e-01 -5.09476960e-01 9.26719844e-01 -1.07740074e-01 1.34855533e+00 2.14463711e-01 2.57320791e-01 -2.04564303e-01 -2.78523624e-01 -5.60231626e-01 -1.06050539e+00 1.21745616e-01 9.60466743e-01 6.14031106e-02 -6.36141300e-01 -4.72455114e-01 -1.99174270e-01]	[9.338930130004883, 9.501349449157715]
8b6a613d-e8fb-4777-a89d-ce055a7e272e	nnqs-transformer-an-efficient-and-scalable	2306.16705	null	https://arxiv.org/abs/2306.16705v2	https://arxiv.org/pdf/2306.16705v2.pdf	NNQS-Transformer: an Efficient and Scalable Neural Network Quantum States Approach for Ab initio Quantum Chemistry	Neural network quantum state (NNQS) has emerged as a promising candidate for quantum many-body problems, but its practical applications are often hindered by the high cost of sampling and local energy calculation. We develop a high-performance NNQS method for \textit{ab initio} electronic structure calculations. The major innovations include: (1) A transformer based architecture as the quantum wave function ansatz; (2) A data-centric parallelization scheme for the variational Monte Carlo (VMC) algorithm which preserves data locality and well adapts for different computing architectures; (3) A parallel batch sampling strategy which reduces the sampling cost and achieves good load balance; (4) A parallel local energy evaluation scheme which is both memory and computationally efficient; (5) Study of real chemical systems demonstrates both the superior accuracy of our method compared to state-of-the-art and the strong and weak scalability for large molecular systems with up to $120$ spin orbitals.	['Honghui Shang', 'Pengyu Zhou', 'Yi Fan', 'Chu Guo', 'Yangjun Wu']	2023-06-29	null	null	null	null	['variational-monte-carlo']	['miscellaneous']	[ 8.59963223e-02 -5.75458229e-01 -1.62330583e-01 -2.78960466e-01 -1.04254782e+00 -6.54466450e-02 3.86866242e-01 3.54532242e-01 -6.41502380e-01 1.20387518e+00 -5.11038378e-02 -5.61321259e-01 -2.08920136e-01 -1.13707316e+00 -6.64975345e-01 -1.24561679e+00 -2.15268061e-02 6.78530991e-01 3.30569863e-01 -5.17884552e-01 5.22545934e-01 7.28253424e-01 -1.64044702e+00 7.77600184e-02 1.04723787e+00 9.94919062e-01 -6.61528483e-02 2.35075235e-01 1.37412712e-01 4.50985312e-01 -1.13163717e-01 -2.24159792e-01 1.78960875e-01 -5.47014832e-01 -1.08191621e+00 -8.45953703e-01 2.63846517e-01 -2.95805670e-02 -3.68493021e-01 1.26756704e+00 8.89971793e-01 5.50999343e-01 6.88072562e-01 -3.61004710e-01 -4.02932078e-01 3.31013680e-01 -1.93536103e-01 1.24094725e-01 -5.06683774e-02 5.66424549e-01 1.05327988e+00 -7.63271153e-01 5.61139762e-01 8.54391873e-01 7.42535114e-01 6.99111104e-01 -1.72899985e+00 -5.62185466e-01 -9.28789675e-01 5.45211792e-01 -1.78565764e+00 -6.20840907e-01 5.77615976e-01 -8.98884237e-02 1.78235829e+00 1.19943067e-01 7.56248593e-01 6.82578385e-01 6.68050587e-01 -1.65568441e-01 1.14717841e+00 -7.08217978e-01 9.27041888e-01 -1.74866453e-01 2.92118669e-01 7.83567131e-01 2.11826578e-01 6.12190366e-01 -6.76845849e-01 -6.94329202e-01 4.78856653e-01 -3.71611770e-03 1.38860941e-01 -5.22657633e-01 -1.12529874e+00 1.11375070e+00 6.29650235e-01 1.88848719e-01 -7.43683875e-01 5.84187210e-01 7.09247112e-01 1.30770756e-02 1.95565090e-01 6.82846606e-01 -3.15594196e-01 -3.84241581e-01 -1.13206565e+00 5.44668674e-01 6.96005821e-01 2.31386885e-01 9.11956131e-01 2.33596757e-01 -9.72384438e-02 5.18951595e-01 1.83112279e-01 5.90849996e-01 2.42845997e-01 -9.18977618e-01 -9.21796337e-02 1.66416302e-01 1.55529350e-01 -2.95061618e-01 -3.74382138e-01 -1.03688903e-01 -8.32119524e-01 2.61755198e-01 1.30914524e-01 5.93372099e-02 -6.02396429e-01 1.61537802e+00 5.80037594e-01 -2.84629852e-01 1.54780775e-01 4.15216029e-01 8.84347320e-01 8.41434419e-01 1.54107124e-01 -5.54886937e-01 1.23691618e+00 -6.82773709e-01 -6.40600264e-01 5.69313407e-01 8.46634984e-01 -4.96951848e-01 6.87421620e-01 1.45263255e-01 -1.38459468e+00 -3.15356672e-01 -1.13779378e+00 -3.27029228e-01 -5.89334011e-01 -1.94067463e-01 1.25967050e+00 9.50490952e-01 -9.85302091e-01 1.38310516e+00 -1.22017419e+00 -1.10676058e-01 3.51596981e-01 1.09294653e+00 -3.52734447e-01 1.02079824e-01 -1.27065885e+00 9.82688367e-01 8.13894033e-01 -2.21109539e-01 -7.28009880e-01 -6.88224912e-01 -5.74407697e-01 4.62210290e-02 2.11578608e-01 -7.57312000e-01 9.22314286e-01 -3.12287599e-01 -2.16326237e+00 3.60531628e-01 -5.22369325e-01 -4.11029279e-01 -7.64274672e-02 5.77434421e-01 -2.92995423e-01 1.32816419e-01 -2.45859604e-02 5.72003186e-01 2.06391484e-01 -3.04906458e-01 1.02521077e-01 -5.49431920e-01 -4.37640607e-01 1.70631975e-01 -1.88674971e-01 -3.74249108e-02 1.80126712e-01 1.05825916e-01 3.10095251e-01 -8.58631611e-01 -5.94356537e-01 -5.56405663e-01 -2.55610347e-01 -4.32726383e-01 2.03491449e-01 -6.09714612e-02 1.29130673e+00 -1.66188276e+00 2.59829432e-01 5.11674464e-01 7.40162581e-02 3.54039818e-01 2.83894420e-01 9.36277866e-01 -3.28534365e-01 -2.26636440e-01 -1.99175522e-01 5.67191839e-02 5.86965419e-02 1.40797198e-01 7.68873766e-02 5.25044978e-01 -3.34860772e-01 1.02761853e+00 -7.18449235e-01 -2.50271261e-01 2.50586778e-01 6.44205987e-01 -1.01611972e+00 -1.86974302e-01 1.13351448e-02 2.45346427e-01 -3.17314625e-01 5.14155328e-01 6.32454216e-01 -4.94222432e-01 2.50024438e-01 -6.19100571e-01 -6.53792918e-01 6.49303436e-01 -1.23327887e+00 1.70232940e+00 1.25095382e-01 -2.27363706e-01 -1.08356610e-01 -1.04065382e+00 5.83328545e-01 5.69912434e-01 5.55267036e-01 -7.89194345e-01 1.00922920e-01 6.52622342e-01 2.87309378e-01 -2.50129730e-01 5.87043226e-01 -7.44668126e-01 2.17540681e-01 2.21702382e-01 3.02403480e-01 -3.34253401e-01 3.51264238e-01 -4.37846929e-02 9.61578608e-01 2.57865936e-01 5.27251184e-01 -1.02389145e+00 5.65944254e-01 3.14673901e-01 3.73614818e-01 6.93543971e-01 -3.63591671e-01 -4.35225852e-02 2.30075821e-01 -8.87084186e-01 -1.23443222e+00 -7.56418943e-01 -7.81027913e-01 1.03474295e+00 4.54048850e-02 -6.61975801e-01 -9.66880918e-01 1.16295457e-01 -1.42638355e-01 5.18934071e-01 -4.20344889e-01 -2.25370571e-01 -3.21478635e-01 -1.42753994e+00 3.07501167e-01 2.68551081e-01 4.22825217e-01 -1.14721620e+00 -3.54657143e-01 6.60242856e-01 2.09164023e-01 -4.50552315e-01 -1.94202796e-01 7.52398074e-01 -1.11264074e+00 -8.10066402e-01 -1.62405655e-01 -2.82274574e-01 3.70984644e-01 1.69935822e-02 9.44724917e-01 -1.89107567e-01 -5.44894874e-01 -3.51797551e-01 3.43433112e-01 -6.24929704e-02 -5.52187741e-01 1.91307902e-01 4.69348937e-01 -4.70641971e-01 7.49646008e-01 -7.48811245e-01 -7.82963991e-01 -1.64906353e-01 -7.23478436e-01 -1.23656847e-01 4.00919974e-01 1.29844904e+00 8.82921100e-01 9.85732675e-02 3.10520202e-01 -6.92168713e-01 5.31037211e-01 -6.80621490e-02 -9.07000005e-01 6.30800873e-02 -1.02981150e+00 5.77211678e-01 6.49623156e-01 2.90870667e-01 -1.00276196e+00 1.51872188e-01 -6.65730178e-01 3.09591234e-01 2.04129517e-02 4.18877780e-01 3.13862443e-01 -6.53286576e-01 1.12051296e+00 3.97691399e-01 2.39439353e-01 -4.03985530e-01 1.69533014e-01 4.71848756e-01 2.23444253e-01 -9.63757813e-01 4.55506086e-01 4.66414452e-01 8.70427370e-01 -1.26138616e+00 -4.96143281e-01 -3.33674103e-01 -8.59421015e-01 3.44457299e-01 1.09876204e+00 -7.36171484e-01 -1.57534206e+00 2.03104034e-01 -9.12536085e-01 -7.02194721e-02 -2.88061261e-01 6.76788807e-01 -5.24290740e-01 5.41870952e-01 -9.30116296e-01 -8.08023751e-01 -8.30070913e-01 -1.75182259e+00 8.71070564e-01 1.70674473e-01 -6.10949993e-02 -9.36704814e-01 3.44338626e-01 3.35447192e-01 6.97960734e-01 3.48446369e-01 1.12704086e+00 -2.18718961e-01 -5.52388251e-01 -1.90603331e-01 -2.29194477e-01 -1.64248154e-03 -2.72889584e-01 4.09514681e-02 -8.67761254e-01 -6.31338418e-01 -4.50285107e-01 -7.19376624e-01 8.30848277e-01 6.36812747e-01 9.23012853e-01 2.36130692e-02 -3.91101569e-01 5.16029716e-01 1.61284769e+00 2.12205932e-01 6.43222392e-01 -5.79415821e-03 6.77806854e-01 1.64256886e-01 -1.50510862e-01 2.38232419e-01 3.87391374e-02 9.02488351e-01 4.33229566e-01 -7.60703161e-02 3.16114336e-01 4.07560952e-02 3.47304583e-01 1.12320936e+00 -6.49442077e-01 4.88829166e-01 -8.21988463e-01 3.62534113e-02 -1.78022456e+00 -1.35234296e+00 -4.16467875e-01 2.52406907e+00 1.15035748e+00 2.22988124e-03 1.17503837e-01 -1.73670556e-02 1.60807833e-01 1.37298718e-01 -6.66485488e-01 -5.92253089e-01 2.79891491e-01 1.05778730e+00 6.28426790e-01 4.25463408e-01 -9.41960096e-01 1.08472538e+00 7.14074564e+00 1.24767900e+00 -1.13948107e+00 4.13110822e-01 2.91335851e-01 -1.41848654e-01 -3.04339156e-02 9.10680443e-02 -9.33385909e-01 2.99183995e-01 1.50968325e+00 2.00813767e-02 7.92630494e-01 7.74926603e-01 3.48121189e-02 1.79191791e-02 -8.70153487e-01 9.54551816e-01 -4.51141626e-01 -2.11206484e+00 -4.65154797e-02 4.97806221e-01 8.84466469e-01 6.58226132e-01 -2.26266816e-01 1.96741000e-01 7.93020986e-03 -1.18144405e+00 4.45717722e-01 2.17545226e-01 1.12270808e+00 -1.10496604e+00 6.21710539e-01 1.21539518e-01 -1.04880273e+00 2.54820198e-01 -7.41544425e-01 -2.64805913e-01 -1.17420658e-01 4.02795285e-01 -1.48754463e-01 5.13417661e-01 6.00297034e-01 3.81706864e-01 -1.16536476e-01 5.08859754e-01 2.26946652e-01 5.85847139e-01 -4.61515695e-01 -5.69603264e-01 6.15870059e-01 -7.36444414e-01 1.55005693e-01 6.56762838e-01 8.52874145e-02 1.61722317e-01 4.68481518e-02 1.18663836e+00 3.42935957e-02 2.08401486e-01 -4.65187579e-01 -1.23659737e-01 4.72347528e-01 9.95631993e-01 -5.84574759e-01 -2.45734751e-01 -1.66392758e-01 6.08798623e-01 2.92282820e-01 1.32576406e-01 -4.19681013e-01 -6.59853995e-01 6.01899147e-01 -1.41856387e-01 3.17947298e-01 -2.58403510e-01 6.57909513e-02 -1.09903371e+00 -4.58022565e-01 -5.71412921e-01 7.25025088e-02 -3.12993318e-01 -1.07133400e+00 6.34411097e-01 -1.72615319e-01 -6.17945492e-01 -1.76264524e-01 -8.92123699e-01 -5.19545615e-01 1.12900698e+00 -1.40000486e+00 -7.34467804e-01 2.65635669e-01 5.42535305e-01 -2.74554402e-01 -1.45422474e-01 1.54042435e+00 2.77490556e-01 -7.56994307e-01 3.81923378e-01 1.08916199e+00 -5.07977426e-01 2.95662016e-01 -1.25668502e+00 2.96596855e-01 4.41130728e-01 -2.01069325e-01 1.18122840e+00 6.95995808e-01 -3.13630551e-01 -1.97065413e+00 -6.93531215e-01 8.75403464e-01 -3.69116217e-01 5.50792992e-01 -1.71314731e-01 -9.30007100e-01 9.49688479e-02 1.15884565e-01 2.39099234e-01 1.00176466e+00 3.09802145e-01 -2.58678533e-02 -2.19004899e-01 -1.35235155e+00 5.25648952e-01 7.19665945e-01 -7.81493485e-01 -1.88209534e-01 7.93309271e-01 2.35336572e-01 -4.29613858e-01 -1.18205297e+00 1.40398473e-01 7.75898457e-01 -1.29473519e+00 1.14529431e+00 -6.16232276e-01 1.88903332e-01 -2.53215998e-01 -2.65220642e-01 -8.06317508e-01 -5.82895577e-01 -9.84514594e-01 2.54434887e-02 3.98973525e-01 3.89338344e-01 -9.41110373e-01 8.04018021e-01 7.74470925e-01 -2.57580489e-01 -8.77611995e-01 -1.52908671e+00 -5.76986670e-01 3.98525536e-01 -1.58830807e-01 6.02951229e-01 6.75583363e-01 3.57740521e-01 3.78735512e-01 -2.22430691e-01 -3.22240740e-02 7.54141331e-01 1.23040237e-01 3.40204090e-01 -1.14103580e+00 -4.72966552e-01 -4.32486266e-01 -2.49958962e-01 -6.95544124e-01 -1.59500563e-03 -1.06046867e+00 -2.57228941e-01 -1.01946366e+00 6.19422197e-01 -8.57630968e-02 -5.08135736e-01 3.76493782e-01 2.03185081e-01 4.93662268e-01 -5.06658673e-01 3.69104564e-01 -6.07857347e-01 8.92500997e-01 1.04828691e+00 1.36365801e-01 -2.12389380e-01 -4.12829578e-01 -2.73016751e-01 4.07471836e-01 3.32635880e-01 -7.54685938e-01 3.11726797e-02 2.97926992e-01 3.19865882e-01 1.66541696e-01 1.97455630e-01 -1.17210555e+00 1.86934873e-01 -1.49065465e-01 1.44728616e-01 -5.48409224e-01 5.34974098e-01 -3.27714413e-01 1.35594264e-01 1.08149266e+00 3.11356373e-02 -1.21873906e-02 8.19080621e-02 3.99110526e-01 -1.02923125e-01 -4.31299061e-01 1.31456733e+00 -3.90067071e-01 -2.64435172e-01 5.36032438e-01 -2.00756058e-01 -3.83908004e-01 7.31591344e-01 -2.36079291e-01 -9.78752822e-02 3.17368478e-01 -2.79286176e-01 -2.80308545e-01 6.52633727e-01 -6.68310940e-01 1.57463737e-03 -1.42920303e+00 -2.46239766e-01 3.36639881e-01 1.59154236e-01 -2.14914843e-01 3.84832531e-01 9.56254363e-01 -1.09444797e+00 7.88708985e-01 -6.12996817e-02 -5.73564291e-01 -1.03054547e+00 3.09468806e-01 6.30493820e-01 -3.35924745e-01 -4.17625815e-01 7.40894020e-01 -4.04426157e-01 -5.43264329e-01 -3.55621070e-01 -7.23493919e-02 2.02666670e-01 -3.97784710e-01 5.22720098e-01 7.37520337e-01 5.21808743e-01 -7.72737861e-01 -4.26287562e-01 4.83357579e-01 -1.51352614e-01 4.97566722e-03 1.67490959e+00 5.18657863e-01 -5.36712110e-01 8.09936225e-02 1.17596734e+00 -3.19359064e-01 -1.00016177e+00 -2.72448242e-01 -2.34561473e-01 5.94657063e-02 7.28405118e-01 -5.68167329e-01 -4.74614143e-01 1.04194534e+00 8.34107995e-01 1.97010338e-01 4.21062559e-01 -4.11466926e-01 9.41590130e-01 1.07043755e+00 6.07436717e-01 -1.35696208e+00 -3.26184541e-01 6.64890170e-01 2.09042281e-01 -1.20476329e+00 4.38256890e-01 1.26136646e-01 -9.94683281e-02 1.35468626e+00 5.87806366e-02 3.77125069e-02 5.85328341e-01 -2.67204940e-01 -3.89461190e-01 -8.07236731e-01 -5.55672467e-01 6.04485832e-02 2.57015854e-01 1.57249093e-01 6.96549296e-01 1.73568085e-01 -5.00762939e-01 1.16977043e-01 -3.47633138e-02 8.65337178e-02 1.60097793e-01 9.91201222e-01 -3.20807129e-01 -1.53265691e+00 -9.31732357e-02 3.31301063e-01 -4.18918967e-01 -4.51214939e-01 1.99749857e-01 5.82648814e-01 8.31153542e-02 6.40687823e-01 -3.59250724e-01 1.70472771e-01 7.43348151e-02 4.74136680e-01 7.56475270e-01 -6.81948245e-01 -7.34437764e-01 -1.11502796e-01 -1.11473829e-01 -8.73386323e-01 -5.36216855e-01 -4.96185839e-01 -1.43891847e+00 -7.77968049e-01 -6.95778966e-01 7.59317100e-01 8.83488119e-01 1.07336020e+00 7.45858550e-01 3.74837667e-01 4.14611161e-01 -1.22827399e+00 -1.12199044e+00 -8.59793425e-01 -8.69600117e-01 2.47903839e-01 4.58969211e-04 -6.75708652e-01 -1.14727803e-01 -4.44088757e-01]	[5.39016580581665, 5.124839782714844]
004fba26-5f98-4ac8-abd3-092cf3bc4b1a	automatic-generation-of-multiple-choice	2303.14576	null	https://arxiv.org/abs/2303.14576v1	https://arxiv.org/pdf/2303.14576v1.pdf	Automatic Generation of Multiple-Choice Questions	Creating multiple-choice questions to assess reading comprehension of a given article involves generating question-answer pairs (QAPs) and adequate distractors. We present two methods to tackle the challenge of QAP generations: (1) A deep-learning-based end-to-end question generation system based on T5 Transformer with Preprocessing and Postprocessing Pipelines (TP3). We use the finetuned T5 model for our downstream task of question generation and improve accuracy using a combination of various NLP tools and algorithms in preprocessing and postprocessing to select appropriate answers and filter undesirable questions. (2) A sequence-learning-based scheme to generate adequate QAPs via meta-sequence representations of sentences. A meta-sequence is a sequence of vectors comprising semantic and syntactic tags. we devise a scheme called MetaQA to learn meta sequences from training data to form pairs of a meta sequence for a declarative sentence and a corresponding interrogative sentence. The TP3 works well on unseen data, which is complemented by MetaQA. Both methods can generate well-formed and grammatically correct questions. Moreover, we present a novel approach to automatically generate adequate distractors for a given QAP. The method is a combination of part-of-speech tagging, named-entity tagging, semantic-role labeling, regular expressions, domain knowledge bases, word embeddings, word edit distance, WordNet, and other algorithms.	['Cheng Zhang']	2023-03-25	null	null	null	null	['semantic-role-labeling', 'part-of-speech-tagging', 'reading-comprehension', 'question-generation']	['natural-language-processing', 'natural-language-processing', 'natural-language-processing', 'natural-language-processing']	[ 6.25406802e-01 5.76566279e-01 6.45061910e-01 -6.78288758e-01 -1.48126864e+00 -9.07613218e-01 5.37376940e-01 4.83408660e-01 -5.28900981e-01 8.86337578e-01 5.74287236e-01 -5.38852632e-01 -1.47628665e-01 -1.07795548e+00 -6.21639431e-01 -5.10161296e-02 4.94716585e-01 7.86996841e-01 4.91503596e-01 -9.77787316e-01 6.88465476e-01 3.18232737e-02 -1.76494610e+00 9.00182307e-01 1.57809329e+00 7.08938062e-01 2.39234865e-01 1.02858293e+00 -9.45058882e-01 9.54365790e-01 -8.92623365e-01 -7.74650633e-01 -1.05774149e-01 -9.51621354e-01 -1.50184143e+00 -3.34342211e-01 3.94279420e-01 8.65841731e-02 2.57790446e-01 7.94171929e-01 7.11697459e-01 4.04259235e-01 7.15435326e-01 -8.47441852e-01 -8.52730155e-01 5.26990771e-01 4.51504976e-01 1.85759187e-01 1.07544327e+00 2.70427555e-01 1.28893030e+00 -9.49893594e-01 5.37737846e-01 1.43419874e+00 4.49145257e-01 1.01148105e+00 -1.06161702e+00 1.85573157e-02 -2.40872070e-01 4.98559326e-01 -7.51288116e-01 -2.61099279e-01 4.54042852e-01 -2.67672956e-01 1.22028649e+00 5.05693376e-01 3.63030851e-01 1.21718609e+00 2.88412184e-01 6.21977389e-01 1.21518791e+00 -9.21485245e-01 3.89841288e-01 8.74414742e-02 4.87440199e-01 7.27291405e-01 -3.87170345e-01 -1.53855845e-01 -3.89676362e-01 -3.55909795e-01 1.28832579e-01 -8.07355642e-01 -1.83746114e-01 3.48762661e-01 -1.06275868e+00 1.17397106e+00 -6.69661537e-02 4.53590989e-01 -3.59022826e-01 -2.39419803e-01 2.98489094e-01 6.11371338e-01 9.06707495e-02 1.26281011e+00 -8.45798314e-01 -5.07995375e-02 -6.01290643e-01 7.26649642e-01 1.09729373e+00 7.52223670e-01 6.77475393e-01 -1.82936490e-01 -1.14137542e+00 1.09611213e+00 2.03933239e-01 3.76607448e-01 9.33384836e-01 -1.02693355e+00 6.59805954e-01 8.44308853e-01 2.63385981e-01 -8.52584958e-01 -2.77262896e-01 1.11526057e-01 -1.27932698e-01 -1.76814005e-01 5.46391249e-01 -2.29902074e-01 -8.77519667e-01 1.88930285e+00 4.17020291e-01 -3.01197141e-01 3.46172124e-01 5.37855148e-01 1.25952423e+00 1.06879222e+00 1.76888287e-01 -8.81571844e-02 1.77649999e+00 -1.12319720e+00 -8.50899577e-01 -2.43621632e-01 7.56512523e-01 -6.38583422e-01 1.51320624e+00 8.91450867e-02 -1.48839188e+00 -7.42579758e-01 -7.17773974e-01 -4.16279674e-01 -6.59554482e-01 -2.74210483e-01 -8.63832831e-02 5.64560235e-01 -1.12850714e+00 4.26442057e-01 -2.10946379e-03 -1.85797289e-01 -7.53443688e-02 7.69946352e-02 -2.28901446e-01 -2.83136498e-02 -1.75252509e+00 1.33113015e+00 6.15452111e-01 -2.40291864e-01 -8.49863529e-01 -8.19908381e-01 -1.32254136e+00 1.29979635e-02 2.36483321e-01 -1.17964494e+00 1.49960577e+00 -1.02654600e+00 -1.76550829e+00 1.03134668e+00 -3.30680728e-01 -4.14879918e-01 -4.22888212e-02 -1.70043617e-01 -4.82324421e-01 2.96523929e-01 3.56499344e-01 6.95207536e-01 9.43705380e-01 -8.57723653e-01 -5.15408218e-01 -2.57707030e-01 8.71702731e-02 2.44492441e-01 1.34719625e-01 2.52106518e-01 2.65956908e-01 -5.37109733e-01 -1.03866428e-01 -4.07150835e-01 -4.08081770e-01 -6.53689027e-01 -2.96336651e-01 -8.21707189e-01 -2.18185261e-02 -1.11261821e+00 1.23177719e+00 -1.48673081e+00 1.47408009e-01 -5.53315543e-02 -5.36058247e-02 5.91652393e-01 -7.90218592e-01 5.99971712e-01 -9.52504128e-02 1.75516605e-01 -4.98397022e-01 1.29213184e-01 2.63060063e-01 1.08728223e-01 -2.91453600e-01 -5.00235796e-01 8.09531927e-01 1.19790626e+00 -1.24102867e+00 -5.19016445e-01 -3.34023461e-02 -7.25091472e-02 -6.83698058e-01 9.21101987e-01 -8.89711261e-01 3.26450288e-01 -3.68293285e-01 2.11654380e-01 4.53481972e-01 2.45602399e-01 -5.90683855e-02 1.10583268e-02 6.64172396e-02 1.10625803e+00 -7.81113565e-01 1.70142186e+00 -7.91376114e-01 1.13817833e-01 -2.94861555e-01 -8.83373201e-01 1.20226228e+00 3.28314900e-01 -3.02322388e-01 -9.19530332e-01 1.53625429e-01 5.10937035e-01 -7.24031106e-02 -1.16661394e+00 7.51441300e-01 -3.82213473e-01 -3.92005503e-01 3.99472773e-01 6.81131899e-01 -5.19757569e-01 5.74347496e-01 1.37298241e-01 1.40498912e+00 3.23935807e-01 3.57841611e-01 -1.08663544e-01 1.12879002e+00 2.97985286e-01 1.96571186e-01 9.55311179e-01 -3.36155891e-02 7.12922215e-01 4.44667727e-01 -1.52410761e-01 -9.38045025e-01 -1.12555516e+00 2.68478692e-01 1.36569142e+00 -5.79614043e-01 -3.42385799e-01 -1.13217962e+00 -1.04672885e+00 -3.32375199e-01 1.55033922e+00 -5.23068488e-01 -2.51836061e-01 -9.34836566e-01 -4.23193663e-01 6.72481239e-01 2.03562692e-01 1.60357505e-01 -1.65914381e+00 -4.17473316e-01 6.53101087e-01 -6.76565111e-01 -7.90939629e-01 -4.78696793e-01 -2.42726207e-02 -5.55680931e-01 -1.17793441e+00 -4.83401656e-01 -1.16459596e+00 4.92417604e-01 -4.01527613e-01 1.62048042e+00 5.50040603e-02 1.43834287e-02 4.18800920e-01 -7.91483164e-01 -3.04175377e-01 -1.14554214e+00 -7.23956078e-02 -3.93713832e-01 -3.94293666e-02 4.83762175e-01 -3.15275341e-01 -2.91072130e-01 -1.01836726e-01 -1.08762395e+00 -1.96872503e-01 5.42640388e-01 9.75599170e-01 4.82272685e-01 -5.41363657e-01 9.24046934e-01 -9.38284755e-01 1.35667062e+00 -5.75599313e-01 -3.21744233e-01 7.21979082e-01 -2.73372650e-01 4.88397688e-01 7.18847573e-01 -1.58955559e-01 -1.38419676e+00 -3.35212827e-01 -1.04111385e+00 3.61836761e-01 -3.82299483e-01 4.50856745e-01 -4.00307089e-01 3.34398568e-01 1.06473577e+00 5.28195262e-01 4.35446464e-02 -2.73638457e-01 8.74301136e-01 7.10257649e-01 6.05102539e-01 -7.60957241e-01 7.49729335e-01 -2.71791965e-01 -3.68875921e-01 -5.94221711e-01 -1.40479159e+00 -2.71109253e-01 -5.97986877e-01 -1.82015728e-02 1.12593031e+00 -2.70367354e-01 -3.88522148e-01 1.95636675e-01 -1.62974060e+00 -1.79138586e-01 -6.19335890e-01 -4.15006327e-03 -6.81262076e-01 4.43324119e-01 -4.83110130e-01 -6.17369413e-01 -7.76370943e-01 -7.78681278e-01 8.85040581e-01 3.66584659e-01 -4.31866437e-01 -9.82317865e-01 4.47515339e-01 9.40604627e-01 4.15690333e-01 8.00444782e-02 1.45638585e+00 -1.35764599e+00 -1.54598773e-01 1.41223863e-01 1.30254179e-01 8.18595588e-01 -9.98119265e-02 -2.03933328e-01 -6.94047093e-01 2.22622693e-01 2.15573415e-01 -6.04292631e-01 6.49959385e-01 -5.34508973e-02 8.90879214e-01 -7.54285276e-01 2.99986422e-01 -5.25394343e-02 1.12081087e+00 2.01276481e-01 1.01800370e+00 2.43023753e-01 3.79817903e-01 1.23568451e+00 7.39136279e-01 -5.48679270e-02 7.83432662e-01 3.63125652e-01 1.23723149e-01 5.37278771e-01 -1.36040568e-01 -5.99531889e-01 5.46963930e-01 1.02433836e+00 6.23600900e-01 -3.33221167e-01 -9.79052067e-01 9.29927111e-01 -1.38599861e+00 -1.03939176e+00 -6.36399627e-01 2.03080392e+00 1.38341510e+00 -3.29697579e-02 -1.23700112e-01 9.67209339e-02 6.22409284e-01 -4.64723632e-02 -4.66608666e-02 -1.03491533e+00 -1.41296864e-01 9.94640708e-01 -3.39402735e-01 7.30591059e-01 -6.68438077e-01 1.14926302e+00 5.81304598e+00 8.33896697e-01 -3.92060071e-01 3.00584495e-01 3.68530124e-01 5.45184255e-01 -9.51904535e-01 4.70770374e-02 -8.45026553e-01 3.31982046e-01 1.48428857e+00 1.96907092e-02 3.79592925e-02 5.26516020e-01 9.76999383e-03 -6.83531240e-02 -1.03088319e+00 4.15905386e-01 5.89035451e-01 -1.50195050e+00 4.71306920e-01 -7.51736164e-01 4.98183876e-01 -4.41424996e-01 -3.76345366e-01 7.96466470e-01 4.25153375e-01 -1.01151240e+00 5.76474786e-01 6.67824924e-01 4.33256745e-01 -5.75006485e-01 8.31625640e-01 4.37044442e-01 -6.41312659e-01 -1.37575850e-01 -4.67094630e-01 5.07887639e-02 3.57506007e-01 3.99893314e-01 -9.82951224e-01 6.31210208e-01 3.72486681e-01 -2.00424671e-01 -8.39344382e-01 7.47240305e-01 -8.35600972e-01 7.30047286e-01 1.08421631e-01 -4.86156523e-01 2.81377286e-01 -9.59104449e-02 5.99314630e-01 1.20053566e+00 3.12612623e-01 2.33467370e-01 -3.29158127e-01 1.11504614e+00 2.17194334e-02 4.87471879e-01 -2.12784111e-01 5.24603911e-02 5.91949165e-01 1.23623657e+00 2.51219515e-02 -4.15992528e-01 -1.90080106e-01 1.08726418e+00 5.50729275e-01 -5.51972575e-02 -4.17695493e-01 -8.31757963e-01 2.97064006e-01 -4.01566923e-02 1.45106345e-01 2.43805200e-01 -1.56498030e-01 -1.14116466e+00 -8.59848224e-04 -1.34088933e+00 7.02176094e-01 -1.07571113e+00 -1.70062697e+00 7.55166173e-01 -1.88767865e-01 -7.20319211e-01 -4.72736835e-01 -6.05645180e-01 -9.12557900e-01 1.10687459e+00 -1.51297176e+00 -8.50328445e-01 -1.29045352e-01 4.86886829e-01 8.76210511e-01 -1.59109324e-01 1.05751872e+00 -4.42971028e-02 -2.05520064e-01 5.09218097e-01 -2.98258781e-01 1.28897980e-01 6.33224726e-01 -1.56128705e+00 6.16977811e-01 6.97979689e-01 1.38003491e-02 5.24425685e-01 6.47714019e-01 -5.52770197e-01 -1.01980972e+00 -1.18143773e+00 1.87812781e+00 -8.77806246e-01 5.37264943e-01 -3.70028466e-01 -1.24318552e+00 3.16011578e-01 6.04475856e-01 -6.11900330e-01 9.59528983e-01 -1.89420804e-01 -2.60482162e-01 1.78230986e-01 -1.40614748e+00 5.60325444e-01 6.68956161e-01 -4.33961421e-01 -1.73616779e+00 4.99206752e-01 1.15022206e+00 -4.13306028e-01 -6.91237032e-01 2.76020914e-01 -9.52951536e-02 -8.66989672e-01 7.16256380e-01 -1.12594283e+00 7.51551688e-01 -3.56545180e-01 -3.27573046e-02 -1.52193558e+00 -2.13418722e-01 -6.53547764e-01 1.79551870e-01 1.46987760e+00 8.41270745e-01 -4.43727165e-01 2.32835069e-01 5.50218701e-01 -5.25896072e-01 -6.88487589e-01 -9.55263674e-01 -5.47993362e-01 4.22010928e-01 -2.94298798e-01 7.72234738e-01 6.80072963e-01 4.43859072e-03 9.14868116e-01 1.29062414e-01 -9.23211947e-02 1.92489460e-01 -2.29058564e-02 4.76655006e-01 -8.84057164e-01 -1.93530038e-01 -1.51792601e-01 1.27613321e-01 -9.77231443e-01 4.13824677e-01 -1.08221948e+00 3.37731272e-01 -1.77518106e+00 -2.18678072e-01 -1.47178009e-01 1.44422710e-01 2.38562673e-01 -7.13749528e-01 -2.60758877e-01 2.86147017e-02 -4.71371651e-01 -5.07042766e-01 8.10824752e-01 1.19488096e+00 5.82630746e-02 -1.61428779e-01 -1.00136973e-01 -7.62607753e-01 4.58642662e-01 9.57101524e-01 -6.66785598e-01 -3.00620109e-01 -4.42506880e-01 6.14011109e-01 3.02217513e-01 2.36848906e-01 -6.19014144e-01 -1.87560543e-01 -2.26728916e-01 5.80404438e-02 -5.53016603e-01 7.87131488e-02 -2.25903720e-01 -5.39545774e-01 3.74965519e-01 -7.37961113e-01 2.88486242e-01 4.52784896e-02 2.58478045e-01 -3.05464536e-01 -1.14442289e+00 6.89379752e-01 -6.00529134e-01 -5.41354477e-01 -2.84005284e-01 -7.19717562e-01 8.48969281e-01 7.38621056e-01 1.60114933e-02 -4.98172611e-01 -3.97109717e-01 -6.61557555e-01 4.65186208e-01 -1.45314068e-01 6.51077926e-01 7.80804813e-01 -1.22086668e+00 -1.23650944e+00 4.79218774e-02 2.52436101e-01 -2.17742845e-01 3.16320062e-01 3.80290717e-01 -5.90528369e-01 4.87802207e-01 -1.87328830e-01 -4.25818823e-02 -1.02944791e+00 2.89133757e-01 5.47176301e-01 -6.30747437e-01 5.77921122e-02 1.21136391e+00 -2.73588061e-01 -1.22755563e+00 -3.14046800e-01 -2.53971606e-01 -8.24611843e-01 2.67278254e-01 7.37312198e-01 2.31145218e-01 3.81475091e-01 -4.79860157e-01 -1.76143840e-01 9.12442580e-02 -2.45528631e-02 -3.61452162e-01 1.07108021e+00 -3.07993796e-02 -4.95403677e-01 1.56080514e-01 1.02052999e+00 7.01393038e-02 -3.91714245e-01 -8.29078481e-02 4.33163434e-01 -1.19003065e-01 -5.30031443e-01 -1.15073895e+00 -1.61956578e-01 8.49731922e-01 2.07941860e-01 3.01375121e-01 9.01886225e-01 1.57891065e-01 1.15971017e+00 4.58460301e-01 3.78721617e-02 -1.23118162e+00 5.41991174e-01 1.05949450e+00 1.25848985e+00 -9.64198351e-01 -7.98502564e-01 -2.73160309e-01 -7.23721504e-01 1.10194981e+00 8.94091308e-01 1.20070912e-01 7.67457634e-02 -4.71153855e-01 2.21560031e-01 -1.89575568e-01 -9.67582405e-01 -3.72717679e-01 4.87961769e-01 7.98151493e-01 4.25143242e-01 -2.76013106e-01 -8.88644695e-01 1.08186519e+00 -6.35027766e-01 -3.30802053e-01 6.00835443e-01 8.53776217e-01 -7.60553122e-01 -1.55912638e+00 -3.89292210e-01 6.53558552e-01 -4.15815264e-01 -4.72254246e-01 -5.31348944e-01 1.68158531e-01 2.10826203e-01 1.40499127e+00 -1.37344509e-01 -2.29241192e-01 6.69903219e-01 7.01492906e-01 5.71596324e-01 -1.24453390e+00 -1.28252161e+00 -7.64059246e-01 6.14065707e-01 -3.42435896e-01 -1.67595282e-01 -5.81258655e-01 -1.14915514e+00 2.38673568e-01 -2.53787041e-01 5.48826039e-01 5.04839420e-01 1.25772345e+00 4.64022577e-01 5.19588828e-01 6.59734249e-01 -7.37990364e-02 -9.82010126e-01 -1.24312449e+00 8.79580528e-02 6.90656126e-01 7.94804767e-02 -6.20978996e-02 -3.51846129e-01 -4.27717082e-02]	[11.493325233459473, 8.209541320800781]
f0fe872f-43c5-4c4b-abd0-f539055753cd	a-generic-approach-to-integrating-time-into	2305.06827	null	https://arxiv.org/abs/2305.06827v2	https://arxiv.org/pdf/2305.06827v2.pdf	A Generic Approach to Integrating Time into Spatial-Temporal Forecasting via Conditional Neural Fields	Self-awareness is the key capability of autonomous systems, e.g., autonomous driving network, which relies on highly efficient time series forecasting algorithm to enable the system to reason about the future state of the environment, as well as its effect on the system behavior as time progresses. Recently, a large number of forecasting algorithms using either convolutional neural networks or graph neural networks have been developed to exploit the complex temporal and spatial dependencies present in the time series. While these solutions have shown significant advantages over statistical approaches, one open question is to effectively incorporate the global information which represents the seasonality patterns via the time component of time series into the forecasting models to improve their accuracy. This paper presents a general approach to integrating the time component into forecasting models. The main idea is to employ conditional neural fields to represent the auxiliary features extracted from the time component to obtain the global information, which will be effectively combined with the local information extracted from autoregressive neural networks through a layer-wise gated fusion module. Extensive experiments on road traffic and cellular network traffic datasets prove the effectiveness of the proposed approach.	['Demin Lu', 'Zonghua Zhang', 'Duc-Thinh Ngo', 'Minh-Thanh Bui']	2023-05-11	null	null	null	null	['open-question']	['natural-language-processing']	[ 6.89497069e-02 -3.27032894e-01 1.29108531e-02 -5.91898859e-01 -4.94327992e-02 -1.19113766e-01 8.86339724e-01 3.11879851e-02 -6.70930594e-02 8.11888576e-01 1.35859445e-01 -3.59131098e-01 -2.04409495e-01 -1.00888467e+00 -4.18497056e-01 -1.03197646e+00 -2.23595232e-01 -2.31868133e-01 3.71513724e-01 -6.07981503e-01 3.08044851e-01 6.37927651e-01 -1.79376638e+00 -2.31537044e-01 1.00910175e+00 1.44112408e+00 2.34562501e-01 6.36827767e-01 -5.09113252e-01 1.09440732e+00 -4.55932975e-01 2.96263337e-01 -9.61213037e-02 -2.18040332e-01 -4.52339761e-02 -2.99894698e-02 -3.78578961e-01 1.29024029e-01 -6.22956812e-01 7.68166244e-01 5.01120351e-02 3.43845308e-01 3.74852955e-01 -1.18860853e+00 -1.00025915e-01 3.57458919e-01 -2.02842057e-01 5.55140674e-01 -1.66658863e-01 1.21286310e-01 3.94856244e-01 -4.17392701e-01 2.85035402e-01 1.00870478e+00 5.21888673e-01 -4.70207222e-02 -6.92731559e-01 -6.71211481e-01 3.24501008e-01 3.54447454e-01 -1.18015647e+00 -5.52781999e-01 1.34030020e+00 -5.75618386e-01 9.43916023e-01 5.83467707e-02 4.11694854e-01 5.86193383e-01 1.08903646e+00 2.50077993e-01 1.17646909e+00 -2.62441725e-01 1.82242185e-01 8.83050412e-02 2.56788850e-01 6.27760112e-01 -2.24236682e-01 2.10156083e-01 -4.75062042e-01 8.73728469e-02 3.01744878e-01 2.93851733e-01 -3.42855342e-02 6.48705512e-02 -8.49855244e-01 6.40446842e-01 6.18125379e-01 7.36448526e-01 -9.88696694e-01 3.32286209e-01 4.10615563e-01 1.83876067e-01 9.00206625e-01 -7.76380002e-02 -3.86289686e-01 -1.52557775e-01 -7.81360567e-01 -3.21743488e-02 5.32219827e-01 4.91666526e-01 1.06614399e+00 5.81098378e-01 1.99607442e-04 5.08923113e-01 3.13474149e-01 5.93898356e-01 5.13554156e-01 -4.53969985e-01 3.15860033e-01 6.06567442e-01 -1.62801787e-01 -1.53044260e+00 -6.89397454e-01 -6.97739005e-01 -9.82291818e-01 -2.57263966e-02 -3.15001383e-02 -4.25301611e-01 -1.14753318e+00 1.77867353e+00 3.21364135e-01 6.68865979e-01 1.38152808e-01 5.09175479e-01 6.09554946e-01 1.00962424e+00 2.90909141e-01 -2.63238966e-01 1.05985379e+00 -5.14569283e-01 -9.98824060e-01 -1.05747961e-01 7.74895430e-01 -5.51685035e-01 -3.07242405e-02 -1.17024682e-01 -4.68235046e-01 -8.33887100e-01 -1.10856056e+00 3.84252161e-01 -9.86032307e-01 1.65256128e-01 8.84587944e-01 4.30044889e-01 -1.11958838e+00 4.83110338e-01 -9.33284223e-01 -4.40868318e-01 2.89525818e-02 3.59083921e-01 -2.07840562e-01 1.02353610e-01 -1.73101175e+00 9.91181135e-01 5.87272465e-01 6.19415700e-01 -4.75284845e-01 -4.00815904e-01 -9.24236655e-01 1.40207157e-01 9.23982188e-02 -3.06157768e-01 8.51232529e-01 -8.68835986e-01 -1.48036563e+00 -7.01810047e-02 -5.15431941e-01 -7.81444490e-01 -4.07160334e-02 4.21924293e-01 -1.00770807e+00 -7.28368536e-02 -3.37571204e-02 2.11190671e-01 7.47140825e-01 -8.21169257e-01 -1.04272723e+00 -4.13799435e-01 -1.79834798e-01 -3.63123193e-02 -3.58295739e-01 -1.47334933e-01 -3.08843940e-01 -4.78045404e-01 2.61251032e-01 -9.97798324e-01 -5.65597951e-01 -7.05594420e-01 6.45077676e-02 -4.15016025e-01 1.22518909e+00 -8.67641747e-01 1.42798889e+00 -2.11657262e+00 -2.21858069e-01 4.82901156e-01 5.78462891e-02 2.16590628e-01 9.07213092e-02 7.17380345e-01 -1.76300351e-02 -4.30885196e-01 -3.01916376e-02 1.43191693e-02 -4.10399765e-01 4.47712064e-01 -5.88172555e-01 3.41030300e-01 4.17081565e-01 7.48555660e-01 -6.40332997e-01 -2.10574195e-01 5.61510623e-01 6.95473313e-01 1.12606838e-01 2.44708434e-02 -1.96457863e-01 7.12028384e-01 -8.90964985e-01 3.36680681e-01 5.55316210e-01 -9.04696584e-02 -1.76925701e-03 -6.46612942e-02 -5.11175215e-01 2.81109303e-01 -6.52031124e-01 1.21681118e+00 -6.47178471e-01 8.24488580e-01 -1.84359238e-01 -1.35620975e+00 1.36875772e+00 4.70744580e-01 6.87711775e-01 -1.02758062e+00 2.60839254e-01 9.01812166e-02 -1.10195510e-01 -4.09140229e-01 5.64798892e-01 8.73499364e-03 -6.45630062e-02 -9.05067250e-02 -8.22500959e-02 4.07355100e-01 5.68215773e-02 1.15350401e-02 9.83571470e-01 7.58080035e-02 1.09798007e-01 -1.10074207e-01 1.02601969e+00 8.48478973e-02 7.62458742e-01 3.28127325e-01 -6.15506135e-02 -1.79987848e-01 3.50603908e-01 -7.84171820e-01 -5.92860937e-01 -3.47083658e-01 4.35793139e-02 8.34946394e-01 3.16987447e-02 -1.61513671e-01 -2.26917461e-01 -2.83514321e-01 -1.91138268e-01 7.52392054e-01 -6.74845934e-01 -3.43172640e-01 -5.74122310e-01 -8.71938825e-01 4.29480858e-02 6.88338995e-01 6.00351989e-01 -8.70895445e-01 -5.02752125e-01 6.67722106e-01 -2.04953067e-02 -1.12167799e+00 1.26650482e-01 2.62241542e-01 -9.08401012e-01 -5.37799120e-01 -3.77057612e-01 -3.77845943e-01 4.27625656e-01 4.67946529e-01 6.53332114e-01 -4.49996553e-02 8.00605342e-02 2.74995029e-01 -1.83149904e-01 -5.90365231e-01 -3.54455143e-01 1.95308961e-02 -6.80959225e-02 5.32075047e-01 2.77056694e-01 -8.50156426e-01 -3.59012574e-01 1.99561581e-01 -7.83158600e-01 1.71974506e-02 6.14762008e-01 6.28070474e-01 3.58363360e-01 6.84174359e-01 8.81109238e-01 -7.52010763e-01 4.98194695e-01 -8.51497650e-01 -7.85511494e-01 1.54261008e-01 -6.98848724e-01 1.99122116e-01 8.00732017e-01 -1.59092382e-01 -1.31324983e+00 1.42296389e-01 -1.12888090e-01 -4.57622409e-01 3.30212037e-03 1.11422646e+00 4.05582376e-02 -5.73906526e-02 1.74812242e-01 6.45693302e-01 2.84936782e-02 -1.92070261e-01 2.75913298e-01 7.34889984e-01 5.79849482e-01 -2.05403477e-01 6.31688535e-01 6.35848582e-01 3.47701818e-01 -9.42491531e-01 -4.78654742e-01 -7.49166906e-01 -5.10795176e-01 -7.07879722e-01 8.07633758e-01 -1.01503468e+00 -5.29883742e-01 5.95042169e-01 -1.01811969e+00 1.09670386e-01 1.80781275e-01 6.77756011e-01 -3.61277312e-01 4.89595570e-02 -1.83760405e-01 -1.16234684e+00 -2.36714318e-01 -8.66421580e-01 7.34031618e-01 4.95393336e-01 2.08669797e-01 -1.36592281e+00 7.42060393e-02 -3.35662141e-02 8.20900083e-01 4.18872684e-01 8.95289838e-01 -7.00832605e-01 -6.21986270e-01 -7.49390960e-01 -1.82933629e-01 1.37741506e-01 3.01684231e-01 5.84745258e-02 -9.88665044e-01 4.72099660e-03 2.21256971e-01 4.68107373e-01 1.07328546e+00 6.55993879e-01 9.71833944e-01 2.09873971e-02 -6.56505644e-01 4.79985714e-01 1.28784978e+00 7.47669995e-01 6.64787352e-01 5.95282018e-02 7.26206422e-01 6.55730784e-01 5.36106229e-01 4.49445695e-01 7.13034630e-01 4.86584723e-01 4.54127759e-01 2.77470648e-02 1.11949392e-01 -2.71154428e-03 3.71216089e-01 1.26918709e+00 -1.25082031e-01 -1.96500719e-01 -9.31978285e-01 5.11005521e-01 -2.15428424e+00 -1.05127704e+00 -1.89461663e-01 2.01503134e+00 -1.32844299e-01 2.46926606e-01 -2.58693993e-01 6.67102635e-02 7.11175859e-01 4.87692952e-01 -6.03075981e-01 -4.66460735e-01 -1.70848787e-01 -1.37507394e-01 6.80830598e-01 1.32165924e-01 -1.23067844e+00 7.68108726e-01 5.87200737e+00 8.26653361e-01 -1.67245400e+00 -1.49578676e-01 6.82469666e-01 6.34661257e-01 -1.28883004e-01 2.05208644e-01 -7.07684219e-01 5.77551007e-01 1.64472246e+00 -2.79412687e-01 1.82700887e-01 6.58820629e-01 4.97194856e-01 -1.84792086e-01 -3.66530925e-01 5.10352850e-01 -1.04334272e-01 -1.43584752e+00 -3.12167317e-01 1.38420969e-01 6.21006727e-01 2.54063964e-01 -1.89962406e-02 4.90946531e-01 1.96520016e-01 -7.12231457e-01 2.74774581e-01 1.23287833e+00 3.27148587e-01 -8.41838241e-01 9.76456404e-01 4.30606276e-01 -1.62321675e+00 -2.93991953e-01 -1.43726230e-01 -2.84959435e-01 8.59748349e-02 7.39574850e-01 -9.47462380e-01 1.06851900e+00 5.72268426e-01 1.06364083e+00 -5.14434040e-01 8.58525455e-01 9.86752138e-02 8.90207589e-01 -3.94296646e-01 -7.48423859e-02 3.87114078e-01 -2.93197244e-01 5.17048717e-01 9.03264105e-01 5.50311029e-01 7.86496997e-02 1.31572366e-01 3.52102876e-01 3.59404683e-01 3.64572965e-02 -8.61534953e-01 -3.51799190e-01 8.93165693e-02 1.44832873e+00 -7.49183357e-01 -4.80205059e-01 -7.08034694e-01 2.31027633e-01 4.35809754e-02 5.49704731e-01 -7.10258126e-01 -4.38592702e-01 5.64181268e-01 -1.10980093e-01 4.75944757e-01 -5.86146295e-01 -1.32661209e-01 -8.96354854e-01 -2.79793330e-02 -2.59096771e-01 3.46498489e-01 -8.17062318e-01 -1.08948457e+00 8.47778022e-01 -3.34669530e-01 -1.46945632e+00 -7.13286877e-01 -3.42517734e-01 -9.10408974e-01 1.21128881e+00 -1.76783383e+00 -1.25522685e+00 -2.52903283e-01 5.77510655e-01 3.99993837e-01 -2.95556337e-01 4.97860491e-01 1.27418907e-02 -6.28445268e-01 -1.92182466e-01 3.94271851e-01 -6.52074516e-02 2.66420335e-01 -8.10789585e-01 2.19838768e-01 1.10056376e+00 -2.86059678e-01 3.38334471e-01 7.99248219e-01 -7.56744385e-01 -1.45618367e+00 -1.30605769e+00 9.66148913e-01 -1.87625922e-02 8.75259042e-01 7.25776255e-02 -1.05521047e+00 4.52048838e-01 1.46670431e-01 1.48806423e-01 3.61402035e-01 1.75104123e-02 -3.32412086e-02 -5.21240115e-01 -6.71020925e-01 4.05583918e-01 3.75516683e-01 -5.56857049e-01 -2.93952346e-01 3.49633060e-02 7.09284067e-01 -9.60818008e-02 -6.69829071e-01 6.60159826e-01 4.41924304e-01 -7.45412529e-01 4.49817181e-01 -3.99828911e-01 1.14237465e-01 -4.87723708e-01 -1.64581925e-01 -1.41365671e+00 -4.88871992e-01 -5.61106622e-01 -1.66806310e-01 1.15717554e+00 2.14850292e-01 -1.07629132e+00 5.35691202e-01 5.24727046e-01 -2.56387919e-01 -5.75055122e-01 -1.12411523e+00 -5.34208298e-01 -2.64711410e-01 -5.83963692e-01 8.01109850e-01 6.98238254e-01 -2.11572886e-01 1.85862035e-01 -4.90058869e-01 4.78002816e-01 1.90358713e-01 1.64099634e-01 6.64420724e-01 -1.46852207e+00 3.23093235e-01 -2.82435954e-01 -8.99200916e-01 -6.99322045e-01 3.88245195e-01 -7.16824114e-01 1.20123297e-01 -1.50143838e+00 -3.76865089e-01 -3.52990717e-01 -8.41727257e-01 2.79376715e-01 -1.11782506e-01 -1.58564866e-01 -4.98018116e-02 1.24669395e-01 -3.64505023e-01 7.82155812e-01 9.13763046e-01 -1.38248736e-02 -3.38854432e-01 3.08229715e-01 -4.10712659e-01 5.42271316e-01 8.93393338e-01 -2.21857935e-01 -5.42600632e-01 -4.27683629e-02 2.03079451e-03 5.63333035e-01 2.30704948e-01 -1.47059369e+00 7.15256155e-01 -2.15884387e-01 4.47746247e-01 -8.93973649e-01 4.20214504e-01 -1.05055904e+00 3.38116437e-01 3.29807580e-01 -5.71671501e-02 2.64519483e-01 3.71825814e-01 9.84826982e-01 -6.32285118e-01 4.28843766e-01 2.46080041e-01 2.25905284e-01 -1.16602254e+00 4.52548683e-01 -7.32318580e-01 -7.67283916e-01 1.06104422e+00 -1.77433982e-01 -2.78386295e-01 -6.43155158e-01 -5.34106255e-01 4.73245978e-01 -2.50174761e-01 6.90860331e-01 3.07722658e-01 -1.25168049e+00 -4.32690233e-01 4.73977774e-01 2.38596275e-01 -5.92183769e-01 7.02015519e-01 1.07624686e+00 -1.54095381e-01 9.22362447e-01 -2.10505962e-01 -6.37346804e-01 -8.15752447e-01 6.51981831e-01 3.42449486e-01 -2.84576654e-01 -4.16555375e-01 2.30346203e-01 1.67070761e-01 -3.45563851e-02 -1.96888939e-01 -3.64808768e-01 -7.64151990e-01 1.10279799e-01 3.58890027e-01 1.78806633e-01 2.86971480e-01 -1.05323064e+00 -3.82424712e-01 5.03342330e-01 2.37765554e-02 2.68530380e-02 1.44242358e+00 -4.23464000e-01 -2.35877231e-01 6.93619788e-01 1.12632263e+00 -3.65389854e-01 -1.11649752e+00 -5.84747851e-01 2.24948183e-01 -4.50962707e-02 5.92898190e-01 -4.78381157e-01 -1.17408848e+00 8.57338071e-01 5.75454950e-01 5.83011270e-01 1.50845313e+00 -2.45290905e-01 8.53993654e-01 2.54757673e-01 2.72518963e-01 -1.09463394e+00 -7.27164745e-01 5.78178287e-01 5.78158379e-01 -1.09404922e+00 -1.49413884e-01 -2.15351537e-01 -4.87560928e-01 1.37720478e+00 2.79240131e-01 -6.41112030e-02 1.13084030e+00 6.66847080e-02 1.67905048e-01 -1.75637141e-01 -1.17066085e+00 -5.12464523e-01 4.05869514e-01 3.82132262e-01 2.89427727e-01 2.10208386e-01 -2.36751035e-01 3.52526397e-01 -4.37900573e-02 1.73065856e-01 1.85156330e-01 7.85177767e-01 -4.43520963e-01 -8.27326298e-01 -2.37070546e-01 5.14660418e-01 -2.58628190e-01 3.24244887e-01 1.64554581e-01 5.08139193e-01 7.09695742e-03 1.31272388e+00 2.25864112e-01 -7.00714648e-01 1.61248326e-01 2.33711258e-01 -2.80959278e-01 -9.99539122e-02 -2.49771044e-01 -3.19982618e-02 4.82648201e-02 -6.01884425e-01 -6.49154961e-01 -5.79732478e-01 -1.18535841e+00 -1.51916400e-01 -2.88965940e-01 1.43505186e-01 1.14007604e+00 1.26415646e+00 6.49418056e-01 9.91420746e-01 1.06972182e+00 -8.02629173e-01 1.71107668e-02 -7.99863458e-01 -4.10830140e-01 -4.70982492e-02 5.62561870e-01 -7.29071438e-01 -2.85469502e-01 -7.13550076e-02]	[6.685892105102539, 2.7856557369232178]
d2222cd0-7e7d-4a1e-829a-a619e91d8d8a	multimodal-analogical-reasoning-over	2210.00312	null	https://arxiv.org/abs/2210.00312v4	https://arxiv.org/pdf/2210.00312v4.pdf	Multimodal Analogical Reasoning over Knowledge Graphs	Analogical reasoning is fundamental to human cognition and holds an important place in various fields. However, previous studies mainly focus on single-modal analogical reasoning and ignore taking advantage of structure knowledge. Notably, the research in cognitive psychology has demonstrated that information from multimodal sources always brings more powerful cognitive transfer than single modality sources. To this end, we introduce the new task of multimodal analogical reasoning over knowledge graphs, which requires multimodal reasoning ability with the help of background knowledge. Specifically, we construct a Multimodal Analogical Reasoning dataSet (MARS) and a multimodal knowledge graph MarKG. We evaluate with multimodal knowledge graph embedding and pre-trained Transformer baselines, illustrating the potential challenges of the proposed task. We further propose a novel model-agnostic Multimodal analogical reasoning framework with Transformer (MarT) motivated by the structure mapping theory, which can obtain better performance. Code and datasets are available in https://github.com/zjunlp/MKG_Analogy.	['Huajun Chen', 'Shumin Deng', 'Xiaozhuan Liang', 'Xiang Chen', 'Lei LI', 'Ningyu Zhang']	2022-10-01	null	null	null	null	['knowledge-graph-embedding']	['graphs']	[-3.78541276e-02 2.01023161e-01 -2.71222770e-01 -2.67703950e-01 -3.86046052e-01 -7.91598260e-01 9.05148327e-01 3.14023286e-01 -2.23705202e-01 4.75894213e-01 4.39511299e-01 -6.23814821e-01 -4.62541848e-01 -1.02928972e+00 -7.57626593e-01 -2.42031157e-01 4.46531087e-01 5.70573807e-01 -1.06236435e-01 -6.74735308e-01 2.74093449e-01 1.05260372e-01 -9.45041180e-01 4.83103096e-01 1.18116629e+00 9.02796328e-01 -2.69301049e-02 3.02962571e-01 -4.90502685e-01 1.12846899e+00 -1.69024140e-01 -1.35034871e+00 -3.27423811e-02 -5.16084015e-01 -1.24606836e+00 -4.74759758e-01 6.98435187e-01 -2.81746358e-01 -9.54457939e-01 1.16989505e+00 4.60103989e-01 3.67394179e-01 8.31790149e-01 -1.39493060e+00 -1.66729307e+00 8.26235235e-01 -2.85884231e-01 1.08202629e-01 7.84597635e-01 1.72874123e-01 1.15754664e+00 -8.52689326e-01 2.84730792e-01 1.54476404e+00 5.63031256e-01 3.71036351e-01 -1.19372725e+00 -6.09976828e-01 9.22019333e-02 1.00314987e+00 -1.13290298e+00 -1.06144205e-01 9.86176014e-01 -1.80275798e-01 6.43922508e-01 2.69462764e-01 7.63599694e-01 1.28419399e+00 -9.46248323e-02 9.61407483e-01 1.18833232e+00 -2.85346806e-01 -5.87056391e-03 -2.75068451e-02 2.76662767e-01 8.01032066e-01 5.38535081e-02 -9.40730944e-02 -4.59579229e-01 1.90007873e-02 7.91714370e-01 2.76457727e-01 -5.35735607e-01 -3.48772377e-01 -1.57516992e+00 8.64142895e-01 1.08266735e+00 1.94310337e-01 -3.24782699e-01 1.52454421e-01 4.29660708e-01 6.54496551e-01 -4.39829558e-01 6.59790218e-01 -8.09514076e-02 2.37734467e-01 -2.45794192e-01 1.27814293e-01 6.30741596e-01 1.34226656e+00 7.24398673e-01 -1.39077172e-01 -4.73336011e-01 8.89560938e-01 3.57932866e-01 7.32307851e-01 4.47984099e-01 -1.09157062e+00 5.72590828e-01 8.97341490e-01 -3.78077239e-01 -1.05587876e+00 -2.30802909e-01 -1.49551019e-01 -7.13041961e-01 -3.35581660e-01 4.12871450e-01 2.06921667e-01 -4.19870675e-01 1.94531965e+00 1.53368071e-01 1.51681984e-02 4.34208989e-01 1.04451156e+00 1.23430097e+00 4.50163871e-01 2.34832630e-01 2.78793633e-01 1.75394833e+00 -1.02816570e+00 -7.64185965e-01 -3.01980525e-01 2.47767150e-01 -6.58506453e-01 1.69028592e+00 2.86270101e-02 -1.04144537e+00 -4.34542865e-01 -8.96669209e-01 -5.83107829e-01 -7.35000432e-01 -1.34310365e-01 9.01248813e-01 3.36556375e-01 -9.15844321e-01 2.13296533e-01 -3.58306795e-01 -6.40980244e-01 4.03666586e-01 5.35901356e-03 -3.90249044e-01 -6.16139472e-01 -1.63005233e+00 1.20632279e+00 6.44164264e-01 1.24078892e-01 -4.75837678e-01 -6.67406797e-01 -9.49554920e-01 1.28692895e-01 4.96400952e-01 -1.35479486e+00 1.22097743e+00 -8.96457791e-01 -1.45063305e+00 6.67571723e-01 3.78636494e-02 2.21773423e-02 2.72077054e-01 -1.99965343e-01 -3.34877819e-01 3.12619150e-01 -6.41493918e-03 6.65764213e-01 5.07539332e-01 -1.13641572e+00 4.47582603e-02 -5.89716077e-01 8.32607388e-01 3.91051352e-01 -6.28260255e-01 -5.76059878e-01 -4.79508251e-01 -7.09449589e-01 -7.67499907e-03 -8.66544604e-01 2.37745941e-01 -6.70082718e-02 -5.22941709e-01 -2.86142081e-01 3.19251657e-01 -8.47686708e-01 9.00339365e-01 -1.96774828e+00 6.87859297e-01 1.66210428e-01 3.37143838e-01 -1.95909962e-01 -4.09213215e-01 7.98535645e-01 6.92810817e-03 -1.69432372e-01 -2.06713408e-01 8.10889229e-02 6.30567968e-01 2.37149850e-01 -5.52923679e-01 -1.52005479e-01 -2.35415599e-03 1.62945724e+00 -1.22280502e+00 -5.73196471e-01 1.63009278e-02 2.66672045e-01 -4.73379642e-01 2.98663467e-01 -3.01731676e-01 3.07770401e-01 -4.43274409e-01 8.24939549e-01 6.85781181e-01 -5.39572477e-01 3.61762077e-01 -8.55411768e-01 4.48256582e-01 -1.01142623e-01 -5.41329682e-01 2.19867134e+00 -4.76524770e-01 4.86839592e-01 -6.01787031e-01 -8.78506243e-01 5.88396847e-01 1.12615265e-01 -1.50263458e-01 -9.47127700e-01 3.63872379e-01 -8.48884359e-02 9.66310278e-02 -5.76714754e-01 6.12735212e-01 -3.49142581e-01 -5.19757043e-04 3.04271549e-01 3.11298162e-01 -2.00719073e-01 -5.20715527e-02 6.40724659e-01 9.73805010e-01 5.82340211e-02 3.14548999e-01 4.23986576e-02 5.29644728e-01 -1.19515106e-01 -1.17608547e-01 5.80835402e-01 -3.15559730e-02 5.35284802e-02 4.38882589e-01 8.83732513e-02 -6.05406284e-01 -1.60964608e+00 7.82102570e-02 1.07420814e+00 5.34075499e-01 -3.06460500e-01 -3.00258845e-01 -4.85480487e-01 1.30329266e-01 1.02838886e+00 -6.36968195e-01 -5.28418362e-01 -1.31796643e-01 -3.41055751e-01 7.03324199e-01 7.31203496e-01 8.49575937e-01 -9.75917757e-01 1.29633650e-01 -2.89278746e-01 -6.12767994e-01 -1.16110158e+00 -3.11105162e-01 -4.41695303e-01 -8.15749586e-01 -1.08627009e+00 -7.42139459e-01 -7.25304604e-01 4.83518034e-01 1.78235516e-01 1.24662244e+00 7.95201026e-03 4.43353038e-03 1.22098756e+00 -5.14111340e-01 -2.23641261e-01 -2.11654678e-01 -3.30392383e-02 -2.51306981e-01 -1.80319056e-01 5.01648962e-01 -7.33275414e-01 -6.69757485e-01 4.97435071e-02 -9.52265382e-01 2.56743640e-01 1.09155715e+00 7.44781554e-01 3.57685894e-01 -4.38846171e-01 7.74054646e-01 -4.79328781e-01 9.92850840e-01 -7.58885860e-01 -1.99472532e-01 6.57181144e-01 -2.12780476e-01 1.48436308e-01 6.49848342e-01 -4.84090269e-01 -1.18716216e+00 -4.81441587e-01 1.07723549e-01 -8.05176497e-01 -5.48572047e-03 8.88312936e-01 -3.60581845e-01 -1.22265831e-01 5.44947505e-01 3.69138539e-01 1.57982446e-02 -1.63579151e-01 1.09735382e+00 2.13913605e-01 6.93987846e-01 -1.08603621e+00 7.90254414e-01 3.54874283e-01 1.52581379e-01 -5.43606222e-01 -8.69220376e-01 4.75350954e-03 -4.02697980e-01 -1.87109783e-01 9.24569011e-01 -7.43181050e-01 -1.22537220e+00 -8.89630839e-02 -1.32634318e+00 -1.90683633e-01 -8.24184343e-02 5.72370708e-01 -5.83848178e-01 7.13471174e-01 -6.75347209e-01 -3.81561548e-01 -2.53974497e-01 -7.12326050e-01 5.97122014e-01 3.49372357e-01 -1.70851037e-01 -1.27320409e+00 9.31634083e-02 6.60310566e-01 4.25901502e-01 -6.08264767e-02 1.44217122e+00 -7.47322679e-01 -8.48951101e-01 -7.57927680e-03 -6.08130336e-01 -1.46443797e-02 -6.10577986e-02 -6.33192956e-01 -9.46101487e-01 2.13468820e-02 -3.66434008e-01 -7.62285471e-01 9.23353791e-01 -1.86867923e-01 1.07974041e+00 -1.85617015e-01 -1.07471466e-01 3.16785634e-01 1.30626166e+00 -6.00189455e-02 7.38701463e-01 2.79416978e-01 9.77713406e-01 5.90329528e-01 3.12330186e-01 1.22541666e-01 1.09033167e+00 4.30188477e-01 4.16653305e-01 1.76382244e-01 -1.10859036e-01 -4.73144054e-01 1.79021269e-01 9.78901803e-01 -2.64653206e-01 -1.37037367e-01 -1.02681208e+00 4.36605334e-01 -1.94527841e+00 -1.18090069e+00 7.48541206e-02 1.86824906e+00 9.64890778e-01 -4.69823778e-01 -2.62636058e-02 -2.28887871e-01 4.18410689e-01 -1.69564024e-01 -6.08267248e-01 -2.09336057e-01 -2.41587415e-01 -6.41520768e-02 7.53853424e-03 4.54992086e-01 -8.04578125e-01 9.78770852e-01 4.95256472e+00 7.87336171e-01 -6.03501439e-01 2.29659537e-03 -7.24131465e-02 2.07430631e-01 -8.45368803e-01 4.93460037e-02 -1.29668981e-01 1.90022469e-01 5.23125768e-01 -3.71108353e-01 8.95014524e-01 3.38581264e-01 -6.08374834e-01 2.30396867e-01 -1.58643699e+00 1.18956351e+00 2.81871885e-01 -1.05327129e+00 8.74631763e-01 -2.78463393e-01 3.57905000e-01 -3.31269115e-01 2.57125676e-01 7.74390817e-01 2.15039164e-01 -1.03746080e+00 3.73792231e-01 9.99610245e-01 4.24996078e-01 -5.78605473e-01 6.58348560e-01 -3.97590250e-02 -1.19046199e+00 -1.81004807e-01 -5.21008551e-01 2.56892797e-02 6.58710822e-02 8.41007829e-02 -3.72772902e-01 1.20516598e+00 5.07123828e-01 7.57880270e-01 -8.46309006e-01 7.79536545e-01 -5.47623694e-01 1.50654957e-01 1.62483275e-01 -1.10164195e-01 -7.96670541e-02 -3.22885573e-01 2.41145760e-01 9.73816872e-01 3.09693635e-01 4.27643389e-01 -3.98458429e-02 1.27611423e+00 -3.41620386e-01 2.95548677e-01 -7.95653224e-01 -3.90284300e-01 6.48394406e-01 1.18833685e+00 -2.37959281e-01 -3.06694835e-01 -9.54159498e-01 1.21581662e+00 6.35974646e-01 6.76601648e-01 -1.02295530e+00 -4.05117184e-01 3.60583514e-01 -3.32114428e-01 -4.53792140e-02 -1.18889928e-01 -1.38469428e-01 -1.49567735e+00 -5.96498102e-02 -8.18708301e-01 7.71237671e-01 -1.16357207e+00 -2.06446218e+00 3.51248384e-01 1.56254128e-01 -1.05432868e+00 4.28931378e-02 -1.02141106e+00 -7.04367518e-01 7.67941415e-01 -1.46495032e+00 -1.64498675e+00 -7.29032874e-01 1.04264712e+00 5.59068285e-02 -2.08963767e-01 8.45493197e-01 2.77836561e-01 -1.92022786e-01 8.80342126e-01 -1.76318601e-01 3.44448000e-01 8.27682853e-01 -1.31570268e+00 3.26422825e-02 2.22638518e-01 5.67667410e-02 1.00207889e+00 4.44291234e-01 -4.81581956e-01 -2.09657121e+00 -7.71152079e-01 4.53426898e-01 -7.54641175e-01 1.18563223e+00 1.77263543e-01 -1.19444418e+00 1.06412089e+00 8.50299180e-01 -6.30405128e-01 1.01902580e+00 2.90702790e-01 -7.58198023e-01 1.19146682e-01 -8.13490570e-01 1.10489690e+00 1.26847649e+00 -8.71293426e-01 -1.19552958e+00 3.56352448e-01 8.50864708e-01 -2.12301150e-01 -1.23421574e+00 3.15029413e-01 5.71576118e-01 -6.56987250e-01 1.30852544e+00 -8.94006073e-01 8.29326034e-01 -1.54084101e-01 -5.46754539e-01 -1.44947946e+00 -4.40065831e-01 -2.83295810e-02 -1.85530305e-01 1.04686320e+00 3.14507544e-01 -8.32414985e-01 2.91541904e-01 6.10583484e-01 -8.69667679e-02 -5.41828156e-01 -7.35698700e-01 -9.02180791e-01 4.03537840e-01 -1.20616257e-01 6.83042407e-01 1.55390275e+00 5.84868252e-01 8.65722835e-01 -1.05995119e-01 2.75043875e-01 6.16398454e-01 4.96653438e-01 8.44657719e-01 -1.15740001e+00 -4.36894417e-01 -6.78969622e-01 -4.92298871e-01 -9.94364321e-01 6.49878561e-01 -1.57660437e+00 -6.18950546e-01 -1.82602549e+00 5.30259609e-01 3.16244304e-01 -4.13282394e-01 6.18735433e-01 -3.89292300e-01 -6.49682656e-02 3.71519834e-01 8.21484625e-02 -6.50242090e-01 1.06569302e+00 1.67942846e+00 -3.89742941e-01 3.62534434e-01 -7.78363466e-01 -8.55487347e-01 5.44005215e-01 9.00937796e-01 3.19185287e-01 -9.41914022e-01 -5.71256816e-01 6.53505147e-01 1.98235765e-01 1.03420520e+00 -6.23156846e-01 4.05118436e-01 -2.05170915e-01 4.07992601e-01 -1.78260952e-01 5.97318530e-01 -9.79607224e-01 -6.30507991e-02 1.07016481e-01 -3.66376817e-01 5.38985193e-01 5.68139434e-01 6.97725415e-01 -2.87853062e-01 -8.44281353e-03 4.03099090e-01 -4.97237332e-02 -9.78260040e-01 2.73201257e-01 6.56347051e-02 3.04266989e-01 7.49073446e-01 9.58707556e-02 -8.34928691e-01 -6.20171189e-01 -7.29326725e-01 4.28090841e-01 3.92927766e-01 4.95652676e-01 8.63941610e-01 -1.90940106e+00 -5.34764171e-01 -3.77998263e-01 6.23028994e-01 -6.51768267e-01 5.83953917e-01 1.11091375e+00 -1.26595050e-01 3.77464652e-01 -5.64812124e-01 -2.36536950e-01 -7.56330848e-01 1.09178531e+00 3.49822491e-01 3.14729780e-01 -4.22571599e-01 5.98213673e-01 3.70405525e-01 -9.03677225e-01 -1.04773350e-01 -9.62261260e-02 -2.65760362e-01 -1.55769676e-01 2.42024258e-01 5.09347081e-01 -4.06611949e-01 -3.17560136e-01 -2.93957829e-01 6.39571488e-01 3.15486975e-02 -2.54617959e-01 5.65405965e-01 -1.05311699e-01 -4.16985333e-01 5.73373497e-01 9.91777599e-01 -8.11163932e-02 -3.72549146e-01 -6.15986466e-01 -1.28951162e-01 -5.11099815e-01 -3.43000084e-01 -9.84505057e-01 -8.03308070e-01 1.13294864e+00 2.74063796e-01 -2.55057868e-02 1.03765392e+00 3.83114964e-01 6.57136738e-01 9.96436894e-01 1.77230835e-01 -5.61498582e-01 5.60831010e-01 6.08380318e-01 1.31691718e+00 -1.37659228e+00 -1.15752779e-01 -3.96776378e-01 -8.70257914e-01 1.08122587e+00 8.88138831e-01 1.97430030e-01 3.60417664e-01 -5.80133796e-01 3.30837108e-02 -4.99015629e-01 -4.41142350e-01 -2.19492391e-01 6.71038151e-01 5.31927526e-01 5.52218199e-01 5.03215417e-02 -7.16950744e-02 9.06186640e-01 -3.82056832e-01 -4.82240170e-02 1.28710002e-01 6.71920955e-01 -9.22173038e-02 -7.67847002e-01 -3.45602602e-01 2.05943793e-01 2.90758789e-01 -5.08732021e-01 -8.13552320e-01 8.48644733e-01 -3.55142355e-01 7.44778395e-01 -3.08156163e-01 -4.02443975e-01 5.85189879e-01 2.65325814e-01 1.06247318e+00 2.34000105e-02 -1.87205210e-01 -6.59774423e-01 -6.50906712e-02 -5.67575574e-01 -3.56614083e-01 -2.29433566e-01 -1.28855419e+00 -7.76257455e-01 2.08855137e-01 7.03303814e-02 9.30018723e-02 8.54293764e-01 3.35273117e-01 5.95054805e-01 -1.76698834e-01 -6.61995590e-01 -4.58477885e-01 -8.39725077e-01 -2.94766217e-01 5.51724792e-01 -4.34884913e-02 -8.24166715e-01 -1.03448793e-01 -8.98292363e-02]	[10.659477233886719, 1.80960214138031]
25bcfc98-3bcc-4448-bdd6-626d7c2ffc76	convolutional-fine-grained-classification	2208.01997	null	https://arxiv.org/abs/2208.01997v1	https://arxiv.org/pdf/2208.01997v1.pdf	Convolutional Fine-Grained Classification with Self-Supervised Target Relation Regularization	Fine-grained visual classification can be addressed by deep representation learning under supervision of manually pre-defined targets (e.g., one-hot or the Hadamard codes). Such target coding schemes are less flexible to model inter-class correlation and are sensitive to sparse and imbalanced data distribution as well. In light of this, this paper introduces a novel target coding scheme -- dynamic target relation graphs (DTRG), which, as an auxiliary feature regularization, is a self-generated structural output to be mapped from input images. Specifically, online computation of class-level feature centers is designed to generate cross-category distance in the representation space, which can thus be depicted by a dynamic graph in a non-parametric manner. Explicitly minimizing intra-class feature variations anchored on those class-level centers can encourage learning of discriminative features. Moreover, owing to exploiting inter-class dependency, the proposed target graphs can alleviate data sparsity and imbalanceness in representation learning. Inspired by recent success of the mixup style data augmentation, this paper introduces randomness into soft construction of dynamic target relation graphs to further explore relation diversity of target classes. Experimental results can demonstrate the effectiveness of our method on a number of diverse benchmarks of multiple visual classification tasks, especially achieving the state-of-the-art performance on popular fine-grained object benchmarks and superior robustness against sparse and imbalanced data. Source codes are made publicly available at https://github.com/AkonLau/DTRG.	['Kui Jia', 'Ke Chen', 'KangJun Liu']	2022-08-03	null	null	null	null	['fine-grained-image-classification']	['computer-vision']	[ 2.91292936e-01 -5.65822646e-02 -5.21675408e-01 -4.87627774e-01 -6.94890738e-01 -5.94188869e-01 6.55110121e-01 9.89453867e-02 7.39196539e-02 5.22859693e-01 2.83199936e-01 1.26915902e-01 -4.23466414e-01 -8.00192475e-01 -8.89462948e-01 -8.96603823e-01 8.76124948e-03 2.03031838e-01 -1.41057268e-01 2.71286233e-03 3.25834095e-01 6.56292558e-01 -1.80396950e+00 5.96252441e-01 8.83042634e-01 1.29364979e+00 1.27480716e-01 1.98055461e-01 -7.99643844e-02 6.90524757e-01 -4.97549176e-01 -6.90526515e-02 4.05202657e-01 -2.47252464e-01 -3.42381686e-01 3.18393558e-01 5.73263407e-01 8.99957716e-02 -4.49794292e-01 1.09397113e+00 4.43575352e-01 1.03449121e-01 9.56717670e-01 -1.46056569e+00 -1.22597241e+00 4.87470120e-01 -9.05018687e-01 2.30084032e-01 -1.08889975e-02 2.23963484e-01 1.10936689e+00 -1.16438794e+00 3.92345905e-01 1.23503447e+00 4.72030550e-01 2.88516641e-01 -1.42892838e+00 -9.42836165e-01 4.05171901e-01 2.87641436e-01 -1.74928677e+00 -1.97679639e-01 1.05246902e+00 -7.01523662e-01 6.41782939e-01 4.15711254e-01 4.62465733e-01 1.11445355e+00 1.50207039e-02 5.72626352e-01 1.15455222e+00 -3.64348650e-01 1.36766851e-01 3.59317921e-02 2.03332692e-01 8.08348835e-01 3.89905721e-01 2.52319664e-01 -5.00058353e-01 -1.29198775e-01 7.78732121e-01 3.01592827e-01 -3.71449351e-01 -9.11488414e-01 -1.08701742e+00 1.10190785e+00 9.21513021e-01 3.54838282e-01 -1.42616719e-01 2.39295624e-02 3.87073785e-01 1.53286442e-01 4.11332875e-01 1.34228945e-01 -3.57030302e-01 3.50520670e-01 -6.52065814e-01 -4.03760634e-02 2.91602492e-01 9.28983092e-01 9.45934832e-01 1.85592771e-01 -7.79498339e-01 1.05598915e+00 3.09539497e-01 3.56763631e-01 7.92959571e-01 -4.78150368e-01 6.72226787e-01 8.78445804e-01 -2.97471255e-01 -1.26622355e+00 -3.92827868e-01 -8.80502582e-01 -1.41433489e+00 1.72420025e-01 3.53275895e-01 3.61677438e-01 -1.11776769e+00 1.74248803e+00 3.43953937e-01 2.39822388e-01 -2.57713288e-01 8.95052552e-01 9.09942329e-01 5.39538860e-01 -6.09612763e-02 -1.43099144e-01 1.34001744e+00 -9.41295564e-01 -4.37055647e-01 1.94694493e-02 5.84566414e-01 -5.59023738e-01 1.25277150e+00 1.85866654e-01 -5.70117474e-01 -7.87378848e-01 -8.97191167e-01 6.73870146e-02 -4.89628017e-01 3.02431315e-01 6.44142747e-01 5.30964732e-01 -7.25930333e-01 3.66070420e-01 -6.71906948e-01 8.74485523e-02 8.95368159e-01 2.56039023e-01 -5.03397048e-01 -2.88152337e-01 -8.88715625e-01 4.35650021e-01 3.91506612e-01 1.33159533e-01 -9.33173180e-01 -9.42654371e-01 -1.02736402e+00 2.14391485e-01 2.25769892e-01 -5.94297290e-01 4.73155290e-01 -9.61838603e-01 -1.09529138e+00 9.72558796e-01 6.17646091e-02 -2.70297974e-01 2.90722221e-01 6.02983199e-02 -2.90781856e-01 -2.39878073e-02 2.08481014e-01 4.70854700e-01 1.26438224e+00 -1.37326896e+00 -2.02012032e-01 -5.77669859e-01 -1.86088756e-01 3.16805989e-02 -4.98482972e-01 -3.04297537e-01 -3.00651968e-01 -1.22204876e+00 2.51750480e-02 -8.26396227e-01 -2.60271281e-01 1.43965006e-01 -6.17870331e-01 -3.01850438e-01 7.38856256e-01 -1.86697558e-01 1.23911071e+00 -2.21535277e+00 1.54518455e-01 3.73040974e-01 3.29840690e-01 3.20179045e-01 -3.62236083e-01 1.94416940e-01 -4.93577600e-01 1.20299667e-01 -2.67498791e-01 -1.90957878e-02 4.30034772e-02 4.16461714e-02 -3.31397593e-01 7.50908077e-01 4.25728917e-01 1.01856017e+00 -6.84946716e-01 -2.31185317e-01 3.80763650e-01 7.49239266e-01 -4.84609842e-01 2.00292319e-01 -4.72603291e-02 4.55508769e-01 -6.65970922e-01 8.12322199e-01 7.84698308e-01 -5.96206844e-01 -9.29286554e-02 -5.47279119e-01 1.14483565e-01 -5.75173236e-02 -1.16984928e+00 1.55547059e+00 -4.47954029e-01 2.93662757e-01 -4.03397709e-01 -1.37655044e+00 1.23806953e+00 -1.49077266e-01 2.66663462e-01 -8.61720741e-01 1.30545020e-01 -2.09058598e-02 3.76780592e-02 -8.71064812e-02 1.49884641e-01 1.07830413e-01 -1.48793757e-01 1.77125111e-01 1.00348808e-01 5.08194305e-02 1.11152172e-01 1.25189140e-01 7.67706573e-01 -1.32675439e-01 3.84077072e-01 -3.44774902e-01 4.76835161e-01 -3.67739171e-01 6.29103482e-01 7.95100272e-01 7.16706039e-03 7.85348535e-01 4.58866507e-01 -3.96570861e-01 -8.53675902e-01 -9.48671162e-01 -4.43718076e-01 1.18284392e+00 1.61586046e-01 -4.42230135e-01 -3.00686389e-01 -7.97906518e-01 3.81251514e-01 3.05000067e-01 -1.11354399e+00 -5.06332397e-01 -4.09969330e-01 -8.77523541e-01 2.36022502e-01 5.60852826e-01 2.40357101e-01 -9.67622936e-01 -1.70318991e-01 -2.74793003e-02 1.92381114e-01 -8.97972643e-01 -5.62801957e-01 4.51355278e-01 -7.18797445e-01 -1.07410252e+00 -8.36966217e-01 -7.84680605e-01 8.61150503e-01 4.24166471e-01 1.06350052e+00 1.16393343e-01 -6.84001505e-01 3.05381358e-01 -4.51119632e-01 -2.60409713e-02 1.21152624e-02 -4.44686227e-02 -1.16262019e-01 4.72928405e-01 1.75287127e-01 -7.12151587e-01 -8.15051556e-01 3.94121259e-01 -8.52788389e-01 -3.19264010e-02 5.81393361e-01 1.29112387e+00 9.85004842e-01 -5.36928475e-02 6.45844638e-01 -9.19267714e-01 3.59295517e-01 -6.25940025e-01 -6.40802205e-01 2.30021939e-01 -5.21229804e-01 1.57585576e-01 8.35903525e-01 -5.19383550e-01 -7.00568676e-01 8.92675966e-02 2.82870948e-01 -8.08169484e-01 -8.99348557e-02 3.76352787e-01 -2.63525397e-01 -2.35546768e-01 7.14815319e-01 2.71025002e-01 -1.61334708e-01 -4.31920350e-01 5.89553654e-01 4.37811285e-01 2.80972302e-01 -6.55613124e-01 1.03245234e+00 4.63756025e-01 8.34309757e-02 -4.55812663e-01 -1.09355724e+00 -4.72467452e-01 -5.75801551e-01 9.33061540e-02 5.08069575e-01 -1.12130284e+00 -4.40539151e-01 4.31058586e-01 -6.38160706e-01 -3.35664093e-01 -4.41630691e-01 2.38887593e-01 -4.81355071e-01 2.49624252e-01 -2.73144573e-01 -3.93961668e-01 -2.81112224e-01 -1.07422316e+00 1.26463032e+00 2.76635528e-01 1.89188957e-01 -8.73478651e-01 -1.65494889e-01 3.55183542e-01 2.84653664e-01 4.35034990e-01 9.51056302e-01 -5.05965114e-01 -6.72540545e-01 -4.26534042e-02 -5.83994448e-01 4.39566284e-01 4.14893717e-01 -7.04616755e-02 -9.11076367e-01 -5.49699605e-01 -2.84029305e-01 -5.54882288e-01 1.15688133e+00 3.97863865e-01 1.86353683e+00 -3.74252081e-01 -3.64487410e-01 9.22928095e-01 1.56731641e+00 -2.06876740e-01 4.82351720e-01 2.22699404e-01 1.09812236e+00 4.45220143e-01 6.83526397e-01 8.98235202e-01 3.91228944e-01 8.84545088e-01 5.33240020e-01 -1.25572503e-01 -4.87505704e-01 -2.85977572e-01 2.36299727e-02 5.87574244e-01 4.13810275e-02 -1.53553367e-01 -6.69797719e-01 5.32789111e-01 -1.68152010e+00 -8.22951496e-01 -4.05015275e-02 2.13060760e+00 8.59651744e-01 -1.35892600e-01 -3.89252454e-02 2.20364973e-01 8.49693298e-01 4.34406966e-01 -6.94733381e-01 1.51046282e-02 -2.96926081e-01 2.53430068e-01 4.88330781e-01 1.33211032e-01 -1.12600100e+00 8.44048142e-01 4.18869543e+00 1.21419919e+00 -1.28335416e+00 5.73058575e-02 9.04638052e-01 -6.05260246e-02 -4.04573768e-01 -3.77122968e-01 -8.94385993e-01 6.55719578e-01 4.32921052e-01 -7.75570571e-02 3.17851275e-01 8.08929861e-01 1.26843620e-02 2.87189364e-01 -8.89643013e-01 1.24932754e+00 1.08868808e-01 -1.59967637e+00 2.85601407e-01 -4.68315696e-03 9.47102010e-01 -9.87006947e-02 4.14402008e-01 3.20509076e-01 1.23585701e-01 -1.30427301e+00 6.16316020e-01 3.24056864e-01 1.18476188e+00 -8.29681933e-01 3.66712719e-01 3.87184061e-02 -1.41626096e+00 -2.93877095e-01 -5.99086940e-01 2.34805703e-01 -2.53463089e-01 8.26670587e-01 -5.19787848e-01 5.23914099e-01 7.64371455e-01 9.96587813e-01 -8.18766713e-01 1.01730311e+00 -1.21119693e-01 4.76182818e-01 -1.39798680e-02 2.01064542e-01 1.52162939e-01 1.18714720e-02 2.48875231e-01 1.14594615e+00 2.45654091e-01 -1.61460787e-02 2.94989973e-01 7.36862361e-01 -1.68006167e-01 7.93979466e-02 -6.35118842e-01 1.01299822e-01 4.25935328e-01 1.39993834e+00 -7.30578065e-01 -1.04920037e-01 -2.63326913e-01 7.23762214e-01 7.19661295e-01 3.78656119e-01 -8.32960784e-01 -3.00669223e-01 7.47336745e-01 1.13309763e-01 7.36927688e-01 6.63008913e-03 -5.08117199e-01 -1.32831895e+00 2.31836006e-01 -1.03668797e+00 6.04037344e-01 -3.44738573e-01 -1.59695053e+00 7.30111837e-01 -2.32256025e-01 -1.53352451e+00 6.10023625e-02 -5.26586831e-01 -3.38718146e-01 7.17258453e-01 -1.66212022e+00 -1.36163175e+00 -6.64738536e-01 1.09838080e+00 3.88536185e-01 -2.74940670e-01 7.34707892e-01 3.88379842e-01 -5.87244034e-01 1.05190980e+00 2.01185733e-01 6.94867224e-02 6.50996983e-01 -1.00402653e+00 5.96251488e-02 6.00183070e-01 3.55376005e-01 4.70521718e-01 3.67263883e-01 -4.06061798e-01 -1.36776936e+00 -1.49614716e+00 2.36412153e-01 -2.93820292e-01 7.23376453e-01 -6.43339872e-01 -1.12153375e+00 5.23744762e-01 -3.46163154e-01 9.52385187e-01 6.85361087e-01 -3.65762822e-02 -9.56158638e-01 -3.43184292e-01 -1.07195568e+00 2.01923683e-01 1.25059319e+00 -5.86268783e-01 -2.93637395e-01 5.54380238e-01 7.39374161e-01 -2.91512966e-01 -8.97895813e-01 6.33161068e-01 4.24971998e-01 -9.25323963e-01 1.17399132e+00 -6.04087830e-01 3.79657596e-01 -4.62013006e-01 -3.08246344e-01 -1.34391558e+00 -7.09876478e-01 -2.04168081e-01 -2.94625044e-01 1.38293588e+00 9.49407890e-02 -6.36943460e-01 8.10904622e-01 1.67843699e-01 -2.49906749e-01 -1.06505406e+00 -9.77841914e-01 -8.82800698e-01 3.74284126e-02 -1.74097404e-01 5.63880920e-01 1.05599260e+00 -4.50962454e-01 1.35238618e-01 -4.85744774e-01 2.53785968e-01 7.25083470e-01 6.11916780e-01 6.65001154e-01 -1.10810792e+00 -3.22734892e-01 -5.99118769e-01 -7.57234156e-01 -8.68890584e-01 3.02626848e-01 -1.28570998e+00 -1.77837625e-01 -1.22683859e+00 2.71966517e-01 -9.59524870e-01 -5.95022857e-01 6.45664513e-01 -1.81945994e-01 7.10102320e-01 2.73310184e-01 2.36612409e-01 -5.79992771e-01 1.07056069e+00 1.45044780e+00 -3.83534670e-01 -2.21154485e-02 -8.83036759e-03 -1.05000222e+00 4.61259872e-01 6.34248316e-01 -6.21731937e-01 -4.76701885e-01 -2.17554256e-01 -1.05442144e-01 -2.04653248e-01 5.08097947e-01 -8.46737206e-01 -7.20005184e-02 -1.41051143e-01 6.23004079e-01 -4.16515470e-01 2.64291495e-01 -7.38751650e-01 1.02123365e-01 2.65025944e-01 -3.65852863e-01 -2.90051907e-01 1.45216867e-01 7.55371988e-01 -4.39502448e-01 8.83190855e-02 1.10266423e+00 9.08537433e-02 -6.74901664e-01 6.47994757e-01 2.41136864e-01 2.87433982e-01 1.12452328e+00 -3.20154518e-01 -6.18233502e-01 -7.28944167e-02 -4.76186723e-01 4.92596030e-02 2.46668279e-01 5.31224906e-01 6.61812425e-01 -1.74577558e+00 -6.96899235e-01 4.65762675e-01 5.07528245e-01 6.92693070e-02 5.94743371e-01 6.66383624e-01 5.85111715e-02 4.05871183e-01 -4.36060160e-01 -8.18219066e-01 -1.01290405e+00 6.77180767e-01 1.95079491e-01 -4.13386106e-01 -6.82075500e-01 1.07911491e+00 8.85041595e-01 -4.32907760e-01 2.50235796e-01 -3.21249902e-01 -2.73877889e-01 1.46980092e-01 5.93605995e-01 1.27157038e-02 1.70928359e-01 -7.47475028e-01 -5.04869640e-01 8.39875460e-01 -2.30149388e-01 6.38914227e-01 1.43871331e+00 -1.89151932e-02 7.57716596e-02 3.51989955e-01 1.51392043e+00 -1.29421223e-02 -1.50669014e+00 -3.75980616e-01 -1.75099581e-01 -8.93469810e-01 -3.19131948e-02 -6.53980494e-01 -1.39277458e+00 7.71530390e-01 8.17084372e-01 2.06925318e-01 1.22163117e+00 2.22027242e-01 3.72941375e-01 -6.85839802e-02 3.76702636e-01 -5.82756639e-01 4.45462197e-01 3.75767678e-01 1.29758418e+00 -1.37561333e+00 -3.52863818e-02 -5.24403632e-01 -5.76174378e-01 9.82485294e-01 7.15557158e-01 -3.31047326e-01 8.04878712e-01 9.34453830e-02 -1.14727132e-01 -2.27727026e-01 -6.41000807e-01 -1.76683947e-01 7.44408786e-01 7.76195109e-01 3.87059391e-01 2.76402473e-01 -3.76844555e-02 6.46115243e-01 -1.11745767e-01 -1.57159373e-01 1.75120622e-01 5.77643871e-01 -1.30500406e-01 -9.87774372e-01 -3.36884141e-01 7.32001841e-01 -2.02346787e-01 -1.84150502e-01 1.74256414e-02 6.16309166e-01 2.01863334e-01 6.25607252e-01 1.04118250e-01 -3.53147864e-01 3.17132115e-01 -4.79278773e-01 4.86773074e-01 -7.50529826e-01 -4.67347443e-01 -1.61396071e-01 -2.92313576e-01 -7.07148850e-01 -3.86501610e-01 -4.14674640e-01 -9.71884727e-01 1.42636731e-01 -3.01137388e-01 -1.37793750e-01 2.84583211e-01 7.28467286e-01 5.55074930e-01 6.31572306e-01 9.01856065e-01 -9.59965765e-01 -6.32493615e-01 -8.60073328e-01 -6.92533493e-01 5.29755712e-01 4.44464058e-01 -1.09227276e+00 -4.54598248e-01 -1.46668315e-01]	[9.631376266479492, 2.1402688026428223]
3b19b719-6d92-4193-8ac5-b4fc28d710a0	netket-3-machine-learning-toolbox-for-many	2112.10526	null	https://arxiv.org/abs/2112.10526v2	https://arxiv.org/pdf/2112.10526v2.pdf	NetKet 3: Machine Learning Toolbox for Many-Body Quantum Systems	We introduce version 3 of NetKet, the machine learning toolbox for many-body quantum physics. NetKet is built around neural-network quantum states and provides efficient algorithms for their evaluation and optimization. This new version is built on top of JAX, a differentiable programming and accelerated linear algebra framework for the Python programming language. The most significant new feature is the possibility to define arbitrary neural network ans\"atze in pure Python code using the concise notation of machine-learning frameworks, which allows for just-in-time compilation as well as the implicit generation of gradients thanks to automatic differentiation. NetKet 3 also comes with support for GPU and TPU accelerators, advanced support for discrete symmetry groups, chunking to scale up to thousands of degrees of freedom, drivers for quantum dynamics applications, and improved modularity, allowing users to use only parts of the toolbox as a foundation for their own code.	['Giuseppe Carleo', 'Nikita Astrakhantsev', 'Vladimir Vargas-Calderon', 'Jannes Nys', 'Gabriel Pescia', 'Clemens Giuliani', 'Christopher Roth', 'Dian Wu', 'Attila Szabó', 'Damian Hofmann', 'Filippo Vicentini']	2021-12-20	null	null	null	null	['quantum-state-tomography', 'variational-monte-carlo']	['medical', 'miscellaneous']	[-5.74423909e-01 -4.36039641e-02 7.70332366e-02 -3.31207603e-01 -1.17741622e-01 -4.78671074e-01 4.60767388e-01 1.33685648e-01 -5.25203228e-01 8.37628007e-01 -3.53394747e-01 -5.93827069e-01 1.40515780e-02 -1.23261023e+00 -4.11087126e-01 -8.34415138e-01 -5.45624495e-01 4.28797781e-01 1.50225744e-01 -7.61285543e-01 9.52559188e-02 6.55478597e-01 -1.43086374e+00 -1.01454221e-02 7.62499332e-01 8.31308424e-01 -2.27844402e-01 7.71912456e-01 5.22456095e-02 6.79775357e-01 1.32264599e-01 -4.10999119e-01 4.59055007e-01 -3.46138746e-01 -7.21902490e-01 -7.44641364e-01 3.20604920e-01 -2.71153837e-01 -6.82582498e-01 9.63800251e-01 6.83459938e-01 4.10896778e-01 3.56472373e-01 -8.71592164e-01 -3.61777872e-01 3.18265975e-01 2.25270331e-01 2.54901171e-01 -1.61349494e-02 6.97719097e-01 9.70796704e-01 -7.15436757e-01 4.69230741e-01 1.01021838e+00 8.64530385e-01 5.90931058e-01 -1.45435548e+00 -3.82065028e-01 -7.05990136e-01 2.23540023e-01 -1.21305668e+00 -2.05057561e-01 2.06918448e-01 -4.01100963e-01 1.56305408e+00 4.57139909e-01 9.17170048e-01 6.84421778e-01 4.49578881e-01 1.83716699e-01 1.03526986e+00 -5.11950850e-01 4.69270617e-01 7.89685398e-02 6.07772052e-01 9.20501232e-01 -5.36384583e-02 5.94612896e-01 -3.59308481e-01 -4.54952508e-01 9.31539476e-01 -1.88959956e-01 1.90673620e-01 -5.05818307e-01 -1.35165250e+00 1.08208227e+00 6.60230100e-01 9.78963152e-02 -3.07118982e-01 4.56824124e-01 7.86230624e-01 2.22102866e-01 1.12780273e-01 5.64222813e-01 -4.23397392e-01 -3.78472865e-01 -5.88438630e-01 7.04793513e-01 1.07827389e+00 3.94468695e-01 1.13660443e+00 3.67472529e-01 -5.57734780e-02 4.68199641e-01 9.89737138e-02 4.08674210e-01 2.19967395e-01 -1.28021777e+00 -2.25680187e-01 2.50096023e-01 -5.35586402e-02 -6.37340844e-01 -8.74300957e-01 -5.88465393e-01 -1.03614223e+00 6.26560688e-01 3.28157216e-01 -2.90426612e-01 -3.73557627e-01 1.48453152e+00 5.72967529e-01 -5.97591996e-02 6.12899102e-02 8.17888319e-01 7.20611513e-01 6.89275086e-01 -1.49403438e-01 -6.57154396e-02 1.20243299e+00 -9.66293931e-01 -3.98942381e-01 4.29509699e-01 1.03638887e+00 -3.66547763e-01 8.99175465e-01 3.45103770e-01 -1.11218786e+00 -3.35812181e-01 -8.74565065e-01 -4.34359819e-01 -7.57070243e-01 -7.21038803e-02 1.52576649e+00 7.89044559e-01 -1.32287395e+00 1.49713206e+00 -1.12502623e+00 -7.22317547e-02 -5.88936312e-03 7.70111918e-01 -2.53474087e-01 5.57133794e-01 -1.46886325e+00 1.15474784e+00 9.44352150e-01 -6.75207227e-02 -5.13360977e-01 -1.09887815e+00 -8.94387841e-01 1.05208494e-01 -1.15894675e-01 -1.06201565e+00 1.17697012e+00 -6.58572614e-01 -1.94973660e+00 7.41254628e-01 1.81979731e-01 -6.71799898e-01 3.10609221e-01 1.53042004e-01 -1.77584618e-01 -3.97763308e-03 1.63864791e-02 5.73107958e-01 3.11347753e-01 -7.56371990e-02 1.90960079e-01 -1.94934443e-01 3.54855627e-01 8.06873068e-02 -1.31223962e-01 -1.45212011e-02 -2.74771755e-03 -1.48645416e-01 -2.42561042e-01 -9.26642060e-01 -6.08618081e-01 -4.35795076e-02 -2.50918448e-01 -2.43825838e-01 4.70095724e-01 -6.06385827e-01 9.17086422e-01 -1.87349069e+00 1.04817428e-01 3.15792352e-01 2.45661199e-01 1.08330719e-01 1.69253692e-01 6.47727668e-01 -6.53362274e-01 -2.77282864e-01 -1.01097308e-01 -4.36531082e-02 3.88952762e-01 2.17087507e-01 -2.29914099e-01 5.08020997e-01 1.14279248e-01 8.34914744e-01 -1.05614007e+00 -1.60170067e-02 5.74905872e-01 5.42128563e-01 -8.51547837e-01 -2.43245736e-01 -2.58432448e-01 7.31198490e-01 -3.41981590e-01 2.37391815e-01 7.70431459e-01 -3.74875277e-01 -1.45397084e-02 -1.75728455e-01 -8.73809338e-01 6.34824157e-01 -1.37777996e+00 1.71929169e+00 -4.75589991e-01 1.58708587e-01 1.72964439e-01 -9.17523623e-01 4.54841226e-01 2.00946406e-01 5.08534670e-01 -6.32351756e-01 1.28943086e-01 3.77715737e-01 2.27242872e-01 -2.51366924e-02 5.51500440e-01 -2.81485260e-01 6.69993982e-02 5.02311051e-01 3.24719667e-01 -4.14779246e-01 6.62044525e-01 3.31668913e-01 9.48869586e-01 3.51094604e-01 2.51153558e-01 -7.09273219e-01 5.90547860e-01 2.74682790e-01 1.50683418e-01 7.24195004e-01 1.04393005e-01 -3.26493196e-02 4.03366327e-01 -7.75366008e-01 -1.40007210e+00 -1.10873687e+00 -7.71534085e-01 1.30751443e+00 -2.60688990e-01 -8.61965477e-01 -6.76331460e-01 1.64726779e-01 -1.24408817e-02 6.90138757e-01 -3.63422066e-01 -5.19943535e-02 -5.30727684e-01 -1.14785182e+00 4.79486257e-01 6.91805258e-02 5.17673373e-01 -1.22787869e+00 -3.56125772e-01 1.82548419e-01 5.65491438e-01 -5.89048207e-01 1.26869917e-01 5.51118731e-01 -9.06401813e-01 -7.80203700e-01 -2.65593737e-01 -3.08268130e-01 1.88239843e-01 -3.23873401e-01 1.25091076e+00 2.18828574e-01 -6.48679435e-01 6.74658641e-02 1.12919852e-01 -7.55648986e-02 -6.49813533e-01 1.02858841e-01 4.56158340e-01 -5.25082648e-01 -3.57457101e-02 -8.59820604e-01 -5.01340985e-01 -1.88834131e-01 -7.48962760e-01 2.43673086e-01 -1.55171841e-01 1.06270623e+00 2.70465583e-01 -9.90711972e-02 -1.23295141e-03 -3.63624662e-01 5.66329181e-01 -2.37635657e-01 -1.11771679e+00 -3.18368912e-01 -5.07849157e-01 4.00887042e-01 9.41390812e-01 2.47096553e-01 -5.08683622e-01 -1.06605202e-01 -6.92983389e-01 -2.02704556e-02 8.56097415e-02 5.20249248e-01 4.00203079e-01 -6.07234299e-01 9.44286346e-01 3.06535542e-01 2.60581583e-01 -3.25527996e-01 5.87056220e-01 3.25897187e-01 4.22214866e-01 -8.20944965e-01 5.88191688e-01 3.55935961e-01 5.28160572e-01 -1.08241320e+00 -6.08451366e-01 -2.63289988e-01 -9.08469915e-01 9.24278200e-02 1.09169996e+00 -8.87958765e-01 -1.36355245e+00 6.08843327e-01 -1.09809470e+00 -5.40605426e-01 -3.04990739e-01 5.01113236e-01 -5.31317711e-01 5.37263930e-01 -1.02107167e+00 -5.89384735e-01 -6.92854226e-01 -1.29629886e+00 5.86719096e-01 3.69160026e-01 3.83922737e-03 -1.25640953e+00 3.41948330e-01 -2.19538480e-01 6.91523254e-01 2.37065345e-01 8.49264920e-01 -4.49577928e-01 -6.15865529e-01 -2.15854198e-01 -4.72996533e-02 5.30983865e-01 -5.02021074e-01 1.82427764e-01 -7.24534869e-01 -3.95030141e-01 -1.74338281e-01 -4.05690610e-01 7.59554505e-01 1.76277757e-01 1.22741735e+00 -1.88244611e-01 -1.09996669e-01 1.00456631e+00 1.25262868e+00 -3.13110888e-01 6.11996651e-01 2.14896232e-01 6.46855533e-01 2.01850280e-01 -6.44467473e-02 4.63786095e-01 1.22762144e-01 8.54488134e-01 4.83146250e-01 -1.95972636e-01 5.05909264e-01 3.60446543e-01 4.10581976e-01 1.00248253e+00 -4.55303729e-01 7.74578512e-01 -8.14722359e-01 -2.08038032e-01 -1.67537665e+00 -1.19661248e+00 -8.68809462e-01 2.46863174e+00 8.88544858e-01 -9.45704356e-02 2.84925252e-01 -1.85084984e-01 3.19908828e-01 -1.34882301e-01 -4.88264233e-01 -9.90368068e-01 2.12473586e-01 1.00119543e+00 7.42362440e-01 6.35096073e-01 -1.11174023e+00 1.08891606e+00 7.60496235e+00 9.06116545e-01 -1.45530343e+00 3.38930964e-01 1.39621884e-01 -8.26734453e-02 2.69448925e-02 3.76305312e-01 -7.11427808e-01 4.31663483e-01 1.28915119e+00 -2.66406327e-01 9.44997609e-01 9.33303595e-01 3.60805452e-01 -2.27136724e-02 -8.86785507e-01 8.22083890e-01 -7.23879158e-01 -1.96723068e+00 -4.26814765e-01 2.17848465e-01 4.34469223e-01 8.63280594e-01 -3.35573733e-01 5.73907495e-01 3.30956191e-01 -1.05995393e+00 4.68917608e-01 3.03532004e-01 7.07170069e-01 -8.85983348e-01 4.58831042e-01 4.86889511e-01 -7.94910431e-01 1.60249010e-01 -7.04043984e-01 -6.70231879e-01 1.31172910e-01 8.31794798e-01 -4.95886743e-01 5.90846658e-01 4.99061286e-01 6.58884287e-01 -3.64137053e-01 8.71554136e-01 -1.17448039e-01 4.57566053e-01 -7.02934742e-01 -1.79223210e-01 3.57611865e-01 -9.01487648e-01 4.79796767e-01 1.14882135e+00 1.24115422e-01 -3.47425565e-02 1.86174840e-01 1.23260307e+00 5.02691567e-01 6.98973835e-02 -4.14130032e-01 -1.71268299e-01 3.37396413e-02 1.64010870e+00 -6.41221583e-01 -4.30780411e-01 -2.17670888e-01 7.73798645e-01 4.56840336e-01 -7.75350258e-02 -9.16819394e-01 -6.94361448e-01 8.73354733e-01 -9.09637138e-02 6.39664531e-02 -8.63943040e-01 -2.22031713e-01 -1.44178104e+00 -4.14593428e-01 -7.06733108e-01 1.49093464e-01 -7.99690008e-01 -1.13958097e+00 3.85925084e-01 -1.50576532e-01 -6.21138394e-01 -3.32891345e-01 -1.27941644e+00 -9.43997562e-01 1.21693206e+00 -9.33594227e-01 -7.82766998e-01 2.51385178e-02 6.12583399e-01 -5.79390407e-01 -1.78431347e-02 1.47184563e+00 3.26361001e-01 -7.15372384e-01 1.73965365e-01 5.47786117e-01 1.42662063e-01 2.25534096e-01 -1.57375193e+00 6.94471061e-01 4.53944057e-01 -1.99970007e-01 1.15757847e+00 8.45124483e-01 -2.95526534e-01 -1.66899168e+00 -6.39797568e-01 5.52711725e-01 -3.07502538e-01 1.27130651e+00 -5.95961452e-01 -7.45211244e-01 6.34681225e-01 2.13510022e-01 1.06213354e-01 7.62986660e-01 4.42247003e-01 -1.88345507e-01 -8.93127248e-02 -9.17432606e-01 6.51814699e-01 6.94055378e-01 -6.53834462e-01 -2.13202506e-01 1.09870636e+00 4.60097581e-01 -8.21494401e-01 -1.13573718e+00 1.00668363e-01 4.24192131e-01 -1.39276683e+00 1.28108490e+00 -4.52946872e-01 1.42286047e-01 -3.78790617e-01 1.89146638e-01 -1.03359103e+00 -3.99607360e-01 -1.02780759e+00 -4.40085083e-02 4.00350869e-01 1.36379763e-01 -1.05892193e+00 5.61061263e-01 5.33828855e-01 -4.07258004e-01 -5.19870341e-01 -1.19028127e+00 -6.21050537e-01 5.07886171e-01 -6.03787303e-01 3.14571977e-01 8.69325757e-01 3.70546281e-01 3.45753968e-01 -1.84806779e-01 -4.83848825e-02 6.15713894e-01 3.33374366e-02 8.17133784e-01 -1.03701973e+00 -9.79462504e-01 -4.69959736e-01 -6.07912958e-01 -7.49461114e-01 2.63470829e-01 -1.47590160e+00 -4.90724772e-01 -1.03438246e+00 7.30499700e-02 -4.50857759e-01 2.36286297e-02 6.96757257e-01 3.21989208e-01 6.04640365e-01 6.01813085e-02 5.81753887e-02 -1.90187067e-01 6.04506195e-01 1.07489908e+00 2.51514673e-01 -2.66882449e-01 -4.58150841e-02 -2.23639429e-01 6.97696686e-01 7.80185580e-01 -3.28173012e-01 1.54643700e-01 -2.01297607e-02 5.60457945e-01 -1.46083891e-01 9.88784373e-01 -1.39815211e+00 -1.31145582e-01 1.11585163e-01 2.67831355e-01 -4.46614474e-01 5.12965322e-01 -1.42275810e-01 1.78936630e-01 8.29122126e-01 9.45983157e-02 2.02647924e-01 4.12585855e-01 -1.37232155e-01 1.68093666e-01 -3.43074411e-01 1.08584237e+00 -4.85346466e-01 -5.72755396e-01 3.73198211e-01 -3.20376068e-01 -2.10506544e-01 8.19724381e-01 -2.86220835e-04 -3.11809570e-01 -1.39544070e-01 -1.08932197e+00 2.02810969e-02 5.81961632e-01 -2.37803429e-01 -3.31325471e-01 -1.21674597e+00 -4.21144098e-01 2.93493569e-01 -2.00089931e-01 -2.19108790e-01 4.39732164e-01 1.11224926e+00 -1.24949360e+00 6.34949684e-01 -3.55503976e-01 -5.71078062e-01 -8.42340648e-01 3.96179885e-01 8.43309760e-01 -4.13567901e-01 -6.95068598e-01 4.72983390e-01 -1.85368285e-01 -1.02595186e+00 -3.89964163e-01 -3.08354825e-01 3.24340731e-01 -4.41961706e-01 6.21744037e-01 3.72951984e-01 4.01717305e-01 -4.88079101e-01 -3.87279779e-01 1.63194656e-01 2.92614967e-01 -2.97916103e-02 1.46330070e+00 3.95711243e-01 -9.93177891e-01 3.52673590e-01 1.03226459e+00 -7.16320351e-02 -7.50608921e-01 2.08286032e-01 -2.81107664e-01 2.99101353e-01 1.65969431e-01 -7.78582692e-01 -4.64399219e-01 1.14689028e+00 5.80757678e-01 1.79369345e-01 4.30371165e-01 -3.93320918e-01 6.57531857e-01 8.62122416e-01 5.88748276e-01 -1.00117671e+00 -3.55276108e-01 1.34138763e+00 5.11961639e-01 -8.26063633e-01 2.45300531e-01 -1.60157681e-01 1.78359076e-02 1.59752333e+00 3.49352181e-01 -5.01139939e-01 7.06283212e-01 2.11050704e-01 -6.34453520e-02 -3.08441192e-01 -4.52082217e-01 -1.44441336e-01 9.57515910e-02 4.54856634e-01 8.95227015e-01 1.26781553e-01 -4.54064041e-01 2.93685496e-01 -7.06642926e-01 2.30981428e-02 6.13080382e-01 6.16793275e-01 -2.81643033e-01 -1.57533360e+00 -3.42288285e-01 2.78446198e-01 -3.37860286e-01 -5.38791537e-01 3.98990273e-01 6.89168751e-01 1.11800723e-01 2.36681432e-01 8.61231238e-02 -2.39070614e-05 -1.88839406e-01 2.37125754e-01 7.03423619e-01 -6.80908144e-01 -9.89515126e-01 -4.80137199e-01 -7.66000059e-03 -7.62391746e-01 5.50450161e-02 -2.99243450e-01 -1.58155429e+00 -7.41908312e-01 -1.93833262e-01 5.15368640e-01 9.52777326e-01 9.80603278e-01 6.14726901e-01 4.53228444e-01 1.30755499e-01 -1.51275706e+00 -8.57164741e-01 -8.02679658e-01 -8.49571526e-01 -6.25676811e-02 1.02061175e-01 -5.44054627e-01 -2.47300208e-01 -6.14084959e-01]	[5.468382358551025, 4.978702068328857]
9a932829-21aa-4bf4-8028-bcb0d7477418	large-scale-pre-training-for-person-re	2203.16533	null	https://arxiv.org/abs/2203.16533v2	https://arxiv.org/pdf/2203.16533v2.pdf	Large-Scale Pre-training for Person Re-identification with Noisy Labels	This paper aims to address the problem of pre-training for person re-identification (Re-ID) with noisy labels. To setup the pre-training task, we apply a simple online multi-object tracking system on raw videos of an existing unlabeled Re-ID dataset "LUPerson" nd build the Noisy Labeled variant called "LUPerson-NL". Since theses ID labels automatically derived from tracklets inevitably contain noises, we develop a large-scale Pre-training framework utilizing Noisy Labels (PNL), which consists of three learning modules: supervised Re-ID learning, prototype-based contrastive learning, and label-guided contrastive learning. In principle, joint learning of these three modules not only clusters similar examples to one prototype, but also rectifies noisy labels based on the prototype assignment. We demonstrate that learning directly from raw videos is a promising alternative for pre-training, which utilizes spatial and temporal correlations as weak supervision. This simple pre-training task provides a scalable way to learn SOTA Re-ID representations from scratch on "LUPerson-NL" without bells and whistles. For example, by applying on the same supervised Re-ID method MGN, our pre-trained model improves the mAP over the unsupervised pre-training counterpart by 5.7%, 2.2%, 2.3% on CUHK03, DukeMTMC, and MSMT17 respectively. Under the small-scale or few-shot setting, the performance gain is even more significant, suggesting a better transferability of the learned representation. Code is available at https://github.com/DengpanFu/LUPerson-NL	['Dong Chen', 'Fang Wen', 'Houqiang Li', 'Lei Zhang', 'Lu Yuan', 'Jianmin Bao', 'Hao Yang', 'Dongdong Chen', 'Dengpan Fu']	2022-03-30	null	http://openaccess.thecvf.com//content/CVPR2022/html/Fu_Large-Scale_Pre-Training_for_Person_Re-Identification_With_Noisy_Labels_CVPR_2022_paper.html	http://openaccess.thecvf.com//content/CVPR2022/papers/Fu_Large-Scale_Pre-Training_for_Person_Re-Identification_With_Noisy_Labels_CVPR_2022_paper.pdf	cvpr-2022-1	['online-multi-object-tracking', 'unsupervised-pre-training']	['computer-vision', 'methodology']	[-8.28403607e-02 -3.10389042e-01 -1.53582171e-01 -3.38613749e-01 -9.64187324e-01 -4.82304513e-01 5.62869668e-01 -2.22300649e-01 -6.35757148e-01 8.06452096e-01 2.44194075e-01 2.27653250e-01 2.70895641e-02 -3.94377589e-01 -7.90266037e-01 -6.00177586e-01 1.52638763e-01 5.90688050e-01 3.01590860e-02 2.63867807e-02 -3.86567235e-01 3.59570533e-01 -1.71152866e+00 1.51261002e-01 5.80148757e-01 7.49628127e-01 1.63038224e-01 6.67042851e-01 1.88720912e-01 7.16588676e-01 -4.37856466e-01 -4.16897982e-01 4.03451502e-01 -4.33859378e-01 -7.45357156e-01 1.93464920e-01 9.67676640e-01 -3.29880893e-01 -7.13240027e-01 1.23864496e+00 6.29744649e-01 5.32843530e-01 4.70473617e-01 -1.17169023e+00 -8.14205170e-01 5.02326846e-01 -5.09275913e-01 3.33914012e-01 1.86302021e-01 5.71298674e-02 8.43218923e-01 -9.61952507e-01 6.90688610e-01 1.23689353e+00 1.08843470e+00 1.12405252e+00 -1.31317937e+00 -9.70964432e-01 2.82278717e-01 3.77347052e-01 -1.55831420e+00 -4.34498578e-01 4.44012880e-01 -5.53375721e-01 4.64489728e-01 2.60482281e-01 3.68691117e-01 1.37538278e+00 -4.95084673e-01 7.99544096e-01 1.20201552e+00 -3.32699418e-01 -1.25140429e-01 3.06454241e-01 6.05429888e-01 5.52847564e-01 2.19322383e-01 4.91051674e-01 -4.20016140e-01 -7.70138428e-02 3.99828225e-01 4.90861058e-01 -3.15651089e-01 1.60017181e-02 -1.20525181e+00 5.48775673e-01 5.36524236e-01 2.33450785e-01 1.43417314e-01 2.23758906e-01 3.93308997e-01 2.02191442e-01 4.89953518e-01 5.89491390e-02 -2.76086390e-01 3.36782932e-02 -1.01580143e+00 1.86659515e-01 4.58285987e-01 1.06143713e+00 8.81781220e-01 -1.99946299e-01 -5.62668502e-01 8.87668610e-01 1.23517282e-01 9.71915007e-01 5.29983878e-01 -8.46924186e-01 2.82687217e-01 1.61237791e-01 4.24966574e-01 -4.95738953e-01 -5.94908595e-01 -5.86847186e-01 -8.06653559e-01 -3.55803967e-02 6.05487466e-01 -4.40299749e-01 -1.20047581e+00 2.02029157e+00 3.66662681e-01 8.05978060e-01 -6.04107194e-02 1.02615428e+00 9.21654701e-01 3.96236300e-01 1.20730065e-01 -1.15606003e-01 1.31986845e+00 -1.40008271e+00 -6.85009181e-01 -4.14981283e-02 8.25649619e-01 -5.56975305e-01 7.50774741e-01 5.13058230e-02 -5.91115654e-01 -1.01498759e+00 -7.83238411e-01 7.58758560e-02 -3.04667324e-01 6.37720168e-01 2.26487339e-01 6.15867674e-01 -1.04506612e+00 7.74536669e-01 -7.03827143e-01 -6.10629082e-01 4.39374745e-01 3.23211700e-01 -4.66358602e-01 -4.26984876e-01 -9.39122200e-01 6.99966192e-01 2.38812238e-01 6.69116303e-02 -1.03543270e+00 -7.58400977e-01 -6.63988173e-01 -7.74297118e-02 4.65012491e-01 -5.21399140e-01 1.10588384e+00 -8.06015849e-01 -1.26429057e+00 9.61804986e-01 -3.92007768e-01 -4.13761377e-01 5.77157915e-01 -4.45221841e-01 -7.22720921e-01 -1.06408127e-01 4.15345550e-01 8.04301262e-01 8.38641167e-01 -1.31905997e+00 -7.43830204e-01 -3.28688562e-01 -1.44324541e-01 5.79907112e-02 -2.38734409e-01 1.10424064e-01 -6.23711824e-01 -8.83481622e-01 -2.77390927e-01 -1.45036530e+00 8.30524601e-03 -1.96074218e-01 -5.39760649e-01 -3.47197324e-01 7.76789367e-01 -7.15213418e-01 9.12372649e-01 -2.25025654e+00 -3.32043990e-02 4.70216088e-02 2.61526883e-01 4.15439814e-01 -3.97184283e-01 1.16225697e-01 -1.53633147e-01 -8.01875666e-02 1.24505192e-01 -8.27650607e-01 -3.20014693e-02 5.46150729e-02 -6.95102587e-02 6.04747355e-01 7.98725337e-03 9.18628693e-01 -1.02339709e+00 -2.82767922e-01 1.57216474e-01 3.84056509e-01 -3.88623923e-01 3.55391800e-01 -3.08455341e-02 9.41726506e-01 4.52229083e-02 6.10835850e-01 7.16835797e-01 -4.43236232e-01 5.29635400e-02 -3.76858085e-01 -1.18657082e-01 -1.45202592e-01 -1.38070107e+00 1.67659569e+00 -1.05091549e-01 6.01270497e-01 -2.87307322e-01 -9.36426997e-01 7.13655710e-01 2.57367015e-01 5.72562218e-01 -5.51336229e-01 5.23374081e-02 6.95727300e-04 -2.69852728e-01 -3.16624492e-01 4.82950956e-01 -5.81577010e-02 -7.31780678e-02 4.15422529e-01 4.34892654e-01 9.18940425e-01 3.01632613e-01 2.75874943e-01 1.04529727e+00 2.49655858e-01 -1.78610310e-02 -4.93396968e-02 4.98586893e-01 -3.34422700e-02 8.26227844e-01 1.10867810e+00 -4.89062339e-01 7.77264118e-01 -2.35583261e-01 -2.44859278e-01 -9.31769848e-01 -1.08014941e+00 -2.09639788e-01 1.31713843e+00 4.30991173e-01 -4.57550853e-01 -5.02900720e-01 -8.80055189e-01 8.88142586e-02 3.07970583e-01 -7.46974409e-01 -4.66859341e-02 -5.67865312e-01 -6.87601388e-01 6.38843596e-01 6.18958712e-01 6.98594928e-01 -7.69206762e-01 2.27533326e-01 -5.23927957e-02 -3.26147735e-01 -1.26513875e+00 -8.89687777e-01 8.72843489e-02 -4.28861082e-01 -1.06322706e+00 -1.05110800e+00 -8.27781558e-01 6.94181740e-01 6.80730164e-01 7.81385243e-01 2.06896171e-01 -2.46494725e-01 6.82374120e-01 -4.53419030e-01 -4.27393652e-02 -1.15744047e-01 -8.82587582e-02 5.52359879e-01 2.06809744e-01 5.28829753e-01 -2.41338149e-01 -5.01008809e-01 5.67557156e-01 -3.42562079e-01 -1.76977560e-01 4.35249835e-01 1.09974730e+00 6.92096174e-01 -1.88440561e-01 6.40181541e-01 -8.28795314e-01 -8.34283698e-03 -4.92816091e-01 -4.90007043e-01 4.49513137e-01 -4.28672463e-01 2.27858475e-03 5.75329304e-01 -7.86424994e-01 -1.12419438e+00 2.79377371e-01 -1.24264754e-01 -8.16981375e-01 -4.29382384e-01 -7.74505083e-03 -6.09475300e-02 -1.18735410e-01 6.62448227e-01 5.36387973e-02 -2.78904408e-01 -8.73899102e-01 5.39796948e-01 7.04346657e-01 9.31704640e-01 -5.12659311e-01 1.07505012e+00 7.35675097e-01 -3.35717678e-01 -5.65588951e-01 -1.35268402e+00 -1.01304746e+00 -8.30018997e-01 -1.73581168e-01 1.04444587e+00 -1.38705158e+00 -7.80373096e-01 4.13755059e-01 -9.10710514e-01 -4.80783492e-01 -4.23659027e-01 6.20778739e-01 -1.53003156e-01 4.37673092e-01 -7.04197109e-01 -6.03344440e-01 -2.69038081e-01 -7.76187897e-01 9.58607495e-01 4.96853530e-01 7.30528533e-02 -7.61720002e-01 3.22259039e-01 5.43397188e-01 1.69760033e-01 -1.18991196e-01 6.87557533e-02 -9.21119571e-01 -4.92923528e-01 -1.33113936e-01 -4.55575258e-01 3.94101024e-01 2.32089773e-01 -5.87765872e-01 -1.25717402e+00 -6.89117134e-01 -3.58902425e-01 -4.30606723e-01 1.05064106e+00 1.99090138e-01 8.90428245e-01 -1.48134559e-01 -6.24910712e-01 7.75773525e-01 1.39155257e+00 -2.03180805e-01 2.68963784e-01 2.87217766e-01 1.07745183e+00 2.84525901e-01 7.07979918e-01 4.73281622e-01 4.69720602e-01 1.01777720e+00 -7.42465854e-02 8.24760832e-03 -5.95423579e-01 -3.37663740e-01 4.89974111e-01 5.63755631e-01 -2.06586495e-01 -4.25723977e-02 -6.37456298e-01 5.35865128e-01 -2.05915022e+00 -1.28049195e+00 -4.63034734e-02 2.18568134e+00 6.14335299e-01 -3.13629776e-01 5.25026858e-01 -2.47409523e-01 1.24452829e+00 -1.04620092e-01 -5.87900877e-01 5.56244671e-01 3.11555248e-03 -9.74433050e-02 7.79553592e-01 3.65046859e-01 -1.57913566e+00 9.84586120e-01 5.16517496e+00 8.23093593e-01 -9.76419687e-01 6.01205528e-01 3.37010175e-01 -2.01505080e-01 2.32244730e-01 -2.01227695e-01 -1.56527257e+00 7.64619529e-01 1.11287916e+00 4.22845855e-02 4.65269655e-01 8.39622736e-01 1.87210292e-01 1.22153111e-01 -1.12683308e+00 1.45471585e+00 3.16506058e-01 -1.29678774e+00 -3.15750301e-01 -6.64350837e-02 1.01681340e+00 2.91587383e-01 -3.68883274e-02 6.49843276e-01 5.68554342e-01 -7.70154953e-01 7.33975768e-01 6.54378057e-01 8.67911816e-01 -5.03619790e-01 8.68607938e-01 2.55584717e-01 -1.42232919e+00 -2.77412325e-01 -5.13467312e-01 2.53876150e-01 3.46825778e-01 4.62716579e-01 -4.96073306e-01 6.74695611e-01 1.11531687e+00 9.94226575e-01 -8.14181626e-01 1.18039024e+00 -1.49085581e-01 5.81602991e-01 -2.93931097e-01 4.25192833e-01 -1.37947842e-01 -4.11933381e-03 4.67925757e-01 1.36852705e+00 4.12995011e-01 6.13588765e-02 6.21510744e-01 5.61642647e-01 -3.73079598e-01 -2.59444416e-01 -3.66402090e-01 2.05939665e-01 4.31702554e-01 1.33947527e+00 -4.90684003e-01 -6.78835869e-01 -6.55378878e-01 1.27483594e+00 5.49312472e-01 4.66844499e-01 -1.18703806e+00 -4.43618484e-02 6.08308077e-01 -2.31135380e-03 5.33290863e-01 -7.05940649e-02 1.81796685e-01 -1.45267379e+00 -1.50735512e-01 -7.58465767e-01 6.28717005e-01 -5.40354192e-01 -1.57623291e+00 6.81812048e-01 4.48972220e-03 -1.47924280e+00 -1.09430462e-01 -4.91873652e-01 -4.63123441e-01 6.21411562e-01 -1.51309872e+00 -1.47947335e+00 -7.64423788e-01 8.01746249e-01 3.92294198e-01 -2.83936054e-01 6.53061271e-01 7.83990979e-01 -9.46876109e-01 1.00251842e+00 4.77931470e-01 4.44661975e-01 1.18287182e+00 -1.17334938e+00 1.41576216e-01 9.78775501e-01 4.78058279e-01 4.00282770e-01 4.06436086e-01 -8.12087119e-01 -8.91959548e-01 -1.65587819e+00 7.04531789e-01 -6.18299723e-01 5.87890506e-01 -4.79547977e-01 -8.11811388e-01 1.03536212e+00 -1.73542589e-01 5.03916323e-01 7.06513107e-01 2.59837121e-01 -4.50464547e-01 -2.30121315e-01 -1.04945838e+00 4.20474350e-01 1.47563708e+00 -5.82866251e-01 -4.80654985e-01 6.62019491e-01 6.80786788e-01 -3.23818237e-01 -8.95862579e-01 2.60765940e-01 3.61850768e-01 -6.11674905e-01 1.15278578e+00 -5.46171248e-01 -3.81204724e-01 -5.07869840e-01 -1.08704716e-01 -1.17145789e+00 -6.21573865e-01 -5.74991167e-01 -8.55414942e-02 1.63092458e+00 8.66464525e-02 -4.58535492e-01 8.78439665e-01 4.81278270e-01 -5.78863099e-02 -1.17042765e-01 -8.12447727e-01 -1.21208251e+00 -2.29268089e-01 -3.67299289e-01 3.34126264e-01 9.93872285e-01 -2.77965873e-01 2.98681468e-01 -9.19969022e-01 4.57365096e-01 9.61528897e-01 -5.29299080e-02 9.73612309e-01 -1.34016812e+00 -3.59425992e-01 1.77334920e-01 -2.69300699e-01 -1.27053511e+00 3.20259869e-01 -1.17099357e+00 2.50205308e-01 -1.07601810e+00 5.62547505e-01 -7.78061450e-01 -5.36607027e-01 6.14028037e-01 -3.51729989e-01 5.43990791e-01 5.44754446e-01 6.84089184e-01 -1.12765932e+00 4.38386440e-01 9.00635302e-01 -3.05749357e-01 -2.02337369e-01 2.01515943e-01 -5.19400358e-01 5.47133386e-01 6.04459941e-01 -7.81108856e-01 -9.03858468e-02 -2.31908545e-01 -4.76817191e-01 -3.69990766e-01 6.89831495e-01 -1.26718390e+00 5.07348657e-01 2.15717673e-01 4.58452880e-01 -5.94221950e-01 4.48862731e-01 -6.24631286e-01 3.22172672e-01 3.21905226e-01 -3.80316377e-01 -1.37885928e-01 1.22495651e-01 8.81119847e-01 5.17342240e-02 -3.75791162e-01 8.80615056e-01 -1.10911146e-01 -1.07348907e+00 4.53316361e-01 -3.37956436e-02 1.12837449e-01 8.27906609e-01 6.65612891e-02 -5.33888698e-01 -2.30842292e-01 -1.17145836e+00 2.93043911e-01 3.72135669e-01 3.63007575e-01 3.09475005e-01 -1.61735940e+00 -6.71017826e-01 1.00046024e-01 1.87902659e-01 -2.97003150e-01 5.87317526e-01 9.09621537e-01 9.66556445e-02 4.25765932e-01 -1.48489729e-01 -8.19415689e-01 -1.49683619e+00 7.80428886e-01 2.97264546e-01 -3.82873684e-01 -8.56623709e-01 8.37175429e-01 2.12236702e-01 -5.31916976e-01 4.65321273e-01 1.00219278e-02 -1.52915865e-01 -6.90685734e-02 7.96921432e-01 5.70993483e-01 -2.28437349e-01 -9.43255424e-01 -4.52620745e-01 6.63841903e-01 -2.67221808e-01 6.06577620e-02 1.04236746e+00 -3.81095886e-01 3.37450534e-01 2.61540473e-01 1.16963148e+00 -1.10568061e-01 -1.44991553e+00 -6.02643728e-01 1.76067889e-01 -3.83734256e-01 -2.48104766e-01 -6.16470814e-01 -1.02694464e+00 5.33644319e-01 1.12157905e+00 -2.74790227e-01 8.10826600e-01 2.88274527e-01 9.23584342e-01 6.77189708e-01 3.36811662e-01 -1.17709327e+00 3.59907925e-01 4.22170162e-01 4.31520849e-01 -1.56154191e+00 -8.61234590e-02 -2.74427325e-01 -6.19158566e-01 6.65422678e-01 6.66027009e-01 -1.32177889e-01 6.73264444e-01 3.61278504e-02 8.76042321e-02 -2.85739899e-02 -3.07318687e-01 -8.56063843e-01 4.41511631e-01 8.12836766e-01 -6.01117276e-02 6.47474900e-02 1.57013521e-01 7.27195621e-01 1.69663325e-01 1.05564103e-01 2.22149163e-01 5.90921938e-01 -2.99161166e-01 -1.04772556e+00 -5.65498412e-01 3.13513309e-01 -3.20983082e-01 4.82685044e-02 7.89124221e-02 7.52898455e-01 5.94262540e-01 9.16454434e-01 8.17104280e-02 -6.56009316e-01 2.84136444e-01 1.24949068e-02 3.69698226e-01 -6.76560283e-01 -5.27775228e-01 9.51441228e-02 1.70315608e-01 -4.28614378e-01 -7.10396290e-01 -7.38868773e-01 -1.10170519e+00 -3.52184713e-01 -4.90630299e-01 3.95381413e-02 1.69883505e-01 1.10549116e+00 3.69354934e-01 3.04301232e-01 5.15662193e-01 -1.06513691e+00 -5.37635267e-01 -1.01793838e+00 -5.63304842e-01 9.18795407e-01 3.06153029e-01 -9.51792419e-01 -4.93709087e-01 2.01593205e-01]	[14.774704933166504, 1.1300017833709717]
43e267b4-2c89-412a-86b2-793112ea173f	spacephish-the-evasion-space-of-adversarial	2210.13660	null	https://arxiv.org/abs/2210.13660v1	https://arxiv.org/pdf/2210.13660v1.pdf	SpacePhish: The Evasion-space of Adversarial Attacks against Phishing Website Detectors using Machine Learning	Existing literature on adversarial Machine Learning (ML) focuses either on showing attacks that break every ML model, or defenses that withstand most attacks. Unfortunately, little consideration is given to the actual \textit{cost} of the attack or the defense. Moreover, adversarial samples are often crafted in the "feature-space", making the corresponding evaluations of questionable value. Simply put, the current situation does not allow to estimate the actual threat posed by adversarial attacks, leading to a lack of secure ML systems. We aim to clarify such confusion in this paper. By considering the application of ML for Phishing Website Detection (PWD), we formalize the "evasion-space" in which an adversarial perturbation can be introduced to fool a ML-PWD -- demonstrating that even perturbations in the "feature-space" are useful. Then, we propose a realistic threat model describing evasion attacks against ML-PWD that are cheap to stage, and hence intrinsically more attractive for real phishers. Finally, we perform the first statistically validated assessment of state-of-the-art ML-PWD against 12 evasion attacks. Our evaluation shows (i) the true efficacy of evasion attempts that are more likely to occur; and (ii) the impact of perturbations crafted in different evasion-spaces. Our realistic evasion attempts induce a statistically significant degradation (3-10% at $p\!<$0.05), and their cheap cost makes them a subtle threat. Notably, however, some ML-PWD are immune to our most realistic attacks ($p$=0.22). Our contribution paves the way for a much needed re-assessment of adversarial attacks against ML systems for cybersecurity.	['Ying Yuan', 'Mauro Conti', 'Giovanni Apruzzese']	2022-10-24	null	null	null	null	['phishing-website-detection']	['adversarial']	[ 3.02109092e-01 5.30762896e-02 9.78473797e-02 1.49127632e-01 -7.74720132e-01 -1.29544187e+00 8.88998508e-01 1.43241420e-01 -3.15538317e-01 5.83766162e-01 -2.40327269e-01 -1.08413053e+00 -5.25438376e-02 -9.71037984e-01 -9.23693299e-01 -7.72083163e-01 -4.76765960e-01 7.81431347e-02 1.36216134e-01 -6.04707003e-01 2.45952204e-01 5.63002706e-01 -8.66255462e-01 1.47564486e-01 4.88997340e-01 6.79030776e-01 -4.85625476e-01 9.12596822e-01 3.96571308e-01 4.80440557e-01 -1.10448074e+00 -9.16882932e-01 7.07382619e-01 -3.07766020e-01 -6.39033556e-01 -6.39934003e-01 4.13926870e-01 -5.45589030e-01 -4.20258522e-01 1.27173591e+00 4.54596758e-01 -3.28395605e-01 6.54361308e-01 -1.49521530e+00 -5.63041210e-01 6.14279807e-01 -2.68390507e-01 1.34543628e-01 3.96974415e-01 8.16020131e-01 8.75348210e-01 -2.42370188e-01 4.14616913e-01 1.27553201e+00 7.13230669e-01 7.38024116e-01 -1.37833643e+00 -6.82387948e-01 -1.88564509e-01 -1.94233581e-01 -1.00059664e+00 -2.38414750e-01 6.65142536e-01 -4.54967588e-01 6.60837471e-01 6.43191755e-01 2.44813666e-01 1.75367939e+00 5.37186086e-01 5.03233433e-01 1.36993957e+00 -2.85897374e-01 4.20385152e-01 5.20122290e-01 1.17036931e-01 3.11778218e-01 7.34541833e-01 7.01566517e-01 1.27132460e-01 -7.60661244e-01 4.68267232e-01 -2.09697247e-01 -2.86922246e-01 -2.54203558e-01 -6.71942353e-01 1.22664309e+00 4.23120290e-01 1.80010647e-01 -2.37901196e-01 2.10614234e-01 6.61905825e-01 7.20923483e-01 1.96166322e-01 1.07243335e+00 -6.45348072e-01 1.31294085e-02 -3.71308863e-01 3.72582197e-01 1.15280557e+00 3.55275273e-01 2.64090419e-01 1.61093712e-01 1.68435797e-01 2.40778849e-01 1.10833153e-01 6.99269056e-01 -1.85984727e-02 -4.88381505e-01 4.42334354e-01 1.44613296e-01 2.48988122e-01 -1.28592682e+00 -1.42462760e-01 -3.71728629e-01 -4.85611260e-01 7.67145038e-01 8.44435394e-01 -4.55137193e-01 -3.98420364e-01 1.87904704e+00 9.68499929e-02 6.04961626e-02 2.16151580e-01 5.47369957e-01 6.74307793e-02 3.85386497e-01 1.81655511e-01 -7.39682391e-02 1.41426897e+00 -2.15373725e-01 -2.32772917e-01 -2.47794434e-01 8.15245748e-01 -5.41167378e-01 1.44793570e+00 4.83372092e-01 -8.31395745e-01 -6.41323701e-02 -1.28681242e+00 8.29376698e-01 -7.20524192e-01 -8.41251910e-01 7.25002587e-01 1.60795689e+00 -7.51861572e-01 6.25915527e-01 -4.42279100e-01 -7.33569637e-02 3.71180773e-01 3.33246768e-01 -3.41931880e-01 1.00770541e-01 -1.75413370e+00 1.14595604e+00 1.62453011e-01 -3.31005782e-01 -1.34194541e+00 -7.57090271e-01 -5.67076862e-01 7.34401271e-02 3.98931563e-01 -3.77590090e-01 8.84076118e-01 -8.75844896e-01 -1.10525715e+00 7.22953200e-01 4.80945736e-01 -8.78203869e-01 8.20422888e-01 -1.67173371e-01 -6.01665139e-01 7.32578933e-02 -3.18106234e-01 -8.38842168e-02 1.01139462e+00 -1.46954381e+00 -5.82145005e-02 -2.76509374e-01 6.31548941e-01 -1.89242527e-01 -5.15547276e-01 2.89366812e-01 4.35106695e-01 -7.28448153e-01 -6.96007073e-01 -1.02999115e+00 -3.44830990e-01 -4.08801943e-01 -6.73418701e-01 2.13935643e-01 9.56289470e-01 -5.77228427e-01 1.19128621e+00 -2.08515048e+00 -3.20471793e-01 3.92510086e-01 2.74125934e-01 8.02013099e-01 -1.73694566e-01 7.24933803e-01 -1.91679284e-01 7.43078411e-01 -1.45177618e-01 1.95546425e-03 3.85188073e-01 -1.05677180e-01 -9.12939012e-01 7.21961617e-01 5.19600213e-02 9.18581009e-01 -6.85068071e-01 1.13071211e-01 2.22839564e-01 4.21662658e-01 -6.03022933e-01 1.16725534e-01 -1.80123299e-01 1.31149799e-01 -4.06815827e-01 5.42057574e-01 7.28284478e-01 1.30143657e-01 6.30066022e-02 1.57487646e-01 2.53419608e-01 1.81327924e-01 -9.28161621e-01 6.68138742e-01 -2.60943979e-01 5.33689559e-01 8.88288394e-02 -6.94155991e-01 6.56138361e-01 1.59497753e-01 8.91493112e-02 -3.57572138e-01 3.05244952e-01 2.83059210e-01 2.52046734e-01 -1.91162750e-01 1.85680509e-01 -1.64300874e-01 -5.03939688e-01 6.48924947e-01 -2.44948342e-01 -5.41108809e-02 -3.12760830e-01 2.69185364e-01 1.52618277e+00 -4.63773370e-01 3.07545364e-01 -4.19158399e-01 5.99794805e-01 -2.26585060e-01 2.22569138e-01 1.22702658e+00 -5.87620616e-01 7.68963248e-03 9.11335588e-01 -4.65268493e-01 -1.11557817e+00 -1.28991532e+00 -1.47493646e-01 8.19572270e-01 1.46047339e-01 -3.18120271e-01 -1.07395792e+00 -1.44027889e+00 1.91256955e-01 9.98972297e-01 -7.18259573e-01 -4.92674738e-01 -5.54657996e-01 -9.87871170e-01 1.32445610e+00 1.45429403e-01 4.02333915e-01 -9.23031986e-01 -4.13750142e-01 -1.28442824e-01 2.29048789e-01 -8.32370639e-01 -2.17365265e-01 5.20944372e-02 -3.83473933e-01 -1.29654920e+00 -4.90530618e-02 -1.99731588e-01 5.24351299e-01 8.59269947e-02 8.33232343e-01 1.76612765e-01 -3.01371127e-01 2.96735793e-01 -2.90108234e-01 -6.33045077e-01 -1.04964530e+00 -1.57209039e-01 3.91851664e-01 -2.29125261e-01 5.39363325e-01 -5.16969323e-01 -4.77735758e-01 4.68716741e-01 -1.02090633e+00 -8.75294685e-01 5.47342777e-01 7.57191241e-01 -3.03646684e-01 4.29383069e-01 8.02561939e-01 -1.12991130e+00 9.85638618e-01 -6.64185762e-01 -6.36153698e-01 1.35777622e-01 -6.31987631e-01 2.68740263e-02 1.05129766e+00 -7.56085932e-01 -7.35219181e-01 -4.39930201e-01 -5.24219632e-01 -2.38472283e-01 -3.81896257e-01 1.58061758e-01 -3.77468407e-01 -4.79814738e-01 1.31127799e+00 1.20924875e-01 1.65942088e-01 -3.51633370e-01 4.26710695e-01 4.05105680e-01 1.61710143e-01 -7.22389758e-01 1.60700047e+00 2.47733444e-01 4.92884479e-02 -7.47359157e-01 -3.05651873e-01 1.27660498e-01 -1.47709340e-01 1.85274947e-02 4.99321371e-01 -3.45184833e-01 -1.11975503e+00 5.30506313e-01 -9.00969028e-01 -4.08718497e-01 -5.24581969e-02 8.95767659e-02 -4.16734874e-01 7.90171325e-01 -7.87626505e-01 -9.62849259e-01 -3.25909734e-01 -1.23423934e+00 4.52496886e-01 -1.77442402e-01 -3.26689094e-01 -1.26268458e+00 1.78806990e-01 3.07654947e-01 7.76627183e-01 6.06576741e-01 1.01574540e+00 -1.39946496e+00 -1.32748157e-01 -8.27535391e-01 2.84803212e-02 6.00785911e-01 -1.35069907e-01 -4.30098101e-02 -1.12082696e+00 -6.88923657e-01 3.19850147e-01 -3.48090738e-01 3.92399848e-01 1.34291515e-01 8.49797666e-01 -9.53794062e-01 -1.70408010e-01 4.26863372e-01 1.53083158e+00 2.87367642e-01 8.63574147e-01 5.34159243e-01 2.85871506e-01 7.14573264e-01 3.63389343e-01 2.81911880e-01 -2.93169171e-01 6.81201041e-01 8.52804482e-01 -1.96674764e-02 4.32065964e-01 -3.49414021e-01 7.92615354e-01 -1.18016988e-01 2.24957988e-01 -3.60981911e-01 -7.97506452e-01 8.37696344e-02 -1.15159166e+00 -1.19222271e+00 -1.69338003e-01 2.53617406e+00 7.59893358e-01 7.58143842e-01 3.27804476e-01 2.38034040e-01 6.03522718e-01 2.40772143e-01 -2.64691830e-01 -7.28838325e-01 -9.37736407e-02 2.81185895e-01 7.68085420e-01 8.73241127e-01 -1.24450707e+00 8.74718308e-01 6.48537445e+00 9.19733167e-01 -9.62553263e-01 3.53917778e-02 6.07308328e-01 1.36270940e-01 -2.28307590e-01 1.81941777e-01 -6.94942892e-01 6.90778077e-01 1.25042236e+00 -1.12811565e-01 4.83709961e-01 1.06426167e+00 6.46151975e-02 3.09436977e-01 -1.04805362e+00 3.50557595e-01 -4.44289995e-03 -1.11179662e+00 -6.29134756e-03 5.15495479e-01 2.93380558e-01 -2.96520263e-01 4.39297915e-01 6.17669582e-01 7.69112825e-01 -1.18438482e+00 4.13164616e-01 -5.78922853e-02 5.87664664e-01 -9.40402329e-01 9.42657828e-01 4.98448908e-01 -5.72250307e-01 -1.74952731e-01 -2.80766338e-01 -1.90016050e-02 -2.66373740e-03 4.00429577e-01 -1.05599463e+00 2.88722605e-01 3.13053101e-01 -1.53976709e-01 -6.33552611e-01 4.12353843e-01 -3.55503261e-01 1.04622364e+00 -1.95966378e-01 -1.52825460e-01 3.70054573e-01 2.05136582e-01 8.26839983e-01 1.24834383e+00 -7.68354163e-02 -7.81803355e-02 7.36630633e-02 7.92256176e-01 1.89979926e-01 -1.30559281e-01 -1.09371662e+00 -1.70972541e-01 6.00477815e-01 9.83729541e-01 -3.75561059e-01 8.17444455e-03 -8.76926333e-02 7.69740641e-01 -1.13191739e-01 1.99275449e-01 -1.04742229e+00 -6.10037327e-01 1.10442293e+00 9.62680057e-02 -2.10825503e-01 -8.07716250e-02 -4.34384882e-01 -1.13516998e+00 -2.11052358e-01 -1.47873306e+00 4.31900680e-01 -6.50019571e-02 -1.59617102e+00 4.13311899e-01 -8.08919296e-02 -9.72913563e-01 -2.53927320e-01 -8.10854077e-01 -8.18225980e-01 1.03073335e+00 -9.84379947e-01 -9.64203238e-01 2.13815778e-01 5.47210515e-01 1.67794436e-01 -3.22399408e-01 9.75715995e-01 3.39348540e-02 -4.54075962e-01 9.83878732e-01 -1.17093980e-01 3.48409474e-01 6.19905174e-01 -1.20711923e+00 8.42225313e-01 1.19931662e+00 -9.53570604e-02 1.08106351e+00 1.11473644e+00 -7.11727560e-01 -1.43589747e+00 -8.03910911e-01 5.52578926e-01 -1.18403697e+00 1.00456583e+00 -6.27890766e-01 -8.57054234e-01 6.13390744e-01 -8.23262036e-02 -2.18175605e-01 6.74286187e-01 8.72808769e-02 -9.31476295e-01 1.93052471e-01 -1.68808532e+00 1.02608573e+00 6.99648976e-01 -6.00642383e-01 -5.02457082e-01 3.14433455e-01 9.53025103e-01 2.01869324e-01 -7.84862101e-01 2.75661826e-01 5.26258945e-01 -1.02116060e+00 1.42605221e+00 -9.80609596e-01 1.79333791e-01 -1.32200564e-03 -2.95404643e-01 -1.20383406e+00 -1.04371421e-01 -8.75508726e-01 -1.35365233e-01 1.13178813e+00 4.31161761e-01 -1.13456106e+00 6.88656271e-01 5.47685146e-01 2.56876409e-01 -5.84568441e-01 -8.75952244e-01 -1.03147149e+00 6.25120878e-01 -6.50043607e-01 4.36958194e-01 1.33992708e+00 1.70598105e-01 -4.37086746e-02 -5.20316899e-01 5.27587831e-01 9.25484478e-01 -6.42268181e-01 1.04374790e+00 -8.95332873e-01 -8.04729700e-01 -6.01246178e-01 -5.05452931e-01 -4.70831454e-01 -1.16424700e-02 -5.85433364e-01 -2.99157679e-01 -5.30615628e-01 -1.07998684e-01 -5.71179807e-01 -2.19616130e-01 4.68884289e-01 -3.56410474e-01 1.82544723e-01 3.54927778e-01 1.74003854e-01 3.60041186e-02 -8.74176770e-02 5.32895267e-01 -1.75880417e-01 -1.22637320e-02 2.11285874e-01 -9.58409905e-01 7.33035803e-01 1.03990805e+00 -5.03186107e-01 -4.54609513e-01 2.66648591e-01 2.21576899e-01 -7.34431297e-02 6.97373748e-01 -7.85871804e-01 -3.25269908e-01 -3.10172856e-01 1.05668113e-01 1.21080503e-01 1.02270760e-01 -7.36733079e-01 -4.09571156e-02 1.07742190e+00 -3.58814836e-01 -1.24913016e-02 1.47237957e-01 5.79465628e-01 3.96661103e-01 -3.27931732e-01 1.00034070e+00 -1.03116550e-01 -1.73463061e-01 9.18436497e-02 -3.60871613e-01 1.51890725e-01 1.26127911e+00 2.25320011e-02 -8.35727572e-01 -5.45463741e-01 -4.26758409e-01 -2.23150447e-01 7.14880526e-01 2.54069299e-01 3.47701609e-01 -8.82985830e-01 -8.11054111e-01 3.17932516e-01 -6.43428266e-02 -9.06838357e-01 2.29651451e-01 5.12161255e-01 -5.95258474e-01 2.37331823e-01 -2.88480759e-01 -6.88777566e-02 -1.32874131e+00 1.05933857e+00 3.32812160e-01 -5.05401790e-01 -4.29967463e-01 7.89802372e-01 2.95599431e-01 -3.13980788e-01 1.58039808e-01 4.27238166e-01 1.53517395e-01 -2.86458611e-01 6.76589966e-01 3.41054589e-01 -1.07257277e-01 -3.32891047e-01 -3.92313480e-01 -2.12045554e-02 -2.40973502e-01 -8.28051865e-02 8.88234615e-01 4.28024083e-01 1.14645213e-02 -1.69513151e-01 1.15403581e+00 4.36544210e-01 -1.08343518e+00 5.27834035e-02 1.22886803e-02 -7.64967978e-01 -5.30879796e-01 -1.00801063e+00 -4.88604218e-01 8.59877050e-01 4.11183685e-01 9.44590271e-01 8.57228279e-01 -3.48338306e-01 6.58353925e-01 4.00635153e-01 6.08734131e-01 -5.82375526e-01 1.83350444e-01 2.94094980e-01 7.10421562e-01 -1.02424359e+00 -8.87268856e-02 -3.04143220e-01 -6.57094836e-01 7.86517203e-01 4.52215135e-01 -3.90991300e-01 4.69837874e-01 4.18739557e-01 -1.04960538e-02 -7.68695027e-02 -5.62537193e-01 5.03737628e-01 -6.99535161e-02 8.54824066e-01 -7.96089992e-02 2.25577056e-01 -4.48055029e-01 5.97040415e-01 -2.84000576e-01 -7.98089445e-01 7.70011723e-01 7.92030036e-01 -4.19978619e-01 -1.32928431e+00 -6.96430385e-01 2.17106551e-01 -8.35236907e-01 -1.11444995e-01 -7.30692387e-01 9.88771856e-01 -8.66219848e-02 1.18673050e+00 -6.52495861e-01 -8.30263317e-01 2.58749276e-01 1.56340703e-01 7.01224953e-02 -3.07678759e-01 -1.02765810e+00 -3.32061350e-01 3.68350476e-01 -5.47374070e-01 3.83714974e-01 -6.14010394e-01 -5.89758933e-01 -1.04550958e+00 -3.63939345e-01 1.68409303e-01 6.49391294e-01 6.94783330e-01 1.68111145e-01 1.44931704e-01 1.05703413e+00 -5.78034103e-01 -1.44250500e+00 -7.07208872e-01 -5.46903610e-01 5.80769598e-01 2.61358052e-01 -5.11244774e-01 -1.07604444e+00 -3.53354126e-01]	[5.826358318328857, 7.687694072723389]
602e6c9a-867e-4380-a7c9-491127ba27ac	rethinking-of-pedestrian-attribute	2005.11909	null	https://arxiv.org/abs/2005.11909v2	https://arxiv.org/pdf/2005.11909v2.pdf	Rethinking of Pedestrian Attribute Recognition: Realistic Datasets with Efficient Method	Despite various methods are proposed to make progress in pedestrian attribute recognition, a crucial problem on existing datasets is often neglected, namely, a large number of identical pedestrian identities in train and test set, which is not consistent with practical application. Thus, images of the same pedestrian identity in train set and test set are extremely similar, leading to overestimated performance of state-of-the-art methods on existing datasets. To address this problem, we propose two realistic datasets PETA\textsubscript{$zs$} and RAPv2\textsubscript{$zs$} following zero-shot setting of pedestrian identities based on PETA and RAPv2 datasets. Furthermore, compared to our strong baseline method, we have observed that recent state-of-the-art methods can not make performance improvement on PETA, RAPv2, PETA\textsubscript{$zs$} and RAPv2\textsubscript{$zs$}. Thus, through solving the inherent attribute imbalance in pedestrian attribute recognition, an efficient method is proposed to further improve the performance. Experiments on existing and proposed datasets verify the superiority of our method by achieving state-of-the-art performance.	['Houjing Huang', 'Xiaotang Chen', 'Wenjie Yang', 'Kaiqi Huang', 'Jian Jia']	2020-05-25	null	null	null	null	['pedestrian-attribute-recognition']	['computer-vision']	[-4.95733730e-02 -4.36872691e-01 5.04719242e-02 -7.14409947e-01 -5.36881745e-01 -3.06921691e-01 6.27407610e-01 -2.92897839e-02 -4.71674949e-01 1.07850885e+00 -2.01487705e-01 -1.39321297e-01 1.47194728e-01 -1.01046979e+00 -6.93788886e-01 -7.71305978e-01 1.37521446e-01 6.52231276e-01 2.29107022e-01 -2.30294272e-01 -7.63590122e-03 2.82943826e-02 -2.02342367e+00 2.33652577e-01 8.67877483e-01 1.31471169e+00 -2.86491096e-01 3.64492655e-01 -1.30131245e-01 5.32274604e-01 -5.58958828e-01 -1.39701903e+00 4.93546605e-01 -3.38691831e-01 -6.97453141e-01 -1.98752433e-02 1.06240714e+00 -3.56735855e-01 -7.62183905e-01 1.19320631e+00 8.07073176e-01 4.39164974e-02 9.70290124e-01 -1.88101649e+00 -8.11398327e-01 2.44518861e-01 -6.93238318e-01 9.47040692e-02 2.96975523e-01 2.38137528e-01 8.13395202e-01 -1.11014485e+00 2.11983919e-01 1.37958372e+00 9.86832559e-01 5.10511637e-01 -1.00055838e+00 -9.54032183e-01 2.95119792e-01 7.89709806e-01 -1.93633020e+00 -4.90501732e-01 4.01262134e-01 -3.78936589e-01 4.80121702e-01 5.40186703e-01 6.71418607e-01 1.35806072e+00 -3.93283874e-01 1.03142893e+00 1.36885476e+00 -1.11603431e-01 8.10073502e-03 3.75623226e-01 3.96798432e-01 5.17583847e-01 4.26742792e-01 -1.61858592e-02 -4.39065099e-01 2.68488284e-03 4.52719241e-01 -2.64711499e-01 3.48128714e-02 -3.99965942e-01 -1.07847929e+00 5.54897070e-01 1.07719088e-02 -4.41597730e-01 -1.80584237e-01 -3.05001557e-01 7.15233803e-01 1.86165318e-01 3.17416303e-02 -1.74396291e-01 -1.38933927e-01 -2.33568981e-01 -5.11543870e-01 3.83207202e-01 4.68299180e-01 1.53712344e+00 7.53587484e-01 7.96791688e-02 -3.29059631e-01 9.77045655e-01 1.22209735e-01 7.06811845e-01 9.62768719e-02 -8.12169552e-01 5.86774886e-01 6.31778419e-01 2.35302359e-01 -8.64434183e-01 1.57973304e-01 -1.68939292e-01 -1.06817162e+00 3.33740741e-01 7.89719403e-01 1.61034465e-01 -8.15570652e-01 1.40851712e+00 2.98507929e-01 2.81498998e-01 1.51952893e-01 8.98488343e-01 1.03005993e+00 3.19932520e-01 5.86148441e-01 2.01473549e-01 1.68191671e+00 -8.59935045e-01 -3.01405281e-01 -2.59832174e-01 2.37749562e-01 -9.10834074e-01 1.10039365e+00 1.54272154e-01 -7.85823941e-01 -9.89103079e-01 -1.13366711e+00 1.97809041e-01 -6.75480843e-01 3.57935667e-01 6.38733149e-01 1.28329551e+00 -7.46160269e-01 2.11669385e-01 -1.04311705e-01 -7.07616329e-01 1.01143038e+00 2.91948736e-01 -6.07396126e-01 -2.72736102e-01 -1.11311674e+00 9.24563706e-01 3.39693934e-01 2.55979057e-02 -8.99395406e-01 -6.69451356e-01 -7.02322841e-01 -2.16461971e-01 2.43690133e-01 -7.49305904e-01 8.00003827e-01 -6.49528325e-01 -1.10151863e+00 1.40230489e+00 -2.70956069e-01 -4.31309670e-01 1.08455265e+00 -2.20001847e-01 -5.67435384e-01 -1.85852394e-01 3.98323625e-01 5.30441284e-01 4.25645739e-01 -1.39224064e+00 -1.01843989e+00 -6.24812961e-01 1.08009435e-01 1.08619109e-01 -2.04298571e-01 1.55753210e-01 -3.77223760e-01 -5.35617113e-01 1.82729233e-02 -8.12166035e-01 1.26091465e-01 1.22285463e-01 -7.07626104e-01 -3.08318526e-01 7.85780013e-01 -4.29440707e-01 7.27994204e-01 -1.95404077e+00 -3.90321910e-01 3.63943875e-02 3.52062471e-02 6.92772627e-01 -1.10891517e-02 1.39405414e-01 -8.54158029e-02 -3.86531428e-02 2.24577766e-02 -4.01050597e-01 2.74559706e-01 3.38812530e-01 -1.51927441e-01 4.68598604e-01 3.80271897e-02 6.88704491e-01 -8.20630312e-01 -8.81268263e-01 4.53305930e-01 3.70276392e-01 2.44146660e-02 2.14856178e-01 3.68471742e-01 3.84406894e-01 -6.69129789e-01 1.08969021e+00 1.17177677e+00 1.92615554e-01 -2.83098191e-01 -3.95010114e-01 -1.11752406e-01 -2.22183421e-01 -1.37682819e+00 1.04819977e+00 9.45347771e-02 4.46790278e-01 -3.12940657e-01 -1.41902184e+00 1.11338723e+00 1.56928807e-01 4.33388919e-01 -7.59856939e-01 2.75323361e-01 9.03273448e-02 -2.88734823e-01 -4.82329100e-01 5.49401820e-01 1.14873096e-01 -9.45687816e-02 -6.57355785e-02 -1.76769646e-03 4.23903793e-01 2.81134993e-01 8.95792246e-02 4.71694589e-01 3.40473175e-01 9.53516364e-02 -3.68724346e-01 9.85588372e-01 -1.02816001e-01 6.30497336e-01 1.19936109e+00 -9.39969659e-01 9.48119938e-01 3.01628828e-01 -7.16048062e-01 -1.50405860e+00 -1.38439906e+00 -3.82054150e-01 1.21441042e+00 4.14865166e-01 -8.25144425e-02 -9.62608933e-01 -6.68730795e-01 -2.82402784e-02 5.24766088e-01 -7.08115995e-01 3.49703059e-02 -4.51184392e-01 -1.21033418e+00 1.12609005e+00 7.45638490e-01 1.31518257e+00 -6.24796867e-01 -1.63616419e-01 1.47699356e-01 -5.99667072e-01 -1.55079520e+00 -3.36947173e-01 -6.97607219e-01 -9.43974704e-02 -1.30110788e+00 -9.59488153e-01 -7.51571417e-01 6.02107465e-01 4.78353083e-01 1.29899526e+00 9.14227962e-02 -3.07104170e-01 3.21603388e-01 -2.54682481e-01 -5.28707027e-01 -1.37135819e-01 -4.03211027e-01 3.93381238e-01 4.48369682e-01 8.47605765e-01 -5.15294075e-01 -9.05805111e-01 9.81736720e-01 -4.66449596e-02 1.25305951e-01 4.63194579e-01 7.35915840e-01 6.46152437e-01 -1.41977057e-01 7.75565147e-01 -7.83822417e-01 -1.00981824e-01 -2.65637666e-01 -3.38097245e-01 5.33962965e-01 -4.17585433e-01 -4.48333561e-01 6.34922802e-01 -4.68234777e-01 -1.03467333e+00 5.42095713e-02 -2.92628467e-01 -3.01877737e-01 -6.35891736e-01 -3.37852091e-01 -7.76583791e-01 -1.81820482e-01 4.52623844e-01 6.57818854e-01 -1.75385490e-01 -1.48333415e-01 2.47622922e-01 7.92765439e-01 1.06081474e+00 -9.13464904e-01 7.15142429e-01 6.11773670e-01 5.61493076e-02 -8.12123179e-01 -6.75568938e-01 -4.91237104e-01 -8.95323753e-01 -3.91910851e-01 8.48087788e-01 -1.20053232e+00 -1.08889353e+00 1.14351630e+00 -7.98422813e-01 1.75850391e-01 -1.19968697e-01 4.56652671e-01 -6.51962399e-01 5.22081137e-01 -2.50842392e-01 -8.90019000e-01 -3.14866215e-01 -1.16357458e+00 8.78471076e-01 4.85568315e-01 4.62868363e-02 -4.06870633e-01 -5.04403532e-01 7.97340989e-01 3.07938069e-01 3.62999827e-01 7.01130450e-01 -6.65511966e-01 -5.33712447e-01 -3.93800706e-01 -7.95934856e-01 3.57594520e-01 3.21524926e-02 5.14822789e-02 -1.19474876e+00 -1.68616176e-01 -6.53591931e-01 -3.05922121e-01 5.88393807e-01 9.77584720e-02 1.08042192e+00 -9.90778655e-02 -3.76149535e-01 5.60378432e-01 1.05265307e+00 1.61878437e-01 1.07906187e+00 5.95393121e-01 8.25543880e-01 5.80769539e-01 7.89513528e-01 4.76680666e-01 9.86503780e-01 7.32641339e-01 2.45502785e-01 -4.44775261e-02 -3.87689233e-01 -2.74954349e-01 -7.79343843e-02 3.32600892e-01 -7.26127446e-01 -9.29054394e-02 -7.11513221e-01 6.23233497e-01 -1.86527395e+00 -1.30804896e+00 -5.69050848e-01 2.68118167e+00 4.81541038e-01 1.46633489e-02 4.47057605e-01 2.71347195e-01 1.03708076e+00 -8.39568954e-03 -4.26016241e-01 5.34805730e-02 -6.16809249e-01 -4.66133282e-02 4.63809848e-01 -1.05973318e-01 -1.56228364e+00 7.81902969e-01 5.77084875e+00 1.05766833e+00 -6.21345162e-01 -5.20841703e-02 1.09132755e+00 3.26046884e-01 4.59677339e-01 -2.71610171e-01 -1.06767130e+00 9.06133175e-01 4.58648801e-01 -3.53715748e-01 -5.32436855e-02 9.50416386e-01 -1.92812741e-01 5.26070707e-02 -1.17552018e+00 1.29740822e+00 1.07069433e-01 -9.92867231e-01 2.28920236e-01 -1.02698348e-01 5.51229298e-01 -4.53547001e-01 2.34402940e-01 8.15281391e-01 4.13147151e-01 -1.11541593e+00 6.98680580e-01 1.11835010e-01 1.02672148e+00 -8.31546187e-01 1.09118664e+00 -4.34437059e-02 -1.57327998e+00 2.36925483e-01 -8.48666787e-01 1.77318752e-02 4.81841974e-02 -4.46902774e-02 -4.40456480e-01 9.60723758e-01 1.08927143e+00 4.62603807e-01 -7.72680342e-01 1.14383268e+00 7.00933114e-02 4.31419373e-01 -2.09858477e-01 9.93933529e-02 1.84586361e-01 -3.02495062e-01 3.76135319e-01 1.18051779e+00 4.42894220e-01 4.11871612e-01 2.15805039e-01 5.15741944e-01 3.29608768e-02 1.64056495e-01 -4.44225341e-01 4.83618498e-01 6.97475731e-01 1.09539461e+00 -4.84079838e-01 -6.18575156e-01 -7.07296729e-01 8.55041802e-01 1.21326894e-01 3.65185201e-01 -1.12421000e+00 -1.13866650e-01 1.09827220e+00 2.09112182e-01 4.35186684e-01 1.46499753e-01 -2.56641269e-01 -1.13505912e+00 9.15423483e-02 -9.67998385e-01 7.06541598e-01 -5.22485256e-01 -1.81072032e+00 5.45048356e-01 8.86076465e-02 -1.58422422e+00 1.65689439e-01 -6.32152498e-01 -6.50416613e-01 9.46415544e-01 -1.35243821e+00 -1.85622239e+00 -8.19992185e-01 1.00150871e+00 5.21952987e-01 -6.37943447e-01 7.69768536e-01 7.86472082e-01 -6.49124980e-01 1.22757256e+00 1.19603746e-01 5.82084477e-01 8.31677258e-01 -1.12334013e+00 7.15872347e-01 7.90658176e-01 -2.15706691e-01 2.28665948e-01 8.04020703e-01 -3.34641129e-01 -1.27543867e+00 -1.26173794e+00 4.52784389e-01 -6.69880152e-01 3.59758407e-01 -3.03670168e-01 -9.81935561e-01 6.83116376e-01 1.24356076e-01 2.73865253e-01 7.70063460e-01 -4.01044711e-02 -5.45855582e-01 -2.41194919e-01 -1.28719413e+00 5.86658716e-01 1.38282156e+00 -2.74816871e-01 -4.40847844e-01 1.51432052e-01 1.44297527e-02 -3.51287246e-01 -8.50403070e-01 4.88180190e-01 7.95831263e-01 -1.11153734e+00 1.70973766e+00 -7.28285253e-01 1.76608920e-01 -3.24989527e-01 -4.96874243e-01 -9.70425308e-01 -2.05288917e-01 -2.30102286e-01 2.03211427e-01 1.71705544e+00 -9.62997507e-03 -7.55710840e-01 1.07735503e+00 7.91274071e-01 -3.74972261e-02 -4.46007013e-01 -1.19567847e+00 -1.06326115e+00 2.87707150e-01 -9.30165425e-02 7.80223727e-01 9.88673329e-01 -6.09907806e-01 3.87254916e-02 -6.82084978e-01 3.42768580e-01 1.35260367e+00 -5.30752763e-02 1.23816812e+00 -1.17371058e+00 2.01507181e-01 -4.50028032e-01 -1.03036785e+00 -7.90291309e-01 -2.22076625e-02 -4.08640891e-01 -1.92082435e-01 -1.35215390e+00 6.10081851e-01 -8.65269244e-01 -3.32486629e-01 2.51842916e-01 -5.74614704e-01 7.78814793e-01 1.02967441e-01 9.98094603e-02 -6.77377701e-01 6.94584250e-01 9.79379714e-01 -4.05115485e-01 5.58922827e-01 1.49016827e-01 -6.95844889e-01 7.69029498e-01 5.69343865e-01 -1.16187185e-01 -1.83796346e-01 -1.50002941e-01 -5.14902174e-01 -4.36942458e-01 8.04711938e-01 -1.25413668e+00 2.34591246e-01 -4.85167280e-02 7.60024905e-01 -8.21319938e-01 6.72579467e-01 -4.75219101e-01 2.22697586e-01 -2.30179969e-02 2.35399067e-01 1.34872258e-01 1.62896141e-01 4.40292329e-01 -2.19270870e-01 -2.85379152e-04 8.72366130e-01 -1.14552706e-01 -1.21757019e+00 7.25552917e-01 -4.90987152e-02 2.94919223e-01 1.32986045e+00 -7.09986269e-01 -7.00907111e-01 -2.66464800e-01 -4.41922843e-01 4.20197934e-01 4.16414052e-01 4.24142540e-01 6.21757567e-01 -1.68076372e+00 -1.24490929e+00 1.75099656e-01 4.21349585e-01 -2.84325242e-01 6.36490226e-01 6.35823429e-01 -4.05477196e-01 7.91094452e-02 -8.75708461e-01 -6.89974308e-01 -1.59383476e+00 5.68654954e-01 2.21628860e-01 2.40147352e-01 -5.05201221e-01 8.47595274e-01 4.95890796e-01 -5.22792041e-01 8.19158778e-02 6.24986827e-01 -2.47658432e-01 1.63454190e-02 6.52409792e-01 6.13578498e-01 -1.97528109e-01 -1.27579463e+00 -4.94910002e-01 6.54273868e-01 -3.23094457e-01 3.46729577e-01 8.65577936e-01 -3.56393367e-01 3.16209018e-01 -5.89121468e-02 9.53271031e-01 -3.35483432e-01 -1.25369203e+00 -1.98477507e-01 -3.84323001e-01 -1.01410806e+00 -7.29240119e-01 -5.56694329e-01 -9.50276911e-01 9.30858850e-01 1.00184548e+00 -1.12885721e-01 7.73230791e-01 -2.60003448e-01 7.86749721e-01 3.03355247e-01 7.36202538e-01 -1.19739139e+00 -1.18832447e-01 1.94349676e-01 3.82839352e-01 -1.74196076e+00 1.47070557e-01 -7.39231765e-01 -9.21512902e-01 6.64203942e-01 1.06990004e+00 2.01803640e-01 1.73127875e-01 -4.87603843e-02 2.56243616e-01 3.22080940e-01 -5.87232523e-02 -2.81838536e-01 1.97930727e-02 1.12416470e+00 1.07060835e-01 1.99771538e-01 -5.04772225e-03 6.83344841e-01 -3.29731524e-01 -2.99475670e-01 5.07436454e-01 7.18434870e-01 -1.32663935e-01 -1.16642714e+00 -4.76226807e-01 3.07009757e-01 -4.33139801e-01 2.15660734e-03 -1.34761035e-01 1.11664224e+00 2.97025353e-01 9.04974878e-01 5.54827675e-02 -4.99432951e-01 6.35486722e-01 2.85204798e-02 5.95746100e-01 2.59940505e-01 -1.66020289e-01 -6.50273502e-01 4.12830383e-01 -2.26698697e-01 -4.01068658e-01 -9.00525451e-01 -6.74677253e-01 -8.97961378e-01 2.88408343e-02 -1.37132287e-01 1.73801795e-01 7.96684146e-01 1.58763930e-01 1.93994612e-01 4.20633644e-01 -5.26500225e-01 -4.34117377e-01 -8.10236156e-01 -3.73613149e-01 1.03652108e+00 -2.51063913e-01 -1.01262391e+00 7.62848333e-02 2.50256211e-01]	[14.538678169250488, 0.9516131281852722]
4e3d356a-b778-4306-bb31-5c11537abe51	machine-vision-based-sample-tube-localization	2103.09942	null	https://arxiv.org/abs/2103.09942v1	https://arxiv.org/pdf/2103.09942v1.pdf	Machine Vision based Sample-Tube Localization for Mars Sample Return	A potential Mars Sample Return (MSR) architecture is being jointly studied by NASA and ESA. As currently envisioned, the MSR campaign consists of a series of 3 missions: sample cache, fetch and return to Earth. In this paper, we focus on the fetch part of the MSR, and more specifically the problem of autonomously detecting and localizing sample tubes deposited on the Martian surface. Towards this end, we study two machine-vision based approaches: First, a geometry-driven approach based on template matching that uses hard-coded filters and a 3D shape model of the tube; and second, a data-driven approach based on convolutional neural networks (CNNs) and learned features. Furthermore, we present a large benchmark dataset of sample-tube images, collected in representative outdoor environments and annotated with ground truth segmentation masks and locations. The dataset was acquired systematically across different terrain, illumination conditions and dust-coverage; and benchmarking was performed to study the feasibility of each approach, their relative strengths and weaknesses, and robustness in the presence of adverse environmental conditions.	['Renaud Detry', 'Eric Kulczyski', 'John Mayo', 'Alex Brinkman', 'Mark Van der Merwe', 'Gerard Maggiolino', 'Peter Ilhardt', 'Tu-Hoa Pham', 'William Seto', 'Barry Ridge', 'Shreyansh Daftry']	2021-03-17	null	null	null	null	['template-matching']	['computer-vision']	[ 4.52500105e-01 9.90857556e-02 3.34776610e-01 -4.93901342e-01 -3.96665752e-01 -6.34104490e-01 7.48940229e-01 2.07412332e-01 -5.90408325e-01 2.94999778e-01 -3.85550678e-01 -3.66378576e-01 -8.55734050e-02 -9.44252133e-01 -7.38406360e-01 -3.21558297e-01 -2.22997800e-01 8.88041973e-01 4.05992746e-01 -6.64605856e-01 3.02104801e-01 1.03918052e+00 -1.85785186e+00 8.79901573e-02 8.29578042e-01 1.15387547e+00 3.69053036e-01 5.93606234e-01 1.46999225e-01 2.55485356e-01 -2.23619670e-01 -2.13767923e-02 7.29292154e-01 -5.45905009e-02 -7.06919193e-01 9.00330767e-02 4.84379917e-01 3.66250835e-02 -8.64504725e-02 6.36420369e-01 9.23525020e-02 2.15422109e-01 7.46048093e-01 -6.97102189e-01 3.07075769e-01 -1.06670067e-01 -4.03580606e-01 4.48708087e-02 1.17333822e-01 2.62754500e-01 8.47177267e-01 -7.73515105e-01 6.99316978e-01 9.91507113e-01 8.65002930e-01 6.87144473e-02 -1.10112870e+00 -1.40184373e-01 1.07416240e-02 -8.23855847e-02 -1.12058580e+00 -4.37950194e-01 2.95845181e-01 -7.26881683e-01 1.03909838e+00 4.49517190e-01 7.79876590e-01 4.58978564e-01 1.30070806e-01 6.04648769e-01 1.10235322e+00 -4.49303359e-01 4.68882918e-01 4.57668006e-02 -7.57358000e-02 6.98839486e-01 5.85475862e-01 2.87659466e-01 -3.84406835e-01 -9.27959085e-02 5.73205173e-01 -2.99891859e-01 -3.40279251e-01 -5.81313968e-01 -1.09511828e+00 8.28821063e-01 6.25097096e-01 -4.31887247e-02 -4.26558137e-01 -5.65151572e-02 2.71915764e-01 2.51960963e-01 5.09667754e-01 6.01442933e-01 -3.43915999e-01 4.39408749e-01 -1.18664491e+00 5.57913363e-01 8.39906752e-01 7.59846747e-01 1.11121511e+00 -6.03920780e-02 2.46527210e-01 5.65537035e-01 6.42990112e-01 8.99326265e-01 1.82579771e-01 -7.01928973e-01 4.50540483e-01 7.08387256e-01 3.33702505e-01 -8.52147877e-01 -7.60636508e-01 -4.05735493e-01 -1.31927669e-01 4.55747515e-01 2.75807738e-01 -1.24072574e-01 -1.10786271e+00 8.05925310e-01 6.51010871e-01 -2.58702099e-01 2.68746406e-01 1.07711399e+00 7.31802940e-01 3.88327450e-01 -5.10800242e-01 5.29606342e-01 1.23883510e+00 -7.11111426e-01 -2.44927704e-02 -8.64046156e-01 6.32475257e-01 -4.91602212e-01 8.01762938e-01 3.94404054e-01 -6.35336041e-01 -2.25364923e-01 -1.49069440e+00 -8.83331299e-02 -6.14538968e-01 5.69509745e-01 5.22054970e-01 5.77837408e-01 -8.11689079e-01 7.78165579e-01 -9.93988514e-01 -8.60763967e-01 2.39239335e-01 2.34286055e-01 -2.52233684e-01 5.49293682e-02 -7.09870100e-01 1.00627410e+00 2.85256326e-01 3.56659800e-01 -1.02739465e+00 -3.72522563e-01 -9.92855012e-01 -2.95286030e-01 1.18033849e-01 -5.40492237e-01 1.34727752e+00 -8.60785544e-01 -1.16792774e+00 1.21566796e+00 9.38843936e-03 -1.00016284e+00 7.99825430e-01 -5.66004455e-01 -1.30319029e-01 8.42842460e-02 1.61402132e-02 5.50732553e-01 5.51304996e-01 -1.36277831e+00 -7.70914257e-01 -6.12815797e-01 -1.29565760e-01 3.37246120e-01 2.91811079e-01 -1.14330776e-01 -1.34118363e-01 -1.33398667e-01 4.60814297e-01 -1.01838648e+00 -3.29963118e-01 2.02420190e-01 -1.91369832e-01 3.98374051e-01 7.37048745e-01 -6.54130816e-01 4.67481911e-01 -1.86834073e+00 -3.44168916e-02 3.81771863e-01 -1.58248693e-01 1.51407972e-01 7.31164292e-02 5.00892043e-01 2.12797612e-01 -2.08565459e-01 -8.74025106e-01 -2.52768070e-01 -9.85669941e-02 1.84223950e-01 -2.64773160e-01 8.40836108e-01 1.61470070e-01 5.17953157e-01 -6.68594718e-01 8.86887982e-02 3.63126278e-01 1.47740126e-01 -6.10862561e-02 1.10810190e-01 -6.68099821e-01 3.09960812e-01 -3.29398334e-01 9.71500337e-01 9.07964945e-01 4.60718989e-01 7.19043612e-02 -2.04339668e-01 -6.71404779e-01 2.27342159e-01 -9.10710752e-01 1.57342553e+00 -3.42743188e-01 8.03232253e-01 6.15323722e-01 -6.53938234e-01 1.48428988e+00 -3.14362377e-01 1.34641692e-01 -9.22392190e-01 3.57884556e-01 7.43411422e-01 -2.30361953e-01 -6.66557014e-01 1.19394648e+00 -3.19234841e-02 9.90990698e-02 1.82346478e-01 -3.54175091e-01 -4.10255462e-01 -8.47669020e-02 -1.05285317e-01 8.94879580e-01 4.65112656e-01 2.22730376e-02 -5.94049335e-01 2.40359351e-01 6.76852643e-01 2.91686505e-01 5.91050327e-01 -6.78229257e-02 9.04156685e-01 1.30172625e-01 -6.44898236e-01 -9.94460762e-01 -8.50217760e-01 -3.59226853e-01 5.97759843e-01 4.80370462e-01 -3.89539707e-03 -6.41934872e-01 -4.03138518e-01 3.81368458e-01 7.02796578e-01 -7.10694492e-01 -9.42888763e-03 -4.51068521e-01 -1.04322517e+00 3.65285575e-01 3.21134388e-01 6.36522174e-01 -8.73849928e-01 -1.45182788e+00 7.40700513e-02 1.77283715e-02 -1.04815233e+00 3.78613293e-01 4.67488229e-01 -8.38890076e-01 -1.37776268e+00 -1.42228559e-01 -1.89246029e-01 5.10249138e-01 4.15237635e-01 1.14427006e+00 -8.30986574e-02 -7.55120039e-01 4.34700698e-01 -5.94286680e-01 -7.25163102e-01 -1.34164497e-01 4.38745227e-03 -2.01681435e-01 2.34523982e-01 1.32718325e-01 -8.64360780e-02 -5.97965658e-01 6.28037035e-01 -8.27934980e-01 -7.02202693e-02 7.21795082e-01 3.61591965e-01 6.56079948e-01 -3.88421178e-01 -4.52197902e-02 -8.17243397e-01 1.35383591e-01 -6.63631737e-01 -7.72214234e-01 2.67106920e-01 -1.81590304e-01 -2.44519800e-01 -4.23083501e-03 2.74059683e-01 -9.64521945e-01 4.74662185e-01 -1.08613774e-01 -1.32781252e-01 -2.47716874e-01 3.81108969e-01 -5.09276949e-02 -5.25700808e-01 1.13687027e+00 1.73615158e-01 1.07627265e-01 -4.79393035e-01 3.40625703e-01 8.28539789e-01 7.51868784e-01 -5.84236205e-01 8.97965372e-01 1.14239585e+00 6.07880317e-02 -1.31873000e+00 -6.79470837e-01 -7.01862931e-01 -8.11281383e-01 -4.86419678e-01 8.19437325e-01 -9.38107014e-01 1.46446878e-03 3.53689104e-01 -7.21122622e-01 -8.51471901e-01 -1.86803684e-01 4.21983331e-01 -4.71144021e-01 1.40791029e-01 -2.35739760e-02 -9.24709260e-01 -7.14295685e-01 -9.29570377e-01 1.29409623e+00 2.82328516e-01 3.61267701e-02 -6.01580799e-01 1.40949681e-01 3.75946105e-01 3.80690902e-01 8.27039480e-01 3.68088156e-01 -4.13875669e-01 -6.37260973e-01 -4.33588833e-01 -3.12906444e-01 4.97420579e-02 -1.39570072e-01 1.45821825e-01 -1.17359161e+00 -5.44035673e-01 9.09099802e-02 -2.21226931e-01 1.30655193e+00 2.77956724e-01 5.50043762e-01 2.51438349e-01 -5.57843029e-01 8.77674937e-01 1.52384722e+00 3.67848352e-02 7.18511939e-01 8.83325815e-01 3.01990211e-01 8.53413939e-01 1.13973224e+00 4.35900241e-01 2.90791988e-01 4.93102133e-01 1.11749971e+00 -2.34879389e-01 2.08381444e-01 9.66214612e-02 1.27160713e-01 -1.10232867e-01 -1.90466523e-01 -1.64684221e-01 -1.01872241e+00 5.81990182e-01 -1.76349545e+00 -4.43584591e-01 -5.06833076e-01 2.52836990e+00 -2.06295267e-01 1.52277857e-01 -4.09141444e-02 -1.56305864e-01 5.13175547e-01 1.16679616e-01 -5.98883033e-01 -4.32370216e-01 -2.26844430e-01 1.10064715e-01 1.05726504e+00 2.20317915e-01 -1.23624599e+00 7.93103158e-01 6.30862570e+00 1.91993400e-01 -1.33861697e+00 -2.51814783e-01 3.56485724e-01 4.82392237e-02 -2.74923414e-01 1.60231084e-01 -8.66141856e-01 -2.35060193e-02 8.90256166e-01 3.60124499e-01 4.57074612e-01 8.63538027e-01 2.19654158e-01 -4.98272717e-01 -7.04787672e-01 4.97817576e-01 1.03537418e-01 -1.29894829e+00 -3.97623837e-01 7.12109059e-02 4.16722029e-01 9.94936109e-01 -3.92020196e-01 -5.18128183e-03 1.20876670e-01 -8.38401258e-01 1.41177690e+00 8.72328520e-01 7.09178805e-01 -5.13818920e-01 6.04600430e-01 1.74744502e-01 -1.23588121e+00 -9.07637998e-02 -3.72227550e-01 -5.03244735e-02 -1.98292091e-01 5.67140579e-01 -9.75453675e-01 8.45032513e-01 1.03921115e+00 4.58769917e-01 -8.51370513e-01 1.42899394e+00 -1.69404387e-01 4.52114522e-01 -7.45348990e-01 7.13014044e-03 3.00865442e-01 -5.49598813e-01 7.22444177e-01 1.19439125e+00 3.05607319e-01 -2.32959852e-01 1.15946233e-01 9.43729460e-01 1.47246748e-01 -2.20469236e-02 -7.60802507e-01 1.38426602e-01 3.13445777e-01 1.69244635e+00 -1.00985968e+00 9.17245522e-02 -1.50151759e-01 5.80398560e-01 1.09689385e-01 -3.43522727e-02 -4.82305050e-01 -4.71230537e-01 5.09464979e-01 6.82209194e-01 3.25280905e-01 -4.54231858e-01 -4.75447565e-01 -7.71454573e-01 1.83212146e-01 -4.87309128e-01 1.89291596e-01 -8.43541861e-01 -8.70411038e-01 5.35227656e-01 -7.35246986e-02 -1.31038892e+00 3.19734484e-01 -7.78567493e-01 -9.60094392e-01 7.68077493e-01 -1.69428241e+00 -1.19554901e+00 -8.66913617e-01 -7.31989071e-02 2.54450947e-01 -5.46550751e-02 5.82162201e-01 -8.98750499e-02 -4.26588982e-01 3.03301420e-02 4.96959269e-01 -3.54077667e-01 3.82363349e-01 -1.27334845e+00 8.33274662e-01 8.16320956e-01 -1.87765554e-01 1.54719055e-01 6.32081449e-01 -9.39153790e-01 -1.70185840e+00 -1.60180056e+00 2.57614583e-01 -1.76443204e-01 4.71233338e-01 -5.62569797e-01 -9.91470158e-01 7.21007347e-01 -2.19914034e-01 2.51688473e-02 1.88898399e-01 -1.93429112e-01 3.80142890e-02 -2.28426725e-01 -1.50659287e+00 2.10534334e-01 6.37868822e-01 -2.74945736e-01 -3.71330738e-01 4.35956120e-01 2.63744295e-01 -7.23066688e-01 -5.66592693e-01 8.14265430e-01 5.10397255e-01 -1.20277095e+00 7.20594883e-01 9.17339697e-02 1.75369859e-01 -6.54021919e-01 -4.53464150e-01 -1.01636779e+00 4.47586477e-02 -4.70259577e-01 3.06014597e-01 8.12950671e-01 5.36253333e-01 -6.61947310e-01 1.01385105e+00 2.43755683e-01 -6.86371028e-01 -5.29651999e-01 -9.67642844e-01 -7.23569810e-01 -1.17944911e-01 -3.66310060e-01 4.82570708e-01 5.18077672e-01 -7.19019949e-01 -7.61821866e-02 1.93346038e-01 5.79682469e-01 5.23337364e-01 3.87848675e-01 9.82668221e-01 -1.44827521e+00 -3.70825678e-02 -1.60161480e-01 -2.51544863e-01 -5.36511779e-01 -4.51540649e-01 -9.65209842e-01 5.62264085e-01 -1.67214966e+00 -5.48105836e-01 -6.78411126e-01 4.79276359e-01 3.29187214e-01 3.89400423e-01 2.15354450e-02 -1.91061437e-01 3.62248331e-01 -2.69748479e-01 6.51461542e-01 6.34661317e-01 -9.32506323e-02 -1.95698723e-01 5.88744432e-02 -6.49055466e-02 6.72791481e-01 8.82372081e-01 -3.35742056e-01 5.61401658e-02 -6.90480173e-01 6.63662404e-02 -1.02817781e-01 5.92967212e-01 -1.60185933e+00 8.39499384e-02 -5.16582765e-02 3.58440310e-01 -1.04740238e+00 6.39508903e-01 -5.04180312e-01 1.59319997e-01 6.00060880e-01 2.16677964e-01 -3.56345624e-01 3.72900486e-01 6.63765371e-01 1.22587368e-01 -6.30030572e-01 1.01048517e+00 -3.77903372e-01 -6.78670585e-01 8.01647902e-02 -3.27349186e-01 -4.58939016e-01 1.07827401e+00 -3.96477968e-01 -1.37470141e-01 3.29365134e-01 -3.84535968e-01 4.56431776e-01 8.98969829e-01 2.06531674e-01 5.79586148e-01 -5.58674753e-01 -7.87156522e-01 3.38660121e-01 3.98106128e-01 5.45489907e-01 3.56871895e-02 5.39665699e-01 -1.20938468e+00 1.65760860e-01 -3.36097702e-02 -8.58602643e-01 -1.00035548e+00 -4.25711498e-02 9.10304129e-01 1.27201200e-01 -7.53432393e-01 7.18194008e-01 -1.32562816e-01 -9.10756946e-01 -1.72476470e-01 -3.98926556e-01 -9.32817981e-02 7.43525773e-02 3.25304121e-01 3.21044624e-01 7.73696363e-01 -6.87082171e-01 -1.95079148e-01 5.88307261e-01 1.18642546e-01 -3.74688730e-02 1.37955308e+00 7.61039034e-02 -1.98486507e-01 2.89240688e-01 6.43204093e-01 -5.14647722e-01 -1.32036328e+00 -1.92955613e-01 3.61053109e-01 -4.10538137e-01 2.01981977e-01 -6.64032757e-01 -9.72417653e-01 7.54372895e-01 8.00300717e-01 7.62875676e-02 7.83732593e-01 -1.53900325e-01 5.66459656e-01 8.27655971e-01 5.51493585e-01 -1.06567800e+00 -3.89214993e-01 6.27341807e-01 1.10613942e+00 -1.15175986e+00 2.14543462e-01 -3.37294340e-01 -3.41409087e-01 1.33579087e+00 5.42594492e-01 -3.07557285e-01 4.25468862e-01 3.41282971e-02 4.03436199e-02 -6.92851543e-01 -3.30250412e-01 -4.18817371e-01 5.83730526e-02 5.71058095e-01 -1.48569150e-02 2.39314750e-01 -9.67505723e-02 2.89642304e-01 -3.58223319e-01 -2.03333363e-01 5.96756697e-01 1.45852387e+00 -1.10706854e+00 -6.43683255e-01 -8.36718738e-01 7.17227876e-01 2.30549172e-01 3.79072160e-01 -6.37370884e-01 8.10278058e-01 1.99002996e-01 7.26499200e-01 2.14392960e-01 -3.48414123e-01 5.05813956e-01 -3.73772562e-01 4.11600977e-01 -6.45914316e-01 -6.30076945e-01 -1.66803464e-01 4.07661647e-01 -3.44034404e-01 -2.73430258e-01 -9.98400867e-01 -1.26657832e+00 2.54529297e-01 -3.46643418e-01 6.68687373e-02 1.31277812e+00 8.73165607e-01 1.58250600e-01 2.33542770e-01 5.61808586e-01 -1.45795524e+00 -3.45768869e-01 -1.06237733e+00 -7.07890093e-01 -4.98805307e-02 1.61379486e-01 -7.42014945e-01 -3.24082881e-01 -2.36273795e-01]	[8.279158592224121, -1.7710552215576172]
278902c8-f018-4e21-bc05-36f61ab78ad3	robust-speech-recognition-via-large-scale-1	2212.04356	null	https://arxiv.org/abs/2212.04356v1	https://arxiv.org/pdf/2212.04356v1.pdf	Robust Speech Recognition via Large-Scale Weak Supervision	We study the capabilities of speech processing systems trained simply to predict large amounts of transcripts of audio on the internet. When scaled to 680,000 hours of multilingual and multitask supervision, the resulting models generalize well to standard benchmarks and are often competitive with prior fully supervised results but in a zero-shot transfer setting without the need for any fine-tuning. When compared to humans, the models approach their accuracy and robustness. We are releasing models and inference code to serve as a foundation for further work on robust speech processing.	['Ilya Sutskever', 'Christine McLeavey', 'Greg Brockman', 'Tao Xu', 'Jong Wook Kim', 'Alec Radford']	2022-12-06	robust-speech-recognition-via-large-scale	https://openai.com/blog/whisper/	https://cdn.openai.com/papers/whisper.pdf	preprint-2022-9	['robust-speech-recognition']	['speech']	[ 1.89316079e-01 3.46412271e-01 -3.64692211e-02 -8.60218525e-01 -1.57089770e+00 -5.64942062e-01 7.43004918e-01 -1.29344538e-02 -5.31343162e-01 8.30924273e-01 5.36706388e-01 -4.40955788e-01 2.50562191e-01 -1.39786974e-01 -8.33222032e-01 -2.84623116e-01 -2.85290480e-01 8.15334201e-01 2.15128317e-01 -3.51479977e-01 -1.69803441e-01 5.93641587e-02 -1.33753633e+00 5.66523254e-01 2.96641260e-01 7.80960023e-01 3.68286334e-02 8.71887326e-01 2.06519470e-01 8.28832865e-01 -6.48478627e-01 -4.50010359e-01 1.23175770e-01 1.57780960e-01 -1.16159177e+00 1.52180359e-01 5.81035614e-01 -2.56891906e-01 -3.89516860e-01 8.78127098e-01 8.60965610e-01 3.51763219e-01 2.81158000e-01 -9.60254431e-01 -5.29209614e-01 9.66004968e-01 -1.26152530e-01 6.99066281e-01 6.30591571e-01 1.11830242e-01 1.08465242e+00 -1.11504591e+00 5.35622597e-01 1.44406116e+00 5.50560296e-01 4.45111573e-01 -1.13486981e+00 -6.36192799e-01 2.00384781e-01 9.68640745e-02 -1.24023151e+00 -1.53998578e+00 1.74328461e-01 -1.88259512e-01 1.56382537e+00 3.76120433e-02 -5.29899225e-02 1.66742456e+00 1.56866565e-01 7.13190079e-01 8.31108034e-01 -4.53010440e-01 2.01046914e-02 1.14263803e-01 9.87832993e-02 7.32125342e-01 -4.13372934e-01 -7.19768479e-02 -1.20841813e+00 -1.71125039e-01 2.17574179e-01 -5.81515729e-01 -2.51733184e-01 3.72175068e-01 -1.41785681e+00 4.94076818e-01 1.95497535e-02 9.49746966e-02 -2.31365129e-01 -1.00824565e-01 7.12803900e-01 8.27409148e-01 1.09147525e+00 2.56807566e-01 -7.89186120e-01 -6.49973929e-01 -1.01463306e+00 -2.82964289e-01 1.08808911e+00 1.39939165e+00 5.30281961e-01 2.35491008e-01 -2.19688434e-02 1.16087079e+00 4.82390672e-02 4.42689240e-01 6.44331157e-01 -9.58433568e-01 9.40399766e-01 -2.77519375e-01 -1.18566439e-01 -2.33954966e-01 -3.40729803e-01 -2.21530586e-01 -5.09111464e-01 -4.53798443e-01 1.52940333e-01 -4.94661063e-01 -8.23202252e-01 1.53553617e+00 -2.23076679e-02 2.42706940e-01 5.85721768e-02 5.45469880e-01 3.63437802e-01 9.50214326e-01 8.41396973e-02 -4.48874503e-01 1.22476351e+00 -1.39219475e+00 -7.18219876e-01 -6.94760203e-01 3.95710319e-01 -1.15461421e+00 1.34998953e+00 6.77433670e-01 -1.48131812e+00 -4.26721692e-01 -7.48874128e-01 -2.34033242e-01 -2.16455698e-01 -2.24336177e-01 4.67296720e-01 4.85184491e-01 -1.41630828e+00 5.22807479e-01 -1.02036428e+00 -5.20489633e-01 1.88474089e-01 2.07235545e-01 -4.83371705e-01 -3.91940698e-02 -1.19115078e+00 9.34245944e-01 1.24736436e-01 -4.56376970e-02 -1.32157004e+00 -5.49864411e-01 -7.37003803e-01 9.32754576e-03 2.29761854e-01 -3.76744211e-01 2.04961395e+00 -8.08384120e-01 -2.00114059e+00 7.62006998e-01 -3.27541828e-01 -6.07791483e-01 4.08005148e-01 -4.46040601e-01 -7.02895522e-01 2.75314003e-01 1.95176736e-01 5.98264635e-01 9.82874632e-01 -5.09059846e-01 -7.28447199e-01 -1.37797654e-01 -2.62556206e-02 2.72056669e-01 -5.35853207e-01 7.22228289e-01 -4.15215611e-01 -7.29041100e-01 -1.93908542e-01 -8.48687589e-01 -5.88586144e-02 -3.64069730e-01 -4.65988934e-01 -3.03381383e-01 5.07873833e-01 -9.14769709e-01 8.06406915e-01 -2.10948443e+00 1.92609951e-02 -1.83058023e-01 -2.46984288e-01 2.85939902e-01 -4.36634719e-01 5.55305243e-01 -7.50471605e-03 2.67424077e-01 -3.95799764e-02 -9.22399104e-01 9.94609073e-02 7.56821334e-02 -6.26475811e-01 4.21455562e-01 2.27752164e-01 6.34621382e-01 -8.89773905e-01 -3.52867514e-01 1.21916607e-02 2.66330183e-01 -5.23181677e-01 4.41317469e-01 -1.17252320e-01 3.92061472e-01 -1.02774344e-01 7.18318045e-01 3.42637450e-02 -1.83443874e-01 4.89597768e-02 2.18414009e-01 1.23578019e-01 1.06302702e+00 -6.65594459e-01 2.05126905e+00 -9.03121471e-01 7.10133970e-01 6.39535666e-01 -8.96752238e-01 6.02649748e-01 9.68187630e-01 2.96611004e-02 -3.83202702e-01 -5.60655072e-02 1.76071793e-01 -1.04068860e-01 -3.73459518e-01 3.82857442e-01 -4.42173362e-01 -1.69040561e-01 8.19719672e-01 6.60906076e-01 -4.16077763e-01 9.18129534e-02 4.21312451e-01 1.17909861e+00 -3.29769492e-01 9.56287906e-02 -2.03881100e-01 2.58572549e-01 -3.01437885e-01 2.87992746e-01 7.37986207e-01 -2.29605824e-01 4.03011799e-01 1.77438810e-01 -2.66129285e-01 -1.03252256e+00 -1.07434833e+00 -1.64777696e-01 1.94652438e+00 -6.27387047e-01 -4.87269968e-01 -8.19260359e-01 -4.49602991e-01 -2.67331243e-01 5.77026606e-01 -9.61544178e-03 3.99906449e-02 -3.07340294e-01 -5.44454932e-01 1.15285039e+00 6.14122629e-01 1.31103501e-01 -1.12501717e+00 2.52655357e-01 4.85596865e-01 -5.59945107e-01 -1.56546414e+00 -7.38430679e-01 4.96175617e-01 -6.19029701e-01 -4.85598594e-01 -3.81350219e-01 -9.17301357e-01 7.88425282e-02 1.52936414e-01 1.31098568e+00 -1.43329561e-01 2.19716914e-02 4.29363579e-01 -3.35491300e-01 -5.43499291e-01 -6.93205774e-01 5.00070870e-01 6.63080812e-01 -1.94874704e-01 2.69034833e-01 -7.57497072e-01 -1.87861342e-02 3.30104858e-01 -4.10154760e-01 -2.79447109e-01 4.78637904e-01 8.43282759e-01 3.37774545e-01 -3.03264380e-01 9.64438319e-01 -8.03423882e-01 8.91493976e-01 -7.14749515e-01 -3.03543121e-01 2.01556265e-01 -5.12304127e-01 -8.10574889e-02 7.21463263e-01 -2.53986299e-01 -1.20225096e+00 -6.96778670e-02 -4.20129269e-01 -3.17525983e-01 -4.96155202e-01 3.59992564e-01 1.28594622e-01 1.58633336e-01 6.53086424e-01 1.42104194e-01 -1.60698190e-01 -6.98671877e-01 4.76747990e-01 1.08460951e+00 5.79788506e-01 -7.09162116e-01 5.83175719e-01 1.55486152e-01 -7.03402340e-01 -1.06493998e+00 -1.10233176e+00 -4.83939260e-01 -6.33855879e-01 1.68272510e-01 7.39452660e-01 -1.32390416e+00 -3.60301435e-01 2.80570537e-01 -1.29915369e+00 -7.45072246e-01 -4.40006107e-02 5.40437281e-01 -7.50194371e-01 2.62186706e-01 -1.43327487e+00 -7.46599793e-01 -5.38743675e-01 -9.59860206e-01 1.31322753e+00 -4.78511989e-01 -2.39989027e-01 -1.04571557e+00 5.17195016e-02 6.59793615e-01 3.56739730e-01 -7.22737968e-01 5.21990061e-01 -1.00207376e+00 -3.20385724e-01 -1.84547141e-01 1.37602016e-01 6.73809707e-01 2.74622366e-02 8.88612941e-02 -1.53171575e+00 -4.54162359e-01 -1.44394219e-01 -9.80772078e-01 8.49987268e-01 9.07785296e-02 1.07874036e+00 -4.97398078e-01 -2.11121161e-02 4.22659993e-01 6.15461051e-01 -2.28408426e-01 -4.22351025e-02 5.93970120e-02 3.18722516e-01 6.08106613e-01 3.44096243e-01 3.10285777e-01 3.56873959e-01 5.67268491e-01 -4.01383080e-02 1.77690357e-01 -1.39927670e-01 -3.83535773e-01 8.59112322e-01 1.59866619e+00 1.76363718e-02 -3.42253238e-01 -1.11224055e+00 6.31995559e-01 -1.55061066e+00 -1.06121111e+00 2.41440073e-01 1.89970887e+00 1.15020764e+00 4.36629087e-01 8.89150351e-02 4.13630940e-02 7.79581130e-01 2.79771149e-01 -2.72825480e-01 -7.65210867e-01 -4.65111770e-02 4.67078030e-01 3.10299963e-01 8.69838238e-01 -1.17646813e+00 1.20369399e+00 8.57174683e+00 7.89012015e-01 -8.96865726e-01 5.69459677e-01 4.87791300e-01 -5.10907054e-01 -9.06233937e-02 -2.42725700e-01 -7.96935558e-01 2.91524321e-01 1.91116869e+00 -3.18829656e-01 9.08852816e-01 6.67165399e-01 2.69228607e-01 2.64227301e-01 -1.35041595e+00 6.86007619e-01 8.35504830e-02 -8.82273495e-01 -1.83260024e-01 -9.38412175e-02 5.02764046e-01 8.43715608e-01 9.76443291e-03 5.74211061e-01 7.06808031e-01 -9.54437673e-01 9.03569579e-01 1.42059043e-01 8.06640625e-01 -5.13154685e-01 6.08597577e-01 6.45318627e-01 -8.77227366e-01 -1.54716074e-01 -4.65291172e-01 -3.83408427e-01 6.24407887e-01 5.04685938e-01 -1.36541355e+00 3.36446673e-01 7.67741263e-01 6.40478253e-01 -4.50007349e-01 5.33604980e-01 -4.96476442e-01 1.16912174e+00 -4.61233765e-01 2.23299682e-01 2.69108474e-01 3.62980306e-01 4.87247884e-01 1.57936883e+00 1.82217494e-01 -9.69215855e-02 3.60479981e-01 8.93690661e-02 -5.25392473e-01 2.35795215e-01 -7.43466675e-01 -6.95891753e-02 7.98638403e-01 1.00460613e+00 -2.87662208e-01 -6.51390314e-01 -6.02043092e-01 9.71756101e-01 8.03241313e-01 4.09410208e-01 -5.65016210e-01 -3.13560426e-01 6.62439227e-01 -1.00431174e-01 6.37516379e-02 -3.92775983e-01 -1.04069635e-01 -1.30103040e+00 8.74269605e-02 -1.12590301e+00 3.57223332e-01 -9.03173804e-01 -1.59768677e+00 9.60636020e-01 -2.03428820e-01 -8.23489726e-01 -7.02208519e-01 -5.33689260e-01 -5.33558249e-01 7.65821517e-01 -1.50443220e+00 -9.11369920e-01 2.56338418e-01 8.08210731e-01 1.09381127e+00 -4.19283211e-01 1.28650284e+00 4.48605061e-01 -6.82328939e-01 6.50688529e-01 -4.67475168e-02 8.93498957e-02 1.25721359e+00 -1.25477302e+00 9.55567062e-01 8.59143078e-01 6.74701571e-01 5.65611243e-01 6.90752447e-01 -2.36812115e-01 -1.19855189e+00 -1.07428241e+00 1.11847150e+00 -7.65609384e-01 1.20685351e+00 -6.69069529e-01 -7.68961370e-01 1.28563190e+00 6.45382524e-01 7.39173293e-02 8.27176034e-01 7.34521389e-01 -5.36289990e-01 -1.74884260e-01 -7.89679885e-01 1.56137839e-01 1.17732120e+00 -1.21797931e+00 -8.74364674e-01 7.09150910e-01 1.13600242e+00 -3.66649300e-01 -1.06181538e+00 5.78707829e-02 3.37349445e-01 -5.73970616e-01 7.37615168e-01 -8.09543490e-01 2.54708290e-01 3.21523607e-01 -2.23028764e-01 -1.57496738e+00 -1.57686576e-01 -1.06739914e+00 -6.96634687e-03 1.13962543e+00 8.48779082e-01 -5.61175525e-01 2.96909362e-01 4.57942545e-01 -7.08895624e-01 -2.28700519e-01 -1.29340887e+00 -9.42157447e-01 1.86170951e-01 -7.48282254e-01 4.12401229e-01 9.25089836e-01 5.74275136e-01 7.45532930e-01 -4.75940377e-01 2.93569058e-01 5.13752282e-01 -3.76883924e-01 5.75086534e-01 -9.62620676e-01 -3.99302900e-01 -2.70010859e-01 -1.18626475e-01 -1.09040713e+00 6.92121983e-01 -1.02512014e+00 3.29940081e-01 -9.92640913e-01 -9.99266580e-02 -2.18103260e-01 -3.69865209e-01 7.56813824e-01 7.28680193e-02 3.81200463e-01 -1.69398123e-03 2.50410676e-01 -7.92464495e-01 4.67800379e-01 7.27433801e-01 -2.56293833e-01 4.66006212e-02 1.26600295e-01 -5.50599694e-01 6.72407627e-01 1.00688446e+00 -4.66877401e-01 -4.91113752e-01 -1.03412223e+00 -1.31365716e-01 1.69051215e-01 -1.88802388e-02 -9.70591009e-01 4.94399607e-01 1.22580677e-01 -1.87417805e-01 -7.50508085e-02 7.09794581e-01 -3.61171156e-01 -4.83683199e-01 -1.93726122e-01 -6.74149036e-01 1.36538208e-01 2.29601905e-01 6.34656250e-01 -5.64545035e-01 6.67933896e-02 4.89007980e-01 -6.85144067e-02 -3.68656397e-01 3.08874071e-01 -9.07249212e-01 2.94704735e-01 6.08340979e-01 4.97199893e-01 -4.41146910e-01 -8.26248884e-01 -1.09178674e+00 1.86322510e-01 4.80671786e-02 7.34995782e-01 3.00070792e-01 -1.01079488e+00 -8.55830491e-01 2.64515042e-01 3.58214192e-02 -1.67924032e-01 -1.70868278e-01 5.92813849e-01 -3.02631259e-02 7.12102830e-01 9.73204225e-02 -4.05688912e-01 -9.55645561e-01 4.79440212e-01 2.38684550e-01 -1.34387121e-01 -4.09385353e-01 1.07706535e+00 -3.29342753e-01 -6.32278860e-01 5.67677140e-01 -3.27066511e-01 3.55115861e-01 8.31120536e-02 7.47048438e-01 2.77198195e-01 7.15072513e-01 -4.72537518e-01 -2.70914465e-01 -2.57159889e-01 -1.97917953e-01 -6.72459066e-01 1.37715483e+00 -2.66121715e-01 6.76878169e-02 9.15562272e-01 1.02117944e+00 1.12294652e-01 -1.12608469e+00 -4.74654794e-01 2.36757025e-01 -7.82229155e-02 7.06693009e-02 -7.62452185e-01 -7.97020733e-01 1.12940371e+00 3.60809863e-02 3.27548206e-01 7.31478810e-01 2.25763664e-01 7.72988319e-01 9.61912513e-01 6.80509508e-01 -1.38182449e+00 1.19682841e-01 9.44921732e-01 9.25038278e-01 -1.18607914e+00 -3.61189663e-01 -2.93129683e-01 -7.62052655e-01 8.59318674e-01 3.71439844e-01 4.05695364e-02 8.51312220e-01 7.30282485e-01 3.90251309e-01 1.43299848e-01 -1.51657927e+00 5.35176583e-02 -1.53432041e-02 7.69679010e-01 8.30703318e-01 1.52476698e-01 3.16017866e-01 6.12037182e-01 -5.39465904e-01 -1.76715881e-01 3.42594028e-01 7.70389259e-01 -4.64294046e-01 -1.07992339e+00 -6.26511648e-02 3.54457498e-01 -7.23631978e-01 -5.44465601e-01 -1.96821690e-01 2.18044460e-01 -4.78718966e-01 1.45039606e+00 6.09267764e-02 -2.81072855e-01 4.12596017e-01 7.78224170e-01 2.29878217e-01 -1.11896968e+00 -6.98597968e-01 1.31599054e-01 7.26401210e-01 -5.64056873e-01 -2.79887199e-01 -7.31836677e-01 -8.79535437e-01 -3.51399869e-01 -2.31849581e-01 2.07135454e-01 6.04737818e-01 9.82111394e-01 4.54053998e-01 3.76029462e-01 7.31903434e-01 -1.09381008e+00 -1.05618584e+00 -1.55215847e+00 -4.99762833e-01 1.44344985e-01 3.85840029e-01 -2.39668652e-01 -4.68195528e-01 3.64183038e-01]	[14.298524856567383, 6.9539947509765625]
14930df2-af6c-4751-81f0-3013c36ce66f	towards-semi-supervised-universal-graph	2305.19598	null	https://arxiv.org/abs/2305.19598v1	https://arxiv.org/pdf/2305.19598v1.pdf	Towards Semi-supervised Universal Graph Classification	Graph neural networks have pushed state-of-the-arts in graph classifications recently. Typically, these methods are studied within the context of supervised end-to-end training, which necessities copious task-specific labels. However, in real-world circumstances, labeled data could be limited, and there could be a massive corpus of unlabeled data, even from unknown classes as a complementary. Towards this end, we study the problem of semi-supervised universal graph classification, which not only identifies graph samples which do not belong to known classes, but also classifies the remaining samples into their respective classes. This problem is challenging due to a severe lack of labels and potential class shifts. In this paper, we propose a novel graph neural network framework named UGNN, which makes the best of unlabeled data from the subgraph perspective. To tackle class shifts, we estimate the certainty of unlabeled graphs using multiple subgraphs, which facilities the discovery of unlabeled data from unknown categories. Moreover, we construct semantic prototypes in the embedding space for both known and unknown categories and utilize posterior prototype assignments inferred from the Sinkhorn-Knopp algorithm to learn from abundant unlabeled graphs across different subgraph views. Extensive experiments on six datasets verify the effectiveness of UGNN in different settings.	['Ming Zhang', 'Wei Ju', 'Yifang Qin', 'Yusheng Zhao', 'Xiao Luo']	2023-05-31	null	null	null	null	['graph-classification']	['graphs']	[ 2.44523793e-01 6.39773130e-01 -5.39137959e-01 -4.00249600e-01 -5.24873614e-01 -6.86335504e-01 3.66691262e-01 1.90612376e-01 7.90959373e-02 7.51483917e-01 -1.22800879e-01 -1.17358007e-01 -7.42586181e-02 -9.88185823e-01 -5.87834060e-01 -8.21753502e-01 2.15660051e-01 7.38116384e-01 1.45756751e-01 2.05193818e-01 -3.48814666e-01 2.03240603e-01 -1.31895673e+00 -3.27064507e-02 8.73875201e-01 9.82646227e-01 3.33242416e-02 1.07594602e-01 -2.53962278e-01 6.56085372e-01 -1.78146109e-01 -5.80687284e-01 2.18713164e-01 -2.78771877e-01 -6.51905477e-01 7.00420856e-01 5.42460859e-01 2.84155994e-03 -5.94700694e-01 1.45902181e+00 1.58421487e-01 8.56934115e-02 6.39862120e-01 -1.58658409e+00 -1.02296937e+00 6.79450750e-01 -6.04284048e-01 -3.00681666e-02 1.48718311e-02 -9.26161483e-02 1.24107695e+00 -8.44617724e-01 7.40231752e-01 1.04795074e+00 4.63368714e-01 6.90819263e-01 -1.21727157e+00 -5.59116900e-01 6.77478373e-01 1.80786729e-01 -1.51580930e+00 -1.24411575e-01 1.06888652e+00 -4.28864270e-01 3.14825386e-01 -7.43941870e-03 5.28653443e-01 1.10031199e+00 -5.55908680e-01 8.86773825e-01 8.47566187e-01 -2.39738256e-01 3.39732170e-01 3.57684493e-01 4.79685813e-01 8.32475245e-01 5.55669665e-01 -8.97152051e-02 -1.95777923e-01 -2.18936279e-02 4.09839451e-01 2.84788460e-01 -6.74218118e-01 -9.51329291e-01 -9.85746324e-01 8.08792531e-01 7.16197014e-01 1.62547052e-01 -1.60125792e-01 -3.04122895e-01 2.63284981e-01 2.06844434e-01 7.64471054e-01 1.29485056e-01 -5.02120614e-01 7.04685152e-01 -5.52484810e-01 -2.93419808e-01 9.42170620e-01 1.20205820e+00 1.11007047e+00 6.98391395e-03 7.21905082e-02 7.69404292e-01 4.40943092e-01 1.49071440e-01 4.58967656e-01 -1.61179528e-01 5.78342021e-01 1.13886929e+00 -2.02801719e-01 -1.20000231e+00 -3.38679969e-01 -6.87089264e-01 -9.38547432e-01 -3.14239800e-01 6.02019429e-01 -8.80572572e-02 -1.24070895e+00 1.79149616e+00 7.69094765e-01 4.52542067e-01 2.31009368e-02 9.64852273e-01 9.51575756e-01 4.25891429e-01 -6.13531955e-02 -2.21554503e-01 1.04039621e+00 -1.08790839e+00 -5.40450752e-01 -4.63954270e-01 6.06650174e-01 -7.26689622e-02 8.29117537e-01 2.28447512e-01 -2.69726068e-01 -5.52267611e-01 -9.66450214e-01 1.91490233e-01 -5.93967795e-01 2.41324410e-01 7.27652073e-01 6.65730655e-01 -8.59861314e-01 5.81699193e-01 -5.69613993e-01 -3.99086267e-01 5.72125196e-01 1.64010346e-01 -5.49051940e-01 -5.46933115e-01 -9.77906048e-01 3.34530354e-01 8.64665031e-01 3.27961266e-01 -9.65118647e-01 -2.34037578e-01 -1.09708846e+00 2.96892136e-01 1.16892505e+00 -3.10730696e-01 6.96840703e-01 -1.10084033e+00 -9.41974401e-01 8.73773217e-01 1.44829497e-01 -6.04373813e-02 3.77674043e-01 3.92006248e-01 -5.97387791e-01 4.51431759e-02 1.88781306e-01 3.14644247e-01 9.45964515e-01 -1.38693571e+00 -6.22216046e-01 -7.73694336e-01 2.65875489e-01 1.72917083e-01 -5.14948487e-01 -7.21278846e-01 -5.02912998e-01 -5.11225641e-01 5.44863641e-01 -9.79561150e-01 -1.77006677e-01 -1.17473103e-01 -6.73974395e-01 -4.71526682e-01 9.07014132e-01 -4.46093351e-01 9.16992247e-01 -2.14244246e+00 1.88143000e-01 2.91206598e-01 8.46741498e-01 5.59834242e-02 -4.82799858e-02 2.29966983e-01 -6.68926761e-02 -1.09845446e-02 -2.54894674e-01 -1.64739937e-01 6.92902431e-02 4.21499312e-01 -2.26399571e-01 5.76283574e-01 2.49577507e-01 8.16244066e-01 -1.11917961e+00 -4.29620057e-01 -1.01682507e-01 -1.65220667e-02 -2.28719056e-01 3.26373965e-01 -2.65296549e-01 4.77470130e-01 -5.91440737e-01 1.05644536e+00 7.24610090e-01 -8.20150077e-01 6.48561537e-01 -2.03079373e-01 5.90672374e-01 -2.90559560e-01 -1.35973275e+00 1.38594365e+00 -7.58775547e-02 2.75621057e-01 -1.52998744e-02 -1.67180979e+00 8.57686579e-01 1.50680631e-01 1.03092328e-01 -7.13204145e-02 1.49854451e-01 2.90409684e-01 -1.28346115e-01 -4.35859382e-01 2.24066809e-01 -2.05529690e-01 1.03839021e-02 3.12885672e-01 5.54566741e-01 2.00112417e-01 4.05656010e-01 2.77387440e-01 9.99788225e-01 8.28142539e-02 3.48739088e-01 -9.90890265e-02 4.36315894e-01 -8.09619799e-02 6.92039967e-01 5.86260974e-01 -4.51628596e-01 5.30403197e-01 5.72223425e-01 -5.06509244e-01 -6.13362610e-01 -1.11605263e+00 1.38400212e-01 1.07795632e+00 5.82415462e-01 -2.34331816e-01 -4.12560463e-01 -1.57200134e+00 -9.96731408e-03 3.37897718e-01 -6.11500680e-01 -3.54386955e-01 -1.20738819e-01 -8.16040277e-01 9.15528834e-02 3.72544557e-01 1.18658148e-01 -8.49332988e-01 1.51050851e-01 1.00725703e-01 -1.07539646e-01 -1.29616737e+00 -3.68694544e-01 1.97081968e-01 -7.61346817e-01 -1.72348690e+00 -5.26117504e-01 -1.22936082e+00 1.25671446e+00 5.32144427e-01 1.09581089e+00 8.95375237e-02 -1.12374522e-01 3.71684283e-01 -3.92136693e-01 -9.39985365e-02 -2.31185734e-01 9.80139151e-02 5.44978306e-04 4.55980778e-01 5.30473232e-01 -5.40164888e-01 -3.17197591e-01 4.30101186e-01 -8.29813421e-01 -2.67067496e-02 5.25609851e-01 9.98718977e-01 8.05030286e-01 2.51309782e-01 8.53318751e-01 -1.38024426e+00 4.12014425e-01 -9.31636393e-01 -5.76247513e-01 5.43136179e-01 -7.25962102e-01 4.77533974e-03 9.15574908e-01 -6.34944618e-01 -7.52981007e-01 8.57374370e-02 4.69990402e-01 -7.70670712e-01 -2.20264062e-01 9.00841475e-01 -5.41953146e-01 -3.42622250e-02 5.36006868e-01 1.84194699e-01 -1.40749887e-01 -3.20684254e-01 5.66188991e-01 5.21239519e-01 4.08599436e-01 -3.98680240e-01 9.34121490e-01 4.71483231e-01 -1.08448543e-01 -6.67209864e-01 -1.11574030e+00 -5.66998005e-01 -6.20140433e-01 -3.66846949e-01 4.43263918e-01 -8.69677484e-01 -1.28573135e-01 3.09205085e-01 -6.30892634e-01 -1.60427809e-01 -3.18669051e-01 4.97639328e-01 -1.00966334e-01 6.74014509e-01 -3.57795924e-01 -6.17348850e-01 -3.37764211e-02 -1.05267417e+00 1.01517749e+00 4.96127844e-01 2.39322975e-01 -1.01999044e+00 -2.39595637e-01 2.93309212e-01 -1.59439936e-01 2.87760586e-01 9.64868307e-01 -1.32170284e+00 -7.01440156e-01 -5.28171360e-01 -4.73797798e-01 3.54132205e-01 3.21833819e-01 -3.26609462e-01 -9.86564159e-01 -5.83748102e-01 -1.85677364e-01 -6.53546214e-01 8.59772503e-01 1.40539780e-01 1.12116551e+00 -2.85908550e-01 -6.76007390e-01 4.95263278e-01 1.43275332e+00 -1.18055239e-01 -2.72697280e-03 -2.95177579e-01 1.24775803e+00 8.21527898e-01 4.40813273e-01 1.65179878e-01 5.61308742e-01 1.96517915e-01 7.50413120e-01 1.07707322e-01 -2.50380579e-02 -5.18203080e-01 -1.65651679e-01 9.58225906e-01 3.04364830e-01 -6.90912008e-01 -8.27654719e-01 6.65525556e-01 -1.86094856e+00 -4.92277265e-01 4.65107486e-02 2.11258197e+00 3.93623263e-01 2.01261237e-01 -1.12618871e-01 8.37374926e-02 1.41503417e+00 2.73489654e-01 -8.52709174e-01 4.52120721e-01 1.15591332e-01 -2.37376586e-01 4.16069776e-01 3.14747803e-02 -1.28606737e+00 7.99135149e-01 4.49733353e+00 8.45280766e-01 -9.92636025e-01 1.05236962e-01 7.91317284e-01 3.90391469e-01 -3.00318629e-01 1.24383040e-01 -6.34511054e-01 4.46926534e-01 6.50792062e-01 -1.85283199e-01 5.55454433e-01 1.23095059e+00 -4.31610048e-01 3.24896336e-01 -1.23044682e+00 9.88074481e-01 3.19915175e-01 -9.76937950e-01 9.23834741e-02 5.67587912e-02 8.52290452e-01 1.22925200e-01 -3.05984229e-01 7.82129586e-01 4.23391819e-01 -6.85520828e-01 4.45511580e-01 1.52427152e-01 7.91021347e-01 -5.72489023e-01 6.34247482e-01 6.04488075e-01 -1.33778167e+00 -1.67158812e-01 -6.20308697e-01 1.87747076e-01 -1.44048229e-01 7.01733053e-01 -9.44998562e-01 7.84772158e-01 3.46306682e-01 1.03087831e+00 -5.26301920e-01 9.05422807e-01 -4.52267796e-01 6.00566208e-01 -2.71119267e-01 1.14010647e-02 2.41692215e-01 -4.32362705e-01 2.87213922e-01 4.82755870e-01 3.30515712e-01 1.44344598e-01 8.34722042e-01 7.89635301e-01 -6.59112573e-01 2.06369132e-01 -7.94669569e-01 -5.09835303e-01 5.11697114e-01 1.42737448e+00 -1.16617835e+00 -4.15980428e-01 -5.25471210e-01 7.67761350e-01 8.43314469e-01 5.35911441e-01 -6.84227586e-01 -3.04507494e-01 3.25998552e-02 -7.68284127e-02 2.44161084e-01 1.29410997e-01 2.64034003e-01 -1.66707706e+00 2.73168474e-01 -5.79440057e-01 8.34192634e-01 -4.29011315e-01 -1.76474214e+00 5.25598347e-01 -1.96403772e-01 -1.33719087e+00 1.22643717e-01 -7.78285027e-01 -5.48389256e-01 3.55473071e-01 -1.47537243e+00 -1.35865867e+00 -5.52018881e-01 4.45562422e-01 4.74870026e-01 -2.38451660e-01 5.35460949e-01 3.09450299e-01 -6.85719132e-01 5.48751771e-01 2.96992123e-01 3.40405732e-01 5.14910221e-01 -1.33225894e+00 8.12364072e-02 9.67076182e-01 4.36538517e-01 2.49637574e-01 3.25349897e-01 -7.20409811e-01 -1.45171773e+00 -1.46042681e+00 5.93157768e-01 -1.21997997e-01 8.20834994e-01 -4.48668271e-01 -1.19396949e+00 8.36243868e-01 -4.28679377e-01 7.29068279e-01 6.48150086e-01 3.26419562e-01 -4.03175741e-01 6.70268992e-03 -1.11946547e+00 4.60540712e-01 1.22133350e+00 -6.39025807e-01 -3.54763627e-01 6.99330091e-01 8.28460455e-01 -3.23314041e-01 -6.67847753e-01 4.63392019e-01 1.48949757e-01 -5.25104165e-01 6.22592449e-01 -8.46864402e-01 1.33965537e-01 -4.25680876e-01 -1.55654028e-01 -1.60298908e+00 -1.13442838e-01 -6.01864457e-02 -3.04226220e-01 1.29520035e+00 3.41728479e-01 -7.94458866e-01 1.14692962e+00 4.56224978e-01 -1.94277734e-01 -5.90450227e-01 -9.41345453e-01 -7.31078267e-01 -3.69238377e-01 -2.10416913e-01 5.31996787e-01 1.41366184e+00 2.52035838e-02 6.11442506e-01 -2.84411550e-01 4.69148487e-01 7.43229568e-01 5.04120409e-01 6.22229755e-01 -1.50257087e+00 -3.20480525e-01 -2.00386778e-01 -7.33195007e-01 -8.15125346e-01 5.89962721e-01 -1.41428816e+00 6.99178427e-02 -1.54424584e+00 2.29017317e-01 -4.22340959e-01 -4.82610643e-01 6.38384342e-01 -5.27547300e-01 1.02271929e-01 -8.19294751e-02 1.01862334e-01 -8.06024194e-01 7.65994251e-01 1.10732949e+00 -6.79603815e-01 7.10312184e-03 3.20181996e-02 -8.18621039e-01 7.28790641e-01 6.19480968e-01 -4.69956428e-01 -8.79367471e-01 -3.38001996e-01 1.01661645e-01 -2.47847941e-02 3.10212046e-01 -8.24099481e-01 8.92235413e-02 -1.13777600e-01 1.46476239e-01 -3.75847608e-01 -9.70477238e-03 -9.60832596e-01 5.95540106e-02 1.00501396e-01 -2.00068384e-01 -4.67177451e-01 -3.23044121e-01 1.40818369e+00 -1.81028396e-01 -1.92906350e-01 6.32897139e-01 -1.92008883e-01 -9.58106697e-01 8.80047917e-01 2.58820385e-01 3.69392365e-01 1.09954214e+00 -1.77475214e-01 -5.03638089e-01 -1.33888140e-01 -1.09389758e+00 5.09389818e-01 4.87815529e-01 4.89677757e-01 5.01088560e-01 -1.42649889e+00 -5.14629006e-01 2.45335579e-01 6.86565876e-01 3.31523508e-01 4.37171698e-01 5.11695027e-01 -6.31000325e-02 1.22822700e-02 2.45907053e-01 -6.68945253e-01 -8.60509872e-01 1.09750032e+00 2.04619765e-01 -3.23695719e-01 -4.69175905e-01 8.18198740e-01 5.85697353e-01 -8.90908837e-01 1.87094390e-01 -2.21632663e-02 -4.04804528e-01 2.82212049e-01 1.81645766e-01 1.41186416e-01 1.56409235e-03 -6.53519809e-01 -1.88087583e-01 1.70614600e-01 -7.39516765e-02 6.27915204e-01 1.14771569e+00 -3.26469958e-01 -1.28398817e-02 3.67814928e-01 1.24756002e+00 -2.84243047e-01 -1.13495052e+00 -7.30129898e-01 1.34625912e-01 -5.21129012e-01 -1.48449287e-01 -4.63413686e-01 -1.47048497e+00 7.35006630e-01 3.61229897e-01 5.53401887e-01 8.34674597e-01 2.33979553e-01 5.10206461e-01 3.82703036e-01 5.30786693e-01 -9.64233279e-01 -5.65308630e-02 1.09140605e-01 3.83248061e-01 -1.65399659e+00 -1.70630142e-01 -7.17749000e-01 -4.73675579e-01 1.09962356e+00 9.31545079e-01 -3.49136628e-02 7.53105342e-01 -5.41737437e-01 -1.73112020e-01 -4.26856935e-01 -5.53289771e-01 -3.23478729e-01 3.23344529e-01 6.75583065e-01 1.41651519e-02 4.00097430e-01 -1.11546181e-01 6.77020013e-01 3.14438224e-01 -2.27667242e-01 5.48906088e-01 7.67742634e-01 -2.65948713e-01 -7.78230011e-01 -4.07308228e-02 9.12865341e-01 -1.87559381e-01 2.50101909e-02 -3.21940750e-01 7.02758253e-01 1.36718929e-01 9.85782146e-01 -3.50691319e-01 -4.71178472e-01 9.49976295e-02 5.09381630e-02 2.41479725e-01 -8.95842254e-01 1.70059547e-01 -9.40818340e-02 1.75506085e-01 -1.55951768e-01 -2.88945168e-01 -2.09411934e-01 -8.79815698e-01 1.65378913e-01 -7.21361220e-01 4.05802757e-01 2.51256436e-01 9.77224708e-01 2.14911163e-01 5.27329981e-01 7.61931837e-01 -8.13226998e-01 -7.15233207e-01 -8.76144230e-01 -1.00502348e+00 4.83900577e-01 1.15807965e-01 -8.72135997e-01 -5.75132430e-01 -9.00955349e-02]	[7.379183769226074, 6.151130676269531]
615ae3aa-fa7c-48a3-8cf6-f59ba3149c94	automatic-gaze-analysis-a-survey-of	2108.05479	null	https://arxiv.org/abs/2108.05479v3	https://arxiv.org/pdf/2108.05479v3.pdf	Automatic Gaze Analysis: A Survey of Deep Learning based Approaches	Eye gaze analysis is an important research problem in the field of Computer Vision and Human-Computer Interaction. Even with notable progress in the last 10 years, automatic gaze analysis still remains challenging due to the uniqueness of eye appearance, eye-head interplay, occlusion, image quality, and illumination conditions. There are several open questions, including what are the important cues to interpret gaze direction in an unconstrained environment without prior knowledge and how to encode them in real-time. We review the progress across a range of gaze analysis tasks and applications to elucidate these fundamental questions, identify effective methods in gaze analysis, and provide possible future directions. We analyze recent gaze estimation and segmentation methods, especially in the unsupervised and weakly supervised domain, based on their advantages and reported evaluation metrics. Our analysis shows that the development of a robust and generic gaze analysis method still needs to address real-world challenges such as unconstrained setup and learning with less supervision. We conclude by discussing future research directions for designing a real-world gaze analysis system that can propagate to other domains including Computer Vision, Augmented Reality (AR), Virtual Reality (VR), and Human Computer Interaction (HCI). Project Page: https://github.com/i-am-shreya/EyeGazeSurvey}{https://github.com/i-am-shreya/EyeGazeSurvey	['Qiang Ji', 'Jarrod Knibbe', 'Munawar Hayat', 'Abhinav Dhall', 'Shreya Ghosh']	2021-08-12	null	null	null	null	['gaze-estimation']	['computer-vision']	[ 2.07050040e-01 -4.43109125e-02 -1.62424520e-01 -3.65402848e-01 -2.16741681e-01 -4.86799896e-01 9.28016379e-02 -3.16144288e-01 -2.20636383e-01 4.52521384e-01 1.27980923e-02 -4.19528246e-01 -1.65742978e-01 2.06231385e-01 -3.13592911e-01 -5.80959678e-01 1.66241691e-01 -1.90263197e-01 3.91030125e-02 -2.42406338e-01 6.19633913e-01 8.71355012e-02 -2.33706617e+00 -2.76059330e-01 9.09266651e-01 8.47986400e-01 2.22086951e-01 9.38267708e-01 3.33228707e-01 5.34709632e-01 -4.22096968e-01 -1.90283284e-01 -6.45325258e-02 -5.58402598e-01 -6.88949466e-01 1.55802175e-01 7.93854415e-01 -1.88861296e-01 2.74482101e-01 1.09852970e+00 6.45636976e-01 1.23209551e-01 3.09189498e-01 -1.71161950e+00 -8.36678147e-01 -3.34280044e-01 -9.77665484e-01 6.10023618e-01 9.51951325e-01 4.00798559e-01 7.60076344e-01 -7.59401739e-01 6.29649520e-01 8.67212176e-01 2.66373754e-01 8.07774961e-01 -9.21694934e-01 -8.96484017e-01 2.92159885e-01 4.28785115e-01 -1.31155384e+00 -7.46144235e-01 6.76653564e-01 -6.34566545e-01 5.53215921e-01 5.90085864e-01 3.77812654e-01 1.04139948e+00 5.12518845e-02 9.56445992e-01 1.36589801e+00 -8.43472838e-01 -8.47132206e-02 4.78311568e-01 5.45658708e-01 6.53821945e-01 1.48530081e-01 4.93751429e-02 -8.11102211e-01 1.93810776e-01 5.50216973e-01 -1.08589955e-01 -8.52625072e-01 -4.36658084e-01 -1.26529050e+00 5.74896812e-01 2.98427641e-01 1.52887907e-02 -1.34730756e-01 -2.94101000e-01 5.82011081e-02 2.74371326e-01 5.70236266e-01 4.33790445e-01 -4.29906934e-01 -4.57971215e-01 -7.81377912e-01 2.09216624e-01 6.69888318e-01 1.05300462e+00 6.13752961e-01 -3.00737768e-01 2.82944776e-02 6.20446801e-01 6.63419366e-01 5.98165929e-01 1.98256612e-01 -9.80594397e-01 1.37923017e-01 2.61294663e-01 2.92081892e-01 -7.14745700e-01 -4.58398581e-01 1.20983489e-01 -1.60351828e-01 7.82997608e-01 7.05155253e-01 -1.84088200e-01 -6.74339235e-01 1.69667041e+00 5.23630619e-01 4.78925519e-02 -4.41925317e-01 1.27098823e+00 1.08986282e+00 5.01754545e-02 -7.69625604e-02 -5.47522902e-01 1.70412064e+00 -8.28891098e-01 -1.05812860e+00 -2.90879160e-01 6.34422421e-01 -1.04511690e+00 1.56710672e+00 6.53768063e-01 -1.10953391e+00 -4.44124609e-01 -8.92377257e-01 -3.93178910e-01 -2.16142684e-01 1.56929448e-01 5.51222384e-01 1.03924048e+00 -1.28984094e+00 -4.84701470e-02 -7.70639360e-01 -8.00181329e-01 2.44859248e-01 6.44859552e-01 -2.00658813e-01 1.15762189e-01 -6.44335806e-01 8.74102592e-01 -1.94182679e-01 2.17749253e-01 -1.76974609e-01 -3.59357446e-01 -1.01987183e+00 -3.10636520e-01 5.24573445e-01 -6.37131274e-01 1.38544118e+00 -1.20663393e+00 -1.67105973e+00 1.20092273e+00 -1.00783873e+00 3.40867490e-01 1.06279232e-01 -3.62636149e-01 -4.86739963e-01 -8.55399445e-02 -1.61553293e-01 5.21637499e-01 7.96904743e-01 -1.18836343e+00 -7.68691123e-01 -7.67442346e-01 1.28237605e-01 3.92801642e-01 -1.99758261e-01 7.42354155e-01 -6.00302577e-01 -1.08219147e-01 -1.07437186e-02 -1.29793370e+00 2.96454161e-01 -3.13368887e-02 -2.69392669e-01 -3.68381977e-01 8.71375799e-01 -5.18943191e-01 1.52171457e+00 -2.18446016e+00 1.82142720e-01 -1.07651204e-01 7.50197649e-01 1.29251480e-01 2.90232778e-01 4.23012115e-02 -2.70623356e-01 -3.47765721e-02 2.25236997e-01 -5.65301836e-01 -1.11587495e-01 -3.30712348e-01 4.21180166e-02 6.61581635e-01 -3.01107138e-01 9.53256190e-01 -9.21041012e-01 -4.68982905e-01 4.05226141e-01 5.83392620e-01 -1.70329362e-01 1.87978268e-01 1.21052071e-01 7.49150693e-01 -2.21541569e-01 8.10746789e-01 5.71233571e-01 -5.47746003e-01 -2.23350361e-01 1.11189425e-01 -4.59189594e-01 1.62203640e-01 -1.13596010e+00 1.43056977e+00 -1.87670663e-01 1.31926942e+00 6.43856898e-02 -1.36205897e-01 4.47884768e-01 2.01358989e-01 1.36234373e-01 -7.81444073e-01 6.40116870e-01 4.33214121e-02 5.37787452e-02 -9.59689975e-01 6.72178388e-01 5.02237439e-01 3.73088926e-01 7.93061614e-01 -1.01241037e-01 4.08091575e-01 1.97848961e-01 -2.60903221e-02 4.81005132e-01 3.41707319e-01 3.74433309e-01 -4.56940383e-02 4.57422704e-01 -3.19758415e-01 1.00145787e-01 2.39224389e-01 -8.40505242e-01 8.80258560e-01 4.25527900e-01 -1.61024600e-01 -5.53651214e-01 -6.06078684e-01 -1.54292062e-01 1.71166039e+00 4.51243937e-01 -3.87683332e-01 -1.02897620e+00 -4.04634595e-01 -4.71798509e-01 5.80260396e-01 -7.43615091e-01 1.42328799e-01 -1.87665999e-01 -4.78480160e-01 -5.12082614e-02 1.78503692e-01 -1.32898435e-01 -1.21920323e+00 -9.03932631e-01 -7.24584341e-01 -2.25468189e-01 -8.63226235e-01 -6.03625178e-01 -3.52011085e-01 -6.19244277e-01 -1.27231836e+00 -7.60975480e-01 -5.21420717e-01 7.89087892e-01 7.48674452e-01 1.05026507e+00 2.01817244e-01 -1.16000354e-01 6.33749366e-01 -3.72220844e-01 -8.03902566e-01 1.58862069e-01 6.62029088e-02 1.88917071e-01 -5.25090992e-02 9.72745299e-01 -3.82436961e-02 -8.94910574e-01 6.69765830e-01 -4.41535801e-01 -1.21825576e-01 1.40543297e-01 5.19040346e-01 2.90321916e-01 -5.32170236e-01 2.12787138e-03 -9.59261775e-01 8.20219874e-01 -3.33647728e-01 -6.68384671e-01 3.03455085e-01 -8.53153646e-01 -3.24414790e-01 -3.69822115e-01 -4.21740979e-01 -9.79740798e-01 -2.91261166e-01 2.17699930e-01 -5.08598030e-01 -5.50663471e-01 2.02152103e-01 -2.30983309e-02 -2.83138841e-01 1.10663497e+00 -2.19951645e-01 2.89039046e-01 -2.73941189e-01 1.51052475e-01 1.10635686e+00 2.16494456e-01 -1.68292582e-01 4.06917870e-01 5.10458231e-01 -4.03806418e-01 -9.74783003e-01 -9.30564821e-01 -7.57645965e-01 -6.64375722e-01 -5.32899499e-01 8.36133182e-01 -7.74573922e-01 -1.28229821e+00 6.06558800e-01 -9.10229802e-01 -3.04079086e-01 1.74215272e-01 6.56920195e-01 -3.69337320e-01 3.64882112e-01 -1.17111869e-01 -1.09488583e+00 -3.63263279e-01 -1.37624836e+00 1.13683593e+00 9.10798609e-01 -6.68588102e-01 -9.40934896e-01 1.51853859e-01 7.38807738e-01 1.50746301e-01 3.21688205e-02 1.49894491e-01 -1.16532743e-01 -5.69590271e-01 -4.72519360e-02 -2.50309259e-01 1.02242924e-01 1.12984829e-01 4.28693056e-01 -1.26874292e+00 -4.08449233e-01 -4.76291701e-02 -1.29863724e-01 2.55247951e-01 8.00526023e-01 8.69892836e-01 2.35013619e-01 -4.18935686e-01 6.36344373e-01 9.57232356e-01 8.19769278e-02 6.77829206e-01 4.83900309e-01 8.75420034e-01 9.53272700e-01 9.17738080e-01 1.17044643e-01 8.46918464e-01 6.62893295e-01 4.54826444e-01 -1.84812456e-01 -6.87398389e-02 2.89765537e-01 2.58292168e-01 3.50044280e-01 -5.65863788e-01 -4.25964445e-01 -9.81700659e-01 3.88124138e-01 -1.72201920e+00 -8.59540641e-01 -5.79797268e-01 2.54909348e+00 5.44763684e-01 -1.05926096e-01 4.89113361e-01 1.61355361e-01 7.23765612e-01 -7.82685950e-02 -5.56009769e-01 -4.04227167e-01 1.82485372e-01 -3.09009217e-02 2.86309421e-01 4.04003739e-01 -9.25230503e-01 8.04363966e-01 5.88630724e+00 3.19775119e-02 -1.57694447e+00 2.98430353e-01 3.23412627e-01 -2.76029736e-01 -4.97652777e-02 -9.80589539e-02 -8.84362698e-01 6.27274513e-01 8.68883252e-01 5.29194064e-02 5.06779373e-01 5.95151722e-01 4.26157415e-01 -6.85607016e-01 -1.14029551e+00 1.33002114e+00 5.43009818e-01 -6.56097710e-01 -9.15977061e-01 3.51386487e-01 3.92643452e-01 2.82061934e-01 5.54799259e-01 -9.55696777e-02 -3.04425538e-01 -9.50283289e-01 5.30179918e-01 6.41733766e-01 1.02975619e+00 -3.76490384e-01 5.13665378e-01 6.72521517e-02 -6.95851505e-01 -1.21766059e-02 1.56115353e-01 -3.33901346e-01 5.21772653e-02 -1.30160019e-01 -5.71658611e-01 1.38504848e-01 1.07695413e+00 7.61259556e-01 -8.81610334e-01 1.10931325e+00 -2.78059453e-01 4.31665808e-01 -1.35024041e-01 -1.57897130e-01 -4.79996800e-01 -2.74957240e-01 6.06059432e-01 6.78495586e-01 -8.94343015e-03 1.22546077e-01 -4.69585925e-01 6.19403541e-01 2.64494061e-01 -5.65650687e-02 -5.12337327e-01 1.91885442e-01 4.10219789e-01 1.28599048e+00 -7.46990442e-01 2.74874926e-01 -6.16087735e-01 7.80109465e-01 8.07388499e-02 8.04706037e-01 -6.75086617e-01 -4.70104784e-01 8.48914504e-01 4.75316435e-01 2.51033660e-02 -1.90023422e-01 -4.98484641e-01 -1.16737771e+00 2.06229210e-01 -9.88640308e-01 3.35714251e-01 -1.29653490e+00 -8.17718387e-01 6.87192976e-01 3.00177503e-02 -1.37527692e+00 -4.45952624e-01 -6.15247428e-01 -4.28515434e-01 1.03640866e+00 -1.58393764e+00 -1.17606151e+00 -8.49963009e-01 7.89312661e-01 4.11615193e-01 -7.99980611e-02 7.32365727e-01 1.96060672e-01 -9.01670933e-01 8.62966835e-01 -1.32224247e-01 -1.92919731e-01 1.15158963e+00 -1.18136871e+00 1.50803730e-01 7.80358672e-01 -5.66721894e-03 9.00588095e-01 1.09124565e+00 -5.65247647e-02 -1.39059436e+00 -1.84251636e-01 8.35704863e-01 -1.34492838e+00 3.91814500e-01 -4.86364275e-01 -8.24693680e-01 8.65063965e-01 5.52914321e-01 -2.67197005e-02 9.92273748e-01 6.75447941e-01 -1.53678060e-01 1.76451340e-01 -8.63612711e-01 8.00277114e-01 8.84426117e-01 -6.06604874e-01 -4.10379916e-01 -1.04388148e-02 4.04505521e-01 -8.28360200e-01 -4.78189379e-01 1.22526437e-01 8.70967269e-01 -1.24560392e+00 6.04846895e-01 -2.00996727e-01 1.20896608e-01 -4.79672045e-01 3.83445680e-01 -8.38845670e-01 2.30742991e-02 -1.01724792e+00 -2.67222315e-01 1.11057425e+00 2.78587759e-01 -8.83277416e-01 6.14259481e-01 1.13825548e+00 2.04533890e-01 -5.98263264e-01 -3.04990679e-01 -1.61819041e-01 -4.34220016e-01 -5.12099922e-01 3.31108987e-01 8.89852405e-01 1.96532860e-01 4.87605959e-01 -3.37018847e-01 1.63227260e-01 6.54240429e-01 -4.84795310e-02 1.23797822e+00 -1.47919965e+00 1.67583793e-01 -4.30459023e-01 -4.83330131e-01 -9.47425246e-01 3.94477546e-02 -7.41088018e-02 -1.11620411e-01 -1.28865767e+00 6.02523983e-02 -1.34368300e-01 -2.96050370e-01 1.67143866e-01 -4.07153249e-01 4.58451033e-01 1.66758552e-01 4.77389961e-01 -8.76031458e-01 1.55613720e-02 1.25249279e+00 5.03750324e-01 -5.85980177e-01 3.52778047e-01 -1.05847347e+00 6.10649586e-01 4.67846036e-01 -2.18448296e-01 -4.69617516e-01 -4.88783330e-01 5.74985802e-01 -1.86652407e-01 3.81709844e-01 -5.21171689e-01 5.85734904e-01 -2.38644183e-01 9.36027020e-02 -4.39744145e-01 2.39930391e-01 -5.95159352e-01 -7.49038905e-02 -2.93310046e-01 9.91160125e-02 1.48690358e-01 2.05957457e-01 3.28869283e-01 -1.23520657e-01 -7.33614191e-02 5.78734279e-01 1.99770555e-01 -7.63424277e-01 1.19151801e-01 3.78132053e-02 7.16944486e-02 1.18711877e+00 -8.88321519e-01 -5.24890363e-01 -5.03845870e-01 -6.72194719e-01 4.12586480e-01 9.77193713e-01 7.43059695e-01 3.86512458e-01 -6.34682357e-01 -2.99572289e-01 3.70773643e-01 4.97231811e-01 -1.07992768e-01 3.36923957e-01 1.27230215e+00 -3.47374707e-01 4.40476775e-01 -3.24758381e-01 -1.00078607e+00 -1.78744519e+00 4.69501108e-01 1.86008066e-01 5.28188527e-01 -1.30948454e-01 9.77990985e-01 3.75149161e-01 9.25957784e-03 4.01610553e-01 -1.18068628e-01 -6.41562045e-01 1.61353141e-01 7.17250645e-01 4.81132805e-01 6.26662821e-02 -1.11147952e+00 -4.01578218e-01 7.29672968e-01 -5.49199469e-02 4.50966060e-02 1.03891563e+00 -9.78856862e-01 -8.89263451e-02 6.40495062e-01 8.10013115e-01 6.86718300e-02 -9.59751964e-01 -7.56753981e-02 -9.94428918e-02 -8.11784089e-01 1.83171958e-01 -6.29668117e-01 -8.96674752e-01 9.53431726e-01 9.75880563e-01 2.79046774e-01 1.37134862e+00 2.48294711e-01 2.48806387e-01 -9.54829007e-02 2.10108116e-01 -9.53330934e-01 -5.96181452e-02 3.81215423e-01 6.82524741e-01 -1.82408500e+00 7.14175105e-02 -4.12357152e-01 -1.01957083e+00 7.65039206e-01 7.50511587e-01 3.83232266e-01 9.80363429e-01 -1.71209976e-01 6.02315664e-01 -6.27905965e-01 -4.74746585e-01 -6.76652431e-01 6.61867976e-01 7.03135669e-01 8.75090599e-01 -1.71830356e-01 -1.64744735e-01 2.05270007e-01 -4.00552839e-01 2.82866538e-01 6.46896482e-01 9.92237151e-01 -4.32584919e-02 -9.41801786e-01 -6.35940254e-01 3.69346261e-01 -5.73998570e-01 1.49333671e-01 -3.57107967e-01 7.33230829e-01 -8.43954682e-02 1.25323665e+00 1.04716167e-01 -2.37197220e-01 4.17128116e-01 -8.71797577e-02 5.88040411e-01 -7.70278752e-01 -3.40680331e-01 2.04498142e-01 -2.09644198e-01 -6.11223102e-01 -7.67436206e-01 -9.26189363e-01 -7.78202593e-01 -4.23968256e-01 -8.30850542e-01 -2.61440456e-01 8.35779130e-01 7.91300178e-01 5.23702919e-01 3.57050329e-01 3.83160889e-01 -1.01027286e+00 1.69048503e-01 -1.02009594e+00 -5.42032897e-01 1.55546859e-01 8.43462229e-01 -9.83146071e-01 -5.77474356e-01 3.59312713e-01]	[14.103230476379395, 0.10896507650613785]
c38e8d34-e11c-41ee-8a38-e2dc92c7b4fe	medical-code-prediction-from-discharge	2106.07932	null	https://arxiv.org/abs/2106.07932v4	https://arxiv.org/pdf/2106.07932v4.pdf	Medical Code Prediction from Discharge Summary: Document to Sequence BERT using Sequence Attention	Clinical notes are unstructured text generated by clinicians during patient encounters. Clinical notes are usually accompanied by a set of metadata codes from the International Classification of Diseases(ICD). ICD code is an important code used in various operations, including insurance, reimbursement, medical diagnosis, etc. Therefore, it is important to classify ICD codes quickly and accurately. However, annotating these codes is costly and time-consuming. So we propose a model based on bidirectional encoder representations from transformers (BERT) using the sequence attention method for automatic ICD code assignment. We evaluate our approach on the medical information mart for intensive care III (MIMIC-III) benchmark dataset. Our model achieved performance of macro-averaged F1: 0.62898 and micro-averaged F1: 0.68555 and is performing better than a performance of the state-of-the-art model using the MIMIC-III dataset. The contribution of this study proposes a method of using BERT that can be applied to documents and a sequence attention method that can capture important sequence in-formation appearing in documents.	['Kyungsun Kim', 'Byeong-Cheol Jo', 'Yeongjoon Park', 'Yongmin Yoo', 'Tak-Sung Heo']	2021-06-15	null	null	null	null	['medical-code-prediction']	['medical']	[ 3.14255834e-01 6.58707172e-02 -8.31868052e-02 -3.41971755e-01 -7.98466027e-01 -3.99056226e-01 1.80666670e-01 8.43506873e-01 -3.42172146e-01 6.53440237e-01 7.94054210e-01 -5.84614515e-01 -2.10346773e-01 -5.58015823e-01 -4.64942545e-01 -3.45881730e-01 -7.48709217e-02 8.56091976e-01 -1.53481647e-01 1.40122682e-01 2.93159008e-01 2.30714232e-01 -1.32601976e+00 1.02557695e+00 7.39338100e-01 1.04402411e+00 1.64182559e-01 8.00211012e-01 -4.69454885e-01 1.58836126e+00 -4.84084338e-01 -5.11445582e-01 -1.88200310e-01 -7.78783321e-01 -1.12748778e+00 -2.55894661e-01 -3.20957333e-01 -1.84871703e-01 -5.13637625e-02 1.07295990e+00 5.36623061e-01 -1.76366284e-01 9.84960079e-01 -7.47493982e-01 -8.01453769e-01 8.11927378e-01 -2.10898459e-01 2.19860733e-01 4.75693107e-01 -3.92824352e-01 8.36867034e-01 -7.21024036e-01 7.63720512e-01 8.98473918e-01 7.63002753e-01 6.21942043e-01 -7.15583146e-01 -4.72229421e-01 -1.47104472e-01 4.69414145e-01 -1.19405329e+00 -1.39644653e-01 2.90391475e-01 -8.81317854e-01 1.26778042e+00 2.91952223e-01 5.18738031e-01 1.01448679e+00 5.89681864e-01 6.35983407e-01 5.51563859e-01 -4.94313866e-01 2.49307200e-01 9.56365168e-02 2.91196197e-01 7.65615046e-01 1.49879143e-01 -1.59149840e-01 -1.93558171e-01 -3.67369741e-01 3.86489362e-01 5.27978361e-01 -3.22274923e-01 3.38745350e-03 -1.42204332e+00 9.35724437e-01 1.57819062e-01 4.48632210e-01 -7.79032171e-01 -8.03941786e-02 9.68047738e-01 8.97360370e-02 1.61875248e-01 2.71858603e-01 -4.30082679e-01 -7.76144028e-01 -7.02155471e-01 6.88219815e-02 7.50374854e-01 1.21868432e+00 -2.02730998e-01 -4.79918808e-01 -5.06528497e-01 8.70760560e-01 2.73226827e-01 3.33518721e-02 9.64785933e-01 -4.44046170e-01 5.12834311e-01 8.18707466e-01 5.63745648e-02 -7.51114964e-01 -4.91529077e-01 -4.29889143e-01 -1.20296550e+00 -4.26761657e-01 -1.42683372e-01 5.13735376e-02 -9.63195801e-01 1.30745196e+00 -5.75128794e-02 3.10961753e-02 2.14179978e-01 4.66591626e-01 9.92920518e-01 7.18300164e-01 1.02492787e-01 -4.02770579e-01 1.68562162e+00 -1.07684982e+00 -1.04724717e+00 3.75306606e-01 9.18899059e-01 -8.20948184e-01 7.92314708e-01 4.09611613e-01 -9.25631344e-01 -5.62005162e-01 -6.12829745e-01 -8.37736800e-02 -1.21023603e-01 2.06709012e-01 2.17056945e-01 3.43405306e-01 -7.48259842e-01 2.84091413e-01 -8.85864258e-01 -5.16554356e-01 3.83620650e-01 1.40220672e-01 -2.19619036e-01 -2.56356969e-02 -9.74521935e-01 7.11444497e-01 3.99201185e-01 -3.97065341e-01 -6.08273625e-01 -7.83281326e-01 -7.08125830e-01 3.56853992e-01 -1.29979983e-01 -8.35694909e-01 1.36305523e+00 -7.21120596e-01 -1.11743915e+00 9.17355359e-01 -1.85160220e-01 -7.49787092e-01 3.71061891e-01 -2.45495498e-01 -6.57058656e-01 2.12682858e-01 1.36033818e-01 4.57393497e-01 6.82072714e-02 -6.48091078e-01 -5.48187375e-01 -1.46631241e-01 -8.74654353e-02 -8.17964450e-02 -4.06171262e-01 2.83831716e-01 -2.25879863e-01 -9.73158717e-01 -2.88382441e-01 -1.02118790e+00 -1.35812461e-01 -1.80224895e-01 -4.42981660e-01 -2.57663429e-01 2.56394506e-01 -9.45549488e-01 1.80840528e+00 -2.12884951e+00 -9.51974392e-02 -8.40863958e-02 1.41545713e-01 3.44645172e-01 2.13645652e-01 8.89190435e-01 -3.27176392e-01 3.30260366e-01 -5.06904900e-01 -4.13451865e-02 -1.85876399e-01 1.15101822e-01 -4.26725417e-01 -4.72146310e-02 2.59300438e-03 7.84185529e-01 -8.72170389e-01 -6.57236934e-01 -6.40179217e-02 4.65643913e-01 -8.83315980e-01 4.70577359e-01 6.75909966e-02 2.73168266e-01 -3.04609358e-01 4.92544562e-01 2.37268806e-01 -8.39970767e-01 2.52532542e-01 -8.51166546e-02 1.14464425e-01 5.33620834e-01 -6.80487156e-01 1.72116590e+00 -3.21179599e-01 4.53420311e-01 -5.61354876e-01 -1.05148101e+00 7.76619196e-01 7.64421344e-01 9.27830756e-01 -1.52597308e-01 2.88896859e-01 9.29143354e-02 2.82960325e-01 -9.75969553e-01 6.29153773e-02 -1.07770488e-01 2.63958070e-02 3.85891795e-01 -2.69883305e-01 2.53650635e-01 2.53233016e-01 2.61302233e-01 1.56159770e+00 -3.29506665e-01 7.65132010e-01 -1.95813239e-01 8.20063412e-01 7.70910010e-02 5.30454934e-01 5.34882605e-01 1.29794627e-02 7.89360166e-01 4.70398992e-01 -8.95685792e-01 -9.13041532e-01 -5.88899255e-01 -4.15432751e-01 3.91220599e-01 -4.34395790e-01 -6.70684218e-01 -6.96834564e-01 -8.67191195e-01 1.78367440e-02 5.73414981e-01 -8.29785287e-01 -2.29424611e-01 -2.13908106e-01 -6.73713446e-01 6.13874197e-01 9.04463172e-01 3.25266331e-01 -1.29690087e+00 -8.80360305e-01 7.88069308e-01 -5.78703046e-01 -1.03175664e+00 -6.63062632e-01 2.58179903e-01 -7.31273413e-01 -1.18165243e+00 -7.32853830e-01 -9.43099558e-01 6.52066946e-01 -4.27980632e-01 1.19466889e+00 1.42559111e-01 -4.07291353e-01 9.14567560e-02 -9.10615861e-01 -6.38628602e-01 -7.56106198e-01 1.23354100e-01 -8.56558681e-02 1.99424159e-02 6.00930572e-01 -2.26457164e-01 -5.39537132e-01 -2.34085232e-01 -1.07472205e+00 3.49593878e-01 5.41921198e-01 1.01259661e+00 4.74195033e-01 -2.80499339e-01 5.77710271e-01 -1.07567227e+00 7.27030933e-01 -5.01930356e-01 -2.43717641e-01 2.88537681e-01 -6.19319499e-01 2.71036029e-01 7.51676083e-01 -1.86571255e-01 -6.68414474e-01 1.08394638e-01 -5.69359422e-01 -2.58177966e-01 -1.30173787e-01 8.31621408e-01 1.60998926e-01 5.02374887e-01 3.09318513e-01 2.37299278e-01 -1.83539808e-01 -7.33211517e-01 -2.54171461e-01 1.37847185e+00 4.15340185e-01 -1.05584808e-01 -1.01939067e-01 1.33324713e-01 -1.87502772e-01 -3.97613943e-01 -7.68415749e-01 -6.41473949e-01 -4.47355866e-01 1.94394127e-01 1.16506350e+00 -6.61376536e-01 -7.80344009e-01 1.59690425e-01 -1.42199564e+00 1.29082695e-01 -4.98572886e-02 6.84347749e-01 -4.48708951e-01 4.84181881e-01 -9.64543402e-01 -4.84014332e-01 -7.86010623e-01 -1.12261569e+00 8.96407187e-01 -4.86770213e-01 -7.51013756e-01 -8.94457102e-01 2.98178464e-01 2.92256266e-01 2.45678112e-01 3.45283478e-01 1.43026781e+00 -8.29416811e-01 -9.41441208e-02 -5.97851761e-02 -2.47697234e-01 2.58122265e-01 6.12922132e-01 -1.45518735e-01 -6.36373818e-01 -8.26363917e-03 -4.00569476e-02 -1.44867077e-01 7.88656950e-01 3.10686439e-01 1.68289769e+00 -5.92000127e-01 -4.90260214e-01 5.00147164e-01 1.31612682e+00 9.22048509e-01 5.92155516e-01 2.32059821e-01 6.61882877e-01 3.34610850e-01 4.02914673e-01 7.86496758e-01 4.58800316e-01 6.50949717e-01 3.02874267e-01 1.28684014e-01 1.63284764e-01 -6.42531514e-02 2.29983982e-02 1.29952455e+00 8.67189988e-02 -4.15202796e-01 -1.37689197e+00 7.71934748e-01 -1.90286100e+00 -1.05167234e+00 -1.39461875e-01 1.87610626e+00 1.19445634e+00 -9.30652022e-02 -2.05681100e-01 3.77210617e-01 5.36892831e-01 -5.27867258e-01 -3.34162593e-01 -7.21610725e-01 4.93137300e-01 6.08080402e-02 2.36384541e-01 2.35512599e-01 -1.00307381e+00 1.39155298e-01 5.88900852e+00 3.36217850e-01 -8.65248263e-01 6.75561652e-02 6.60600126e-01 4.36932929e-02 -1.70066267e-01 -6.07280493e-01 -5.77306271e-01 1.10719836e+00 1.33767426e+00 -2.26383939e-01 2.75779497e-02 8.83933127e-01 -8.62227827e-02 3.72255355e-01 -1.30809724e+00 1.33304048e+00 1.90577149e-01 -1.52316272e+00 3.28536838e-01 1.30981550e-01 7.84717143e-01 -1.31523505e-01 -2.81531990e-01 2.13314340e-01 3.37984741e-01 -8.97276044e-01 4.85589951e-01 4.52259779e-01 9.81345177e-01 -6.33426130e-01 1.33107603e+00 2.79057056e-01 -1.08435440e+00 -2.43041396e-01 -2.13492140e-01 2.25968748e-01 2.47259483e-01 6.94974065e-01 -1.06855500e+00 5.20061910e-01 8.23254764e-01 1.03467345e+00 -1.68225482e-01 1.06567371e+00 1.84296846e-01 5.96283376e-01 2.02894986e-01 -2.18957588e-02 2.53988355e-01 1.61958799e-01 2.49984600e-02 1.49285316e+00 7.03779876e-01 3.54409218e-01 1.72895700e-01 3.37579161e-01 -2.49767527e-01 3.38881016e-01 -6.31152153e-01 -3.11871469e-01 2.69038141e-01 7.43351936e-01 -7.39049017e-01 -7.08170414e-01 -4.13905382e-01 1.09896576e+00 -1.67561788e-02 -8.91487151e-02 -1.01834691e+00 -5.57708621e-01 5.53703964e-01 6.01848923e-02 5.05851865e-01 4.24447656e-01 -1.13053724e-01 -1.05267119e+00 -5.12705110e-02 -1.05937624e+00 6.50672376e-01 -7.82526374e-01 -1.13432133e+00 1.11217368e+00 -2.71363050e-01 -1.62281299e+00 -5.94278753e-01 -5.22963226e-01 3.94267291e-02 5.05982518e-01 -1.22199380e+00 -7.34095871e-01 -3.83676827e-01 4.59204018e-01 6.61239147e-01 -2.96690077e-01 1.24215484e+00 7.85198033e-01 -2.38337055e-01 5.11699736e-01 4.84327376e-01 4.75099206e-01 6.02061987e-01 -1.17298687e+00 3.23605478e-01 4.34842199e-01 1.08433604e-01 7.07309306e-01 2.16226354e-01 -6.19398952e-01 -8.36072087e-01 -1.46024060e+00 1.61304843e+00 -3.86013061e-01 1.88216314e-01 -1.25334207e-02 -9.34536755e-01 7.63292909e-01 2.79863447e-01 -1.11203976e-01 1.08007801e+00 -4.40138161e-01 -7.50181079e-02 -3.39936912e-02 -1.11698556e+00 1.72078744e-01 1.07697165e+00 -4.47123587e-01 -7.66223848e-01 4.32100028e-01 8.57291222e-01 -4.66454148e-01 -1.09284019e+00 3.22577477e-01 5.32479584e-01 -7.28610933e-01 8.49168599e-01 -7.29125381e-01 9.17876005e-01 -1.11278340e-01 -1.72589701e-02 -1.09975731e+00 -5.87058604e-01 -2.78322369e-01 -9.51170996e-02 1.17417395e+00 3.01538467e-01 -4.71422076e-01 3.50757897e-01 2.95724601e-01 -2.42665380e-01 -8.46338451e-01 -7.50874341e-01 -6.56874478e-01 -2.43469447e-01 -2.05370620e-01 8.87278080e-01 1.16752636e+00 2.97550172e-01 2.28609100e-01 -3.27763736e-01 -2.30664030e-01 2.04670846e-01 1.26653463e-01 1.25597820e-01 -1.37429082e+00 -2.26815373e-01 -3.67503881e-01 -4.61492449e-01 -5.74360609e-01 -2.12538168e-01 -1.03785777e+00 -2.52531525e-02 -1.92583466e+00 4.68759835e-01 -3.55833143e-01 -6.54535770e-01 6.35428786e-01 -1.22003011e-01 -1.19164675e-01 8.46177414e-02 2.72282809e-01 -4.23922718e-01 2.75999248e-01 7.15205729e-01 -3.45986217e-01 -2.14661378e-02 -2.16417998e-01 -6.49698913e-01 4.87950176e-01 8.62613678e-01 -8.85202885e-01 -4.80490595e-01 -4.05297428e-01 3.41826290e-01 3.29228848e-01 -1.06768690e-01 -1.08954573e+00 1.77438110e-01 9.00897831e-02 1.77542433e-01 -7.60659337e-01 -1.36969328e-01 -1.11950171e+00 3.27842295e-01 1.03540230e+00 -7.88190067e-01 6.26569867e-01 1.92306533e-01 4.45657820e-01 -4.54911351e-01 -3.55462223e-01 4.98654246e-01 -3.11446249e-01 -3.78792167e-01 9.92216617e-02 -7.88149297e-01 1.78633437e-01 8.85401547e-01 6.83552474e-02 2.55212206e-02 -3.04808766e-01 -4.58663791e-01 -9.11440775e-02 1.40479028e-01 5.95944285e-01 7.67634630e-01 -1.45185292e+00 -7.81984270e-01 4.10614789e-01 5.83455861e-01 -3.46802384e-01 2.46651359e-02 8.13347876e-01 -9.69377697e-01 9.24584031e-01 -2.55428106e-01 -4.74883884e-01 -1.59753585e+00 8.63904953e-01 6.28242046e-02 -7.43025541e-01 -7.37305760e-01 7.10298598e-01 1.30459934e-01 -2.78892368e-01 4.00127828e-01 -9.95380640e-01 -5.55332422e-01 2.15190221e-02 8.84785116e-01 1.53373942e-01 4.11499202e-01 -4.15759206e-01 -6.85649633e-01 6.11544311e-01 -1.58292010e-01 1.14717118e-01 1.40027094e+00 9.38953161e-02 -2.65727460e-01 2.29656622e-01 1.41127527e+00 -2.16032699e-01 -3.97717476e-01 3.41942534e-02 1.85299352e-01 -1.30434111e-01 -1.76845610e-01 -1.05993974e+00 -8.69609833e-01 1.09965026e+00 7.97706664e-01 4.61693257e-02 1.18609047e+00 -1.52785361e-01 9.27218437e-01 3.19293737e-01 2.67836630e-01 -6.70976877e-01 -5.52364402e-02 5.94172537e-01 9.84422982e-01 -1.03208411e+00 -4.32651997e-01 -1.80487156e-01 -8.71402740e-01 1.02948439e+00 1.45660713e-01 4.17776346e-01 7.93359816e-01 4.29221809e-01 2.24566370e-01 -1.48572057e-01 -9.87015247e-01 1.72817603e-01 2.97743469e-01 3.46258730e-01 9.95526254e-01 -6.29977137e-02 -5.33561110e-01 7.59335399e-01 -9.64731798e-02 5.55609643e-01 4.63115633e-01 1.16229820e+00 -6.84115663e-02 -1.22492039e+00 -3.20184469e-01 6.69885159e-01 -1.09258819e+00 -5.12972474e-01 -2.16501594e-01 2.37940922e-01 1.98843569e-01 8.98166895e-01 1.28250569e-01 -3.93724591e-01 1.84184566e-01 3.47686142e-01 5.24195135e-02 -7.70698726e-01 -9.05711830e-01 -3.06993008e-01 6.54684603e-02 -2.56073862e-01 -4.44586426e-01 -4.65666085e-01 -1.48909450e+00 -1.22695476e-01 1.87416375e-01 5.01483917e-01 2.48237222e-01 5.13768017e-01 8.54727864e-01 1.07449567e+00 2.06362531e-01 2.97789425e-02 -3.07548583e-01 -1.14516771e+00 -3.14118564e-01 7.41667569e-01 6.09677792e-01 -3.82010251e-01 1.24293499e-01 7.99967647e-01]	[8.000174522399902, 6.841665744781494]
71347f7a-bc9e-486c-8399-047891f4e6f7	knowledge-distillation-from-ensemble-of	2108.09183	null	https://arxiv.org/abs/2108.09183v2	https://arxiv.org/pdf/2108.09183v2.pdf	Boosting of Head Pose Estimation by Knowledge Distillation	We propose a response-based method of knowledge distillation (KD) for the head pose estimation problem. A student model trained by the proposed KD achieves results better than a teacher model, which is atypical for the response-based method. Our method consists of two stages. In the first stage, we trained the base neural network (NN), which has one regression head and four regression via classification (RvC) heads. We build the convolutional ensemble over the base NN using offsets of face bounding boxes over a regular grid. In the second stage, we perform KD from the convolutional ensemble into the final NN with one RvC head. The KD improves the results by an average of 7.7\% compared to base NN. This feature makes it possible to use KD as a booster and effectively train deeper NNs. NNs trained by our KD method partially improved the state-of-the-art results. KD-ResNet152 has the best results, and KD-ResNet18 has a better result on the AFLW2000 dataset than any previous method.We have made publicly available trained NNs and face bounding boxes for the 300W-LP, AFLW, AFLW2000, and BIWI datasets.Our method potentially can be effective for other regression problems.	['Victor Samun', 'Andrey Sheka']	2021-08-20	null	null	null	null	['head-pose-estimation']	['computer-vision']	[-4.45908725e-01 3.91437590e-01 -1.24176286e-01 -7.18727827e-01 -7.19917059e-01 -1.91654816e-01 4.59438980e-01 -3.69553030e-01 -6.96249783e-01 8.20001781e-01 3.94437462e-01 -1.22211330e-01 2.16368988e-01 -9.70808804e-01 -9.31254089e-01 -6.83211923e-01 8.39079469e-02 7.83472002e-01 6.42816126e-01 -3.97361130e-01 -1.95880026e-01 7.00460792e-01 -1.57262659e+00 2.37630904e-01 6.58164978e-01 1.06447482e+00 -7.65897846e-03 3.68283570e-01 -9.35121626e-02 7.55078614e-01 -5.28565466e-01 -5.54840088e-01 1.68232977e-01 3.34463045e-02 -8.94636810e-01 -5.02611101e-01 8.80906940e-01 -5.49749136e-01 -4.04380113e-01 6.47694707e-01 1.09298539e+00 3.68195057e-01 6.51616096e-01 -1.15869045e+00 -6.16652787e-01 6.12509072e-01 -6.93449974e-01 -4.47932817e-02 2.47923359e-01 -1.54117718e-01 5.67293227e-01 -1.29692280e+00 6.06960535e-01 1.47777462e+00 8.70400786e-01 1.11581552e+00 -9.59561348e-01 -1.34776318e+00 1.51064783e-01 3.82493764e-01 -1.61584008e+00 -6.01117134e-01 6.34521067e-01 -2.08448827e-01 9.51024234e-01 -1.12359419e-01 5.40184498e-01 1.06346381e+00 -4.02208388e-01 9.35715735e-01 1.02487564e+00 -4.02137518e-01 5.63804954e-02 7.42326602e-02 4.00448471e-01 8.81467760e-01 1.09927334e-01 1.35169938e-01 -6.50713027e-01 -5.27521595e-02 4.66111153e-01 -2.75090665e-01 -2.74058729e-01 5.85807823e-02 -4.35782403e-01 1.06836522e+00 8.80072534e-01 1.15697458e-01 -1.15189970e-01 6.88775554e-02 2.51211256e-01 1.73528977e-02 7.80441284e-01 5.68703897e-02 -6.20475888e-01 2.66474396e-01 -1.04055977e+00 5.17684996e-01 9.72128808e-01 8.72874737e-01 6.39473975e-01 1.24544248e-01 -2.41791710e-01 1.23541760e+00 4.33306366e-01 6.26270652e-01 5.72203636e-01 -6.41424716e-01 2.40128636e-01 4.69634652e-01 -1.88701883e-01 -6.33407176e-01 -8.06586921e-01 -2.84146786e-01 -8.17545354e-01 2.87729383e-01 5.29075384e-01 -3.25032771e-01 -1.34139323e+00 1.74243450e+00 5.69350362e-01 3.60066205e-01 -1.05310880e-01 7.02567101e-01 1.72833121e+00 4.73994941e-01 1.52700365e-01 3.35120521e-02 1.33849311e+00 -1.19810593e+00 -6.62309825e-01 -1.63839057e-01 8.13075185e-01 -6.20478630e-01 7.49278307e-01 3.02990854e-01 -9.82589960e-01 -5.87927043e-01 -8.90608490e-01 -1.92092210e-01 -5.62164545e-01 2.00336948e-01 5.98061383e-01 7.14870214e-01 -1.59289777e+00 4.48837966e-01 -4.80733871e-01 -2.49948785e-01 6.12152278e-01 6.84001505e-01 -5.29971063e-01 -1.65062189e-01 -1.25146782e+00 1.22236836e+00 2.50775248e-01 2.18513370e-01 -7.18106627e-01 -7.16149211e-01 -9.60689068e-01 -1.15024231e-01 2.74643868e-01 -3.36080551e-01 1.36054564e+00 -5.25448024e-01 -1.91804469e+00 8.72056007e-01 -2.70561308e-01 -3.81026804e-01 4.79072452e-01 -3.18880290e-01 -2.66897649e-01 -2.06561804e-01 -1.71749994e-01 1.15104640e+00 7.77405798e-01 -1.15803552e+00 -4.88028198e-01 -4.87759203e-01 -1.22726761e-01 -5.36331125e-02 -1.67062581e-01 1.88374221e-02 -6.99381709e-01 -4.40153927e-01 -4.46940549e-02 -9.94398654e-01 1.59142334e-02 -2.13941813e-01 -2.36217871e-01 -8.65574479e-01 8.06010962e-01 -6.89011335e-01 1.31310844e+00 -1.93144262e+00 -1.19090915e-01 2.62308002e-01 1.99753612e-01 7.73824155e-01 -4.55192089e-01 -1.74843356e-01 -3.70122761e-01 -8.15860778e-02 1.52930856e-01 -7.84285426e-01 -1.94454074e-01 2.82629997e-01 -2.68761307e-01 4.45336819e-01 1.60509810e-01 9.30302441e-01 -4.54363316e-01 -4.42655683e-01 -1.29939288e-01 6.60561562e-01 -9.04646099e-01 3.59424233e-01 -5.89116141e-02 3.42311054e-01 -1.03878140e-01 6.08680964e-01 1.05692220e+00 6.90740198e-02 -1.24580756e-01 -2.41171554e-01 4.49500456e-02 1.85052633e-01 -1.17097700e+00 1.28130245e+00 -5.01225591e-01 4.99737352e-01 -1.41868725e-01 -9.17410970e-01 1.14076531e+00 2.59217680e-01 3.93616483e-02 -7.20165253e-01 8.74874964e-02 2.25142971e-01 -5.83105758e-02 -3.04595351e-01 3.38339090e-01 -2.78232619e-02 1.77079961e-01 2.15003833e-01 5.35767794e-01 -1.60754845e-01 -1.66108012e-02 -1.10196069e-01 7.20700145e-01 2.98220098e-01 1.42626479e-01 -2.02749088e-01 6.68077886e-01 -4.01573867e-01 7.11182296e-01 6.62592471e-01 -9.40290168e-02 5.78816116e-01 2.68431455e-01 -7.52616763e-01 -5.92044830e-01 -8.18998575e-01 -3.68685156e-01 1.64069617e+00 -3.82188976e-01 -3.48056495e-01 -9.25382733e-01 -9.63385880e-01 1.65501922e-01 6.20059967e-01 -9.37437296e-01 -2.03894868e-01 -9.90779757e-01 -7.00662196e-01 8.32701385e-01 1.01667643e+00 8.81399274e-01 -1.19241869e+00 1.47976592e-01 1.05802894e-01 -2.23391727e-01 -1.06950545e+00 -3.59165132e-01 2.81506062e-01 -6.40238345e-01 -9.99847293e-01 -9.36356664e-01 -9.36490238e-01 7.90918231e-01 -2.24659801e-01 1.04882455e+00 2.66416430e-01 -5.43836243e-02 1.73718836e-02 -2.89579868e-01 -7.62101591e-01 2.15149187e-02 3.08956295e-01 3.09799582e-01 -1.81634679e-01 6.10156775e-01 -2.71424383e-01 -5.17281115e-01 4.51468676e-01 -3.73896569e-01 -2.45777503e-01 3.69785607e-01 8.16580176e-01 3.64831477e-01 -4.33488488e-01 7.34884143e-01 -9.79891896e-01 3.07063311e-01 -2.21424416e-01 -6.64508224e-01 4.61896546e-02 -5.37814856e-01 2.36838698e-01 4.64781672e-01 -5.93425989e-01 -1.27123940e+00 2.29483783e-01 -8.12399209e-01 -1.72334343e-01 -2.29808837e-01 1.09971903e-01 -1.08640336e-01 -3.44023257e-01 8.07745934e-01 -9.42140967e-02 -1.23482384e-01 -7.87544250e-01 3.90548140e-01 6.86654687e-01 4.72828478e-01 -5.90083599e-01 8.43322933e-01 2.08144858e-01 -1.49912253e-01 -8.88160944e-01 -1.13519084e+00 -3.11337918e-01 -7.14100480e-01 3.17850411e-02 8.94847870e-01 -9.51061845e-01 -1.06758440e+00 6.71828985e-01 -1.27202487e+00 -6.98186398e-01 -1.97173789e-01 4.03329670e-01 -8.91070589e-02 -2.03309000e-01 -6.25916481e-01 -5.70819616e-01 -4.77066278e-01 -9.54861701e-01 9.78676081e-01 4.58590239e-01 -1.97686538e-01 -8.66671264e-01 1.54245853e-01 4.36044306e-01 6.28253281e-01 -1.73020676e-01 6.57174110e-01 -1.02010667e+00 -1.86973050e-01 -1.57604918e-01 -2.80320495e-01 3.20591956e-01 -3.60271782e-01 -1.64439112e-01 -1.43786228e+00 -2.84106523e-01 -2.94451237e-01 -6.14829957e-01 1.31977141e+00 4.10833359e-01 1.37432575e+00 -1.45501345e-01 -3.90506297e-01 8.22559118e-01 1.02139103e+00 1.28426388e-01 7.78127015e-01 2.85813361e-01 6.89631224e-01 4.61115181e-01 2.76609629e-01 6.44565895e-02 6.25564456e-01 7.69275904e-01 2.66766787e-01 -3.25823128e-01 -4.06818748e-01 -2.18128324e-01 3.85310501e-01 6.74914062e-01 -5.94016612e-01 5.19317612e-02 -9.27592099e-01 3.43902379e-01 -1.63206542e+00 -7.62399197e-01 2.29265373e-02 2.05497146e+00 1.05492210e+00 3.02636120e-02 2.64498979e-01 -1.79158282e-02 4.75203544e-01 1.11234188e-01 -2.77318776e-01 -3.72903377e-01 1.70506626e-01 7.20550835e-01 4.78156745e-01 5.81291080e-01 -1.05387342e+00 1.29956615e+00 6.50010157e+00 8.77130151e-01 -1.21444869e+00 3.46960604e-01 5.58128357e-01 -1.13027945e-01 2.21949816e-01 -5.20236850e-01 -1.75808418e+00 1.67802960e-01 9.71288741e-01 4.35886174e-01 3.66761476e-01 1.07176852e+00 -1.73807010e-01 -2.27677748e-01 -1.09449697e+00 9.17643249e-01 4.11371112e-01 -1.14855766e+00 -1.35784104e-01 -8.58253762e-02 6.77232563e-01 1.90225154e-01 3.73201654e-03 1.08066857e+00 5.96046925e-01 -1.31828904e+00 4.40392345e-01 3.37373644e-01 8.47684562e-01 -9.81892645e-01 9.79933441e-01 3.35475475e-01 -1.28933096e+00 1.00359805e-02 -5.48158109e-01 6.98388591e-02 -2.57367671e-01 3.68395418e-01 -1.02842331e+00 1.65138677e-01 9.94918346e-01 2.77728409e-01 -6.99720502e-01 8.70219171e-01 -6.31856263e-01 8.52776587e-01 -5.31246305e-01 -1.16626382e-01 -2.94317361e-02 2.49507099e-01 3.35558392e-02 1.33570421e+00 -1.29279777e-01 3.19701642e-01 4.94647399e-03 4.61356759e-01 -4.81116772e-01 1.68277338e-01 -4.88787502e-01 5.88240564e-01 4.51680332e-01 1.36208379e+00 -4.23656762e-01 -3.55219990e-01 -4.69450563e-01 6.38435364e-01 7.99652159e-01 3.26248556e-01 -8.29849303e-01 -3.58162969e-01 5.29884577e-01 9.63361040e-02 4.62815315e-01 6.75578564e-02 -2.35151313e-02 -8.85362446e-01 -2.94564158e-01 -7.56539345e-01 4.14635092e-01 -6.45080328e-01 -1.22316337e+00 8.94248486e-01 1.80487275e-01 -6.73145711e-01 -2.80868649e-01 -8.13896060e-01 -6.71998680e-01 9.65129137e-01 -1.64650154e+00 -1.30093789e+00 -4.38557506e-01 7.37273157e-01 3.37255567e-01 -2.27122188e-01 1.13433707e+00 4.66086090e-01 -7.05674946e-01 1.00673544e+00 -2.06701562e-01 5.36676347e-01 1.02570319e+00 -1.18227816e+00 3.36965859e-01 3.41618389e-01 -1.74956083e-01 5.40900111e-01 2.61960059e-01 -6.18093967e-01 -9.23927784e-01 -1.31165564e+00 1.08722222e+00 -4.32696998e-01 2.17131615e-01 -4.19799298e-01 -1.06871605e+00 8.09666038e-01 2.81379931e-02 4.27337706e-01 6.61129832e-01 4.34045166e-01 -5.26881874e-01 -2.89024830e-01 -1.33416128e+00 3.69654328e-01 1.00459492e+00 -3.86820942e-01 -8.07884812e-01 2.65552312e-01 6.78118289e-01 -7.15763509e-01 -6.54302895e-01 6.07324064e-01 7.16784656e-01 -5.42327464e-01 1.08901966e+00 -5.96061170e-01 2.63441205e-01 -2.63991449e-02 -4.10341434e-02 -1.61740541e+00 -3.30621302e-01 -1.90203607e-01 -2.06937328e-01 1.26105464e+00 4.50785100e-01 -7.27109015e-01 9.86065149e-01 6.77779019e-01 6.26630858e-02 -9.95733321e-01 -9.33254600e-01 -6.88917816e-01 4.49200839e-01 -4.13533121e-01 6.78513527e-01 8.67351055e-01 -4.38739449e-01 4.23190594e-01 -2.25990057e-01 -2.31800731e-02 5.09651661e-01 -3.02301198e-01 8.83461475e-01 -1.55066705e+00 9.51942205e-02 -2.96449035e-01 -9.38896686e-02 -9.93665636e-01 6.26014233e-01 -1.12658346e+00 2.87960470e-02 -1.42949533e+00 1.52164355e-01 -3.71976644e-01 -9.57108103e-03 1.13187754e+00 -2.20970541e-01 6.23081684e-01 1.32312194e-01 -2.85417318e-01 -2.49002948e-01 4.93076086e-01 1.35657310e+00 -3.89725417e-02 -3.44031096e-01 1.06351577e-01 -4.71025497e-01 1.19886708e+00 7.23614037e-01 -5.06853998e-01 -8.21369365e-02 -1.90664351e-01 -1.15719862e-01 -2.70438164e-01 1.02156028e-01 -8.24516416e-01 5.10493636e-01 2.60333389e-01 9.23027873e-01 -8.85561466e-01 4.42085534e-01 -6.43475175e-01 -3.65851700e-01 3.79237175e-01 -6.63933232e-02 -2.90493727e-01 4.56659943e-01 6.68973178e-02 1.99597590e-02 -2.96421558e-01 1.10139084e+00 1.45922974e-01 -7.07812309e-01 4.31582719e-01 -9.66568962e-02 2.25136697e-01 6.98939562e-01 -5.37129957e-03 -4.91101921e-01 -3.86222303e-01 -9.75868225e-01 4.27753687e-01 -2.67107010e-01 4.38370764e-01 7.26804912e-01 -1.40453565e+00 -8.34522605e-01 6.24090493e-01 -5.81532866e-02 3.15665245e-01 -5.18968003e-03 8.23041499e-01 -3.24543238e-01 3.54715198e-01 -1.34199694e-01 -4.25436676e-01 -1.49561191e+00 1.79910898e-01 4.19968128e-01 -3.18156153e-01 -4.28925663e-01 1.36451817e+00 1.64786726e-01 -9.65732694e-01 6.81411624e-01 -1.57535121e-01 -6.52013600e-01 3.47710252e-01 7.98562586e-01 4.68698204e-01 2.82928973e-01 -7.91333258e-01 -6.02819145e-01 7.83142686e-01 -3.70945454e-01 5.70186786e-02 1.67447555e+00 4.37894642e-01 -4.09647217e-03 -1.57462254e-01 1.30073643e+00 -6.88018836e-03 -9.76759255e-01 -3.73137414e-01 -1.64595321e-01 -7.91223198e-02 1.41141772e-01 -8.99194777e-01 -1.34123957e+00 8.21644843e-01 6.91763163e-01 -5.01509309e-01 1.04557943e+00 1.55419141e-01 5.84396243e-01 5.68824410e-01 2.84035325e-01 -1.09821439e+00 1.91973612e-01 1.05262411e+00 1.08317685e+00 -1.28557336e+00 -1.14242882e-02 -2.97916502e-01 -4.55660403e-01 1.20799696e+00 1.14616394e+00 -1.46550208e-01 9.56067860e-01 5.12638330e-01 1.25514179e-01 -1.03100829e-01 -5.82118034e-01 -4.66067225e-01 6.87941790e-01 5.61135709e-01 6.32759511e-01 2.01263893e-02 -1.43640727e-01 9.19831395e-01 -3.93262297e-01 1.56113133e-01 3.01773474e-02 6.45023286e-01 -4.94279861e-01 -1.17032242e+00 -4.14294660e-01 3.67074043e-01 -4.28974777e-01 -6.49429038e-02 -3.37119371e-01 1.08511710e+00 3.89589697e-01 7.05664575e-01 4.53474186e-02 -5.77905118e-01 6.84375286e-01 4.37184393e-01 6.07539833e-01 -6.32843852e-01 -6.85614765e-01 -5.75491488e-02 1.41009703e-01 -7.88620532e-01 -1.78050622e-01 -2.55590141e-01 -1.42118263e+00 -4.55623388e-01 -5.83833516e-01 -1.82533078e-02 6.38840854e-01 9.79629040e-01 -1.08750194e-01 4.51665699e-01 4.28606600e-01 -1.11060584e+00 -4.94948715e-01 -1.37255597e+00 -4.64641422e-01 1.06043600e-01 2.68043876e-01 -1.01990831e+00 -3.87843609e-01 -1.73760280e-01]	[13.630918502807617, 0.34267160296440125]
88e044a7-bc6b-4fa6-938e-a6fb8cabf56d	building-accurate-low-latency-asr-for	2305.18596	null	https://arxiv.org/abs/2305.18596v1	https://arxiv.org/pdf/2305.18596v1.pdf	Building Accurate Low Latency ASR for Streaming Voice Search	Automatic Speech Recognition (ASR) plays a crucial role in voice-based applications. For applications requiring real-time feedback like Voice Search, streaming capability becomes vital. While LSTM/RNN and CTC based ASR systems are commonly employed for low-latency streaming applications, they often exhibit lower accuracy compared to state-of-the-art models due to a lack of future audio frames. In this work, we focus on developing accurate LSTM, attention, and CTC based streaming ASR models for large-scale Hinglish (a blend of Hindi and English) Voice Search. We investigate various modifications in vanilla LSTM training which enhance the system's accuracy while preserving its streaming capabilities. We also address the critical requirement of end-of-speech (EOS) detection in streaming applications. We present a simple training and inference strategy for end-to-end CTC models that enables joint ASR and EOS detection. The evaluation of our model on Flipkart's Voice Search, which handles substantial traffic of approximately 6 million queries per day, demonstrates significant performance gains over the vanilla LSTM-CTC model. Our model achieves a word error rate (WER) of 3.69% without EOS and 4.78% with EOS while also reducing the search latency by approximately ~1300 ms (equivalent to 46.64% reduction) when compared to an independent voice activity detection (VAD) model.	['Nikesh Garera', 'Abhinav Goyal']	2023-05-29	null	null	null	null	['action-detection', 'activity-detection', 'automatic-speech-recognition']	['computer-vision', 'computer-vision', 'speech']	[ 1.46071807e-01 -3.90408307e-01 -8.57475176e-02 -1.63499296e-01 -1.68515885e+00 -4.38474625e-01 3.84567380e-01 -7.34639540e-02 -7.13438988e-01 2.28313193e-01 3.99437785e-01 -8.63520205e-01 2.64098287e-01 -5.87011203e-02 -3.74913961e-01 -4.13320780e-01 3.29203159e-02 3.32090229e-01 6.01004958e-01 -2.58726090e-01 4.16597212e-03 5.65236926e-01 -1.45572555e+00 3.95042300e-01 5.39263189e-01 1.21891785e+00 7.05271602e-01 1.46762252e+00 -1.60212874e-01 8.78871202e-01 -1.13663375e+00 -1.08603463e-02 -2.05066592e-01 -2.01542303e-01 -5.86945951e-01 -2.50257969e-01 2.56333977e-01 -5.97109854e-01 -9.43662465e-01 5.63992620e-01 1.18489838e+00 4.11643893e-01 1.96866751e-01 -8.91120195e-01 -1.41114697e-01 6.10219359e-01 -2.45954171e-01 7.90935636e-01 3.39583308e-01 1.36915624e-01 1.12938178e+00 -9.23146665e-01 -4.18231562e-02 1.15824103e+00 3.45440000e-01 8.46265316e-01 -7.20532715e-01 -6.47394896e-01 -5.06905355e-02 5.42549789e-01 -1.36646771e+00 -1.29035366e+00 4.68070209e-01 2.80492574e-01 1.69663703e+00 7.18792796e-01 3.01850736e-01 1.18356574e+00 -1.32684499e-01 1.14188325e+00 3.85873646e-01 -3.29519629e-01 1.89326569e-01 -7.93911442e-02 -1.91596195e-01 2.48096243e-01 -6.79007590e-01 -1.99477356e-02 -1.05816591e+00 -2.18949199e-01 3.85912150e-01 -3.57096642e-01 -3.75449479e-01 7.11829901e-01 -1.14556968e+00 5.86272717e-01 -4.57225367e-02 2.68663645e-01 -4.50313807e-01 5.49250126e-01 7.25470245e-01 3.86816502e-01 4.99852449e-01 -3.17480825e-02 -5.46332657e-01 -9.37886357e-01 -1.36816072e+00 -3.00012738e-01 8.33283842e-01 1.15496922e+00 -1.55657977e-01 8.55160296e-01 -3.96314114e-01 1.45025611e+00 4.23863083e-01 7.94576406e-01 8.14059675e-01 -8.31641078e-01 6.94867611e-01 -4.03751820e-01 -1.59565464e-01 -2.94144183e-01 8.35332423e-02 -5.92773736e-01 -5.75452805e-01 -4.36773032e-01 -1.04541801e-01 -8.10822025e-02 -1.06713283e+00 1.30213749e+00 6.51111677e-02 2.93021381e-01 1.06716998e-01 8.14617693e-01 7.38198638e-01 1.11170518e+00 -2.78437257e-01 -4.25950259e-01 1.32049119e+00 -1.09282291e+00 -9.94996488e-01 -1.95745900e-01 5.37391126e-01 -9.52796340e-01 1.10875762e+00 3.17310005e-01 -1.26804507e+00 -3.27879965e-01 -7.57285714e-01 -7.64940530e-02 2.22247597e-02 1.39694601e-01 -6.16688207e-02 7.93950558e-01 -1.45217073e+00 1.34915322e-01 -9.38438177e-01 -3.30213159e-01 -7.95329213e-02 4.46691245e-01 1.96675345e-01 2.16128469e-01 -1.30274439e+00 5.41144490e-01 -3.58981609e-01 2.57443249e-01 -1.09471917e+00 -6.00180089e-01 -6.84730887e-01 3.20525527e-01 4.48969096e-01 2.42805984e-02 2.04704332e+00 -2.47930080e-01 -1.90381742e+00 3.84962201e-01 -7.93217480e-01 -7.00163126e-01 4.66240227e-01 -2.64086574e-01 -9.11835551e-01 3.00111175e-01 -2.43287385e-01 3.25277150e-01 9.69539046e-01 -7.00403988e-01 -9.02668834e-01 8.15913305e-02 -4.12554383e-01 2.19471380e-01 -4.61296469e-01 7.80172884e-01 -7.74025381e-01 -7.38145113e-01 -1.45538390e-01 -7.03481376e-01 8.20268467e-02 -2.95457095e-01 -2.82775342e-01 -2.85235316e-01 1.22530365e+00 -9.85185146e-01 1.72272921e+00 -2.16609454e+00 -4.41339701e-01 -1.35483712e-01 -8.10623616e-02 8.18835318e-01 -2.32321143e-01 5.53396940e-01 2.98095405e-01 2.51190186e-01 8.21419954e-02 -5.48116267e-01 6.02791384e-02 1.57078490e-01 -5.95781147e-01 2.07302868e-01 2.25386564e-02 7.21221328e-01 -6.42036378e-01 -4.80945498e-01 3.63894790e-01 6.08989835e-01 -3.97237420e-01 5.51148832e-01 2.26988681e-02 7.12606460e-02 -1.64560407e-01 8.95528376e-01 5.12027591e-02 1.65813506e-01 -1.24998190e-01 1.47043124e-01 -2.20602527e-01 9.86781597e-01 -7.93911934e-01 1.28527606e+00 -9.32180405e-01 1.14721370e+00 5.33524156e-01 -6.41470015e-01 7.47210205e-01 9.53787804e-01 1.24179803e-01 -1.06407785e+00 3.96614149e-03 4.78224188e-01 1.22230342e-02 -4.50038433e-01 8.41042519e-01 1.85821518e-01 1.50282696e-01 3.04046452e-01 -1.14330754e-01 4.12254967e-02 -2.85326660e-01 1.61875144e-01 1.40632570e+00 -5.09491920e-01 -2.20132604e-01 1.72395602e-01 5.34222901e-01 -4.20965075e-01 2.91664034e-01 9.20690894e-01 -3.87097508e-01 6.97727144e-01 -1.89398378e-02 3.13577026e-01 -8.74517143e-01 -8.80620480e-01 2.70177498e-02 1.46016204e+00 -3.53904337e-01 -4.17998731e-01 -6.56108201e-01 -2.72170484e-01 -3.18321019e-01 9.06601310e-01 1.74758673e-01 1.36282267e-02 -9.80559766e-01 -2.57231355e-01 1.14695919e+00 4.26573902e-01 3.19427580e-01 -1.18964434e+00 -5.49689949e-01 5.41170120e-01 -4.56194371e-01 -1.61607146e+00 -1.05250931e+00 2.66118646e-01 -6.48541749e-01 -1.83646977e-01 -1.04306710e+00 -7.02882230e-01 -2.47673094e-01 6.29230797e-01 8.35811496e-01 -1.05904201e-02 -1.57277957e-01 5.41276574e-01 -5.65923929e-01 -1.18896604e-01 -7.07277596e-01 2.81275123e-01 1.28458247e-01 -1.41073480e-01 1.02748506e-01 -3.66146654e-01 -5.99205315e-01 5.33834875e-01 -7.33719170e-01 -6.10901296e-01 3.79528880e-01 8.70187640e-01 3.05549383e-01 -3.12589079e-01 8.34618032e-01 -2.50568658e-01 8.35716069e-01 -3.54661673e-01 -3.34669590e-01 2.62601972e-01 -5.24002969e-01 -1.82418168e-01 5.91448307e-01 -7.29560792e-01 -9.02502120e-01 -3.17484796e-01 -8.77323091e-01 -7.52184153e-01 4.56902012e-02 4.17430222e-01 2.23155208e-02 1.22077800e-01 2.02549949e-01 7.29585409e-01 -2.19786987e-01 -6.30583167e-01 9.24353302e-02 1.69974804e+00 5.09912550e-01 -1.81557789e-01 3.26390296e-01 -6.61821738e-02 -7.41581261e-01 -1.63778174e+00 -1.66015089e-01 -9.57888246e-01 -5.82965724e-02 -2.13132933e-01 4.18615609e-01 -9.78533447e-01 -8.71406138e-01 4.91253316e-01 -1.16499901e+00 -5.17526925e-01 -1.04762189e-01 5.89622498e-01 -4.70779419e-01 5.65119088e-01 -9.55103695e-01 -1.48716128e+00 -8.39371085e-01 -1.15659356e+00 1.31411862e+00 -1.66922182e-01 -1.89045355e-01 -5.59120297e-01 -3.44511062e-01 5.18763065e-01 9.32090342e-01 -7.89207041e-01 5.64834416e-01 -9.46897864e-01 -3.79308015e-01 -1.98074281e-01 -6.21892251e-02 3.98824066e-01 -6.66284636e-02 -8.76807496e-02 -1.34082246e+00 -3.99354905e-01 7.35605359e-02 -1.16810091e-01 7.62626588e-01 3.59740704e-01 1.10657620e+00 -4.63520914e-01 -6.70695081e-02 3.37683767e-01 8.81766915e-01 7.77659774e-01 6.08660281e-01 -1.10546783e-01 5.01306832e-01 2.32616067e-01 6.01691067e-01 4.97686684e-01 1.64701566e-01 9.81630325e-01 2.01056764e-01 8.88220668e-02 -4.93006647e-01 -2.58848041e-01 8.60515356e-01 1.74561155e+00 4.86001462e-01 -9.32662487e-01 -8.88730824e-01 9.18019950e-01 -1.56677794e+00 -8.39394510e-01 -6.36567846e-02 2.32679987e+00 8.09887528e-01 1.72830939e-01 2.02154800e-01 3.64176363e-01 8.52033973e-01 4.59469289e-01 -6.02184892e-01 -8.08363259e-01 1.25958294e-01 4.83947754e-01 5.52217007e-01 7.62556136e-01 -5.16626954e-01 1.07652974e+00 6.18007326e+00 1.32501912e+00 -1.41642106e+00 4.90060180e-01 3.04742396e-01 -4.31409061e-01 -1.70468479e-01 -4.94598597e-01 -9.26839411e-01 4.04501021e-01 1.79253674e+00 -8.78608301e-02 5.81297278e-01 7.08526134e-01 7.34182894e-01 9.00709480e-02 -8.86037946e-01 1.21523726e+00 1.09581970e-01 -9.20749128e-01 -7.28188902e-02 1.67049706e-01 -1.03893019e-01 5.84822953e-01 6.10598288e-02 4.63374823e-01 -6.35712594e-02 -9.66289818e-01 9.13122952e-01 -9.73881930e-02 1.34768367e+00 -6.35946035e-01 5.71167111e-01 4.39057827e-01 -1.55370498e+00 -5.99142984e-02 -1.42940402e-01 2.61724293e-01 5.67835391e-01 1.91750214e-01 -1.04999959e+00 5.04544191e-03 6.64090455e-01 1.22573942e-01 -1.15074709e-01 1.01120305e+00 1.50459826e-01 1.25819457e+00 -6.12524092e-01 -4.58781600e-01 3.99605572e-01 4.09802824e-01 9.22609925e-01 1.72165430e+00 6.30768478e-01 7.85411447e-02 -3.07065636e-01 2.76478827e-01 -2.32100502e-01 1.88126764e-03 -2.99323559e-01 -1.74723431e-01 9.53400195e-01 8.00309658e-01 -3.85011584e-01 -3.11394364e-01 -2.48693779e-01 1.18664169e+00 -1.09429806e-01 5.87687135e-01 -9.76929545e-01 -7.06041217e-01 8.56115818e-01 3.56888846e-02 6.07346356e-01 -4.41451877e-01 3.42278183e-01 -9.63518620e-01 1.75423980e-01 -1.07304108e+00 4.35748473e-02 -6.54657781e-01 -7.24916458e-01 8.34502876e-01 -4.61960405e-01 -1.08295631e+00 -5.89677215e-01 -2.50927955e-01 -6.70547724e-01 8.61850798e-01 -1.63402653e+00 -6.51892543e-01 1.17235847e-01 4.29015428e-01 1.38234568e+00 -3.21569651e-01 7.31105030e-01 8.30334306e-01 -5.79061389e-01 1.12414980e+00 3.26557130e-01 2.73877755e-02 4.65359509e-01 -1.00259221e+00 9.67714310e-01 8.02924216e-01 3.28932583e-01 4.01050776e-01 5.68804443e-01 -3.20195138e-01 -1.76038730e+00 -9.83466685e-01 1.15145624e+00 -8.38651508e-02 7.64563262e-01 -5.46035945e-01 -9.07872558e-01 2.92548299e-01 1.63039103e-01 -5.35211936e-02 4.89323556e-01 -1.50266141e-01 -1.49743170e-01 -2.55006343e-01 -7.92139411e-01 6.52925551e-01 9.10911977e-01 -1.12590420e+00 -2.66617596e-01 2.88216472e-02 1.21369970e+00 -3.39768231e-01 -7.56617546e-01 2.22903222e-01 6.53983176e-01 -6.62749171e-01 8.65685046e-01 -2.76444018e-01 -3.26012701e-01 -9.96518228e-03 -5.54088295e-01 -1.01131785e+00 1.74265862e-01 -1.20796859e+00 -4.86940205e-01 1.38449693e+00 5.67046702e-01 -6.04280174e-01 6.99013293e-01 4.23499286e-01 -3.91422123e-01 -6.01523936e-01 -1.48101699e+00 -9.10029233e-01 -4.53706354e-01 -1.03092027e+00 2.25038722e-01 3.48901987e-01 -7.97382072e-02 4.12791193e-01 -5.35117209e-01 2.35828459e-01 2.72724003e-01 -5.11590779e-01 3.15016240e-01 -4.76013929e-01 -2.89149761e-01 -3.46735269e-01 -5.18279662e-03 -1.84355831e+00 -7.01986849e-02 -5.44753969e-01 3.98848295e-01 -1.32541609e+00 -3.32414776e-01 -1.97582051e-01 -3.79736364e-01 2.19352007e-01 3.02607536e-01 -7.86415264e-02 3.37795883e-01 1.02348216e-01 -5.68154216e-01 8.06997061e-01 6.66332662e-01 -6.51236996e-02 -3.10955048e-01 3.23148996e-01 -9.48173702e-02 2.58039057e-01 7.24120259e-01 -5.15552759e-01 -3.26711416e-01 -6.27806485e-01 -3.13581079e-01 7.41977453e-01 -4.01865765e-02 -8.69959295e-01 6.71219289e-01 2.59490430e-01 -2.76625007e-01 -9.01324272e-01 7.79910445e-01 -5.69790959e-01 -2.16024458e-01 4.49156135e-01 -5.89367628e-01 1.04226246e-01 2.94841826e-01 5.79445243e-01 -4.04771864e-01 -1.79850504e-01 5.10419607e-01 1.84566155e-01 -4.47742909e-01 1.38410792e-01 -1.08828568e+00 2.17456430e-01 2.90205389e-01 -1.61988139e-01 -1.56187415e-01 -1.04427946e+00 -4.35614228e-01 2.36181468e-01 -2.75728136e-01 6.49214089e-01 1.02604079e+00 -8.91303718e-01 -7.08557010e-01 1.05048820e-01 -1.14791684e-01 -3.30753535e-01 7.34046698e-02 7.06914842e-01 -4.56907809e-01 8.39166045e-01 7.11441576e-01 -6.30561411e-01 -1.73128462e+00 6.76161125e-02 3.74167621e-01 1.07172966e-01 -5.04327834e-01 8.04990172e-01 -2.05060661e-01 -6.93540275e-02 1.00712228e+00 -3.99811566e-01 1.08425885e-01 -7.10249469e-02 5.64376295e-01 7.11060047e-01 3.79045576e-01 -6.33152068e-01 -4.63968396e-01 1.03309825e-01 -1.08681932e-01 -6.77273631e-01 9.83014405e-01 -5.26999891e-01 4.33579683e-01 5.75752139e-01 1.40404725e+00 2.31337920e-01 -7.27360070e-01 -2.83055186e-01 -9.60728247e-03 -3.00033063e-01 5.35460293e-01 -6.77826583e-01 -8.85031700e-01 1.37544560e+00 6.67547047e-01 3.05212796e-01 9.69378829e-01 6.34516180e-02 1.67211497e+00 4.68717813e-01 1.65738061e-01 -1.20972228e+00 3.21651250e-03 8.28968823e-01 8.64391446e-01 -1.04637647e+00 -5.91141224e-01 -5.19513674e-02 -5.48097789e-01 1.04979146e+00 2.04203740e-01 5.06179273e-01 4.50759530e-01 5.03244936e-01 2.29952306e-01 1.80063471e-01 -1.21013832e+00 -3.87521416e-01 1.99635893e-01 3.91379535e-01 3.71328205e-01 -3.62808183e-02 1.34354681e-01 2.17340976e-01 -1.48712516e-01 -4.61309791e-01 3.40158165e-01 8.55660379e-01 -6.29440904e-01 -7.82249570e-01 -2.17337623e-01 4.17336464e-01 -9.07779038e-01 -5.25301397e-01 -2.53365338e-01 2.63825148e-01 -8.81051302e-01 1.55894899e+00 1.57472640e-01 -5.91600478e-01 4.03538793e-01 1.04832739e-01 -2.07832962e-01 -4.76072550e-01 -8.13055098e-01 7.64721453e-01 5.44505537e-01 -5.52498579e-01 4.12904285e-02 -4.63784128e-01 -1.16849244e+00 -3.95132512e-01 -7.29273736e-01 2.91156799e-01 1.00735343e+00 7.96512187e-01 4.86668855e-01 7.37515271e-01 8.25458944e-01 -5.97919583e-01 -8.39589894e-01 -1.09263206e+00 -5.01403689e-01 -4.01947737e-01 8.02581549e-01 6.22107498e-02 -6.35868549e-01 -2.81179935e-01]	[14.497220039367676, 6.55946159362793]
6144584d-8487-4c0b-abc5-dadb3126848d	regression-bugs-are-in-your-model-measuring	2105.03048	null	https://arxiv.org/abs/2105.03048v1	https://arxiv.org/pdf/2105.03048v1.pdf	Regression Bugs Are In Your Model! Measuring, Reducing and Analyzing Regressions In NLP Model Updates	Behavior of deep neural networks can be inconsistent between different versions. Regressions during model update are a common cause of concern that often over-weigh the benefits in accuracy or efficiency gain. This work focuses on quantifying, reducing and analyzing regression errors in the NLP model updates. Using negative flip rate as regression measure, we show that regression has a prevalent presence across tasks in the GLUE benchmark. We formulate the regression-free model updates into a constrained optimization problem, and further reduce it into a relaxed form which can be approximately optimized through knowledge distillation training method. We empirically analyze how model ensemble reduces regression. Finally, we conduct CheckList behavioral testing to understand the distribution of regressions across linguistic phenomena, and the efficacy of ensemble and distillation methods.	['Stefano Soatto', 'Yi Zhang', 'Yuanjun Xiong', 'Yi-An Lai', 'Yuqing Xie']	2021-05-07	null	https://aclanthology.org/2021.acl-long.515	https://aclanthology.org/2021.acl-long.515.pdf	acl-2021-5	['negative-flip-rate']	['computer-vision']	[-1.04618192e-01 2.15030294e-02 -3.63170922e-01 -6.13311231e-01 -6.26992583e-01 -5.27183712e-01 5.13251722e-01 -3.12664032e-01 -8.31131876e-01 1.01581967e+00 1.95258513e-01 -5.23301363e-01 -6.60658181e-02 -3.86932969e-01 -1.02062309e+00 -3.94402415e-01 4.40182745e-01 1.25401407e-01 -8.45656544e-02 -1.62425086e-01 3.19126040e-01 2.63949841e-01 -9.52082396e-01 1.08616762e-01 1.19759357e+00 8.61106992e-01 -2.32975315e-02 5.26535153e-01 1.14970103e-01 1.23510802e+00 -6.75829828e-01 -8.23911726e-01 3.05094004e-01 -1.42397895e-01 -5.01232326e-01 -8.42763901e-01 7.92402864e-01 -3.17442775e-01 -3.05134445e-01 1.11829269e+00 5.58130085e-01 2.58954316e-01 7.05933809e-01 -1.22767794e+00 -1.01397586e+00 1.21098292e+00 -6.46918297e-01 5.91202378e-01 -2.58239150e-01 2.79439211e-01 9.75323498e-01 -6.91033840e-01 4.49519664e-01 1.22247350e+00 1.08251619e+00 5.29030085e-01 -1.45891941e+00 -1.18301356e+00 5.61995566e-01 5.20030782e-02 -1.36056459e+00 -7.66903281e-01 5.52731872e-01 -6.44156098e-01 1.23228109e+00 9.80542749e-02 3.35316449e-01 1.38607180e+00 3.86109620e-01 4.70407724e-01 1.13211203e+00 -2.49150336e-01 -5.22959717e-02 4.71980333e-01 8.60234678e-01 7.29719818e-01 8.32181990e-01 2.35318050e-01 -7.95164764e-01 1.56276599e-01 5.00904024e-01 -3.05968732e-01 -4.32991624e-01 2.02670798e-01 -8.36354852e-01 7.46143699e-01 4.60762888e-01 1.46035492e-01 -2.50495136e-01 5.67359149e-01 3.63776475e-01 5.08514404e-01 7.64547408e-01 1.01271975e+00 -9.34953809e-01 -2.50530928e-01 -8.18355024e-01 4.39393014e-01 9.39449072e-01 7.65825689e-01 6.33400142e-01 2.49178261e-01 -4.27279472e-01 9.72707033e-01 2.46181160e-01 3.71049345e-01 7.10021794e-01 -1.13936377e+00 7.94601262e-01 4.64656383e-01 1.32953990e-02 -1.05411482e+00 -3.98138911e-01 -6.42534494e-01 -7.35056937e-01 -1.81758314e-01 2.03281969e-01 -7.53942251e-01 -5.56828618e-01 2.07405472e+00 -1.98373079e-01 -2.14854162e-02 -3.43323946e-02 5.61962962e-01 7.14253545e-01 3.07178855e-01 3.26425999e-01 -2.78644115e-01 8.67465019e-01 -1.18747890e+00 -9.36115921e-01 -6.69417500e-01 1.09094763e+00 -2.34900787e-01 1.11441433e+00 3.21153849e-01 -1.00546610e+00 -3.90343517e-01 -1.16762090e+00 -4.09336299e-01 -3.75018418e-01 2.64815181e-01 6.46725237e-01 8.26586306e-01 -1.06214237e+00 9.07347322e-01 -7.86255062e-01 1.06531307e-01 6.62809014e-01 3.28811973e-01 -1.95680916e-01 1.89756513e-01 -1.20552242e+00 1.54338062e+00 4.91656452e-01 2.24290371e-01 -5.56308806e-01 -1.28947878e+00 -6.63101315e-01 1.39801484e-02 3.49055737e-01 -7.64185786e-01 1.45265889e+00 -1.04554379e+00 -1.53380370e+00 7.30067611e-01 -3.51183355e-01 -6.19088531e-01 4.83617455e-01 -7.00058997e-01 -2.27560267e-01 -1.00011289e+00 -2.24155128e-01 4.96533215e-01 5.66507041e-01 -1.16933632e+00 -5.39789021e-01 -3.68473858e-01 6.67980164e-02 2.73248792e-01 -3.06996465e-01 -9.19550210e-02 -1.76844910e-01 -6.30713999e-01 -4.06238556e-01 -9.32623506e-01 -5.17457277e-02 -5.02073765e-01 -4.54938263e-01 -2.10168183e-01 2.30906531e-01 -1.10637045e+00 1.82169294e+00 -1.98535478e+00 2.48314843e-01 5.49583361e-02 4.77977902e-01 -2.61056609e-02 -3.03405493e-01 -3.23382258e-01 -1.29665494e-01 8.11025023e-01 -3.40054750e-01 -4.08022374e-01 2.47483160e-02 1.23249941e-01 -4.39159751e-01 2.39923686e-01 3.23701262e-01 1.03917956e+00 -5.18354058e-01 -2.70318836e-01 -3.68929297e-01 3.02569062e-01 -8.82372022e-01 -2.43442178e-01 -1.94286913e-01 1.85533807e-01 -1.70981094e-01 4.01807010e-01 6.51106238e-01 -1.58798829e-01 1.04764134e-01 -5.12472451e-01 -1.59949437e-01 5.34050107e-01 -8.10628891e-01 1.47990942e+00 -5.12981653e-01 1.00511622e+00 -3.31930488e-01 -1.02932477e+00 6.55842781e-01 -2.38234758e-01 1.08736344e-01 -9.42182004e-01 1.11552022e-01 6.66792840e-02 3.84216607e-01 -3.14865112e-01 7.88163066e-01 4.58083414e-02 1.51597649e-01 3.70526016e-01 3.15204076e-02 2.14346908e-02 2.79420376e-01 -1.02982394e-01 1.05623567e+00 1.33588567e-01 2.16067806e-01 -3.18202019e-01 1.58286579e-02 -1.78616680e-02 6.29518807e-01 1.09534299e+00 -6.65729567e-02 5.57294600e-02 7.93001115e-01 -5.25752045e-02 -6.93282068e-01 -8.04261804e-01 -1.71037808e-01 1.42709243e+00 -3.44558984e-01 -2.18467429e-01 -5.95517874e-01 -8.03717673e-01 1.44695774e-01 1.17426193e+00 -8.76854956e-01 -6.98399365e-01 -6.06860280e-01 -1.18245697e+00 7.88264751e-01 6.23955369e-01 7.14706182e-01 -7.88988352e-01 -3.21017534e-01 -1.31025165e-01 -2.36825377e-01 -1.00074267e+00 -2.26782247e-01 4.18261826e-01 -1.00955284e+00 -8.71406436e-01 -2.47702181e-01 -4.20572102e-01 5.39792299e-01 -7.74525777e-02 1.40615666e+00 1.15316771e-01 9.11768004e-02 1.04035109e-01 -8.03413391e-02 -6.38200581e-01 -2.64241606e-01 6.84917986e-01 1.93774179e-01 -6.18008196e-01 4.28952694e-01 -1.60599396e-01 -2.75536448e-01 7.05027282e-02 -3.97397727e-01 -2.70044450e-02 6.04333341e-01 6.59641445e-01 4.34347928e-01 -5.28253168e-02 5.58952928e-01 -1.20202577e+00 1.31681192e+00 -6.22435629e-01 -5.49684465e-01 5.11259794e-01 -1.05983865e+00 5.22995889e-01 2.34311834e-01 -6.37410164e-01 -1.36690462e+00 -4.12538081e-01 2.61198997e-01 -4.80325632e-02 4.94746506e-01 7.89168417e-01 2.42737815e-01 -1.78580940e-01 9.04204726e-01 -2.27872640e-01 -1.97192922e-01 -4.29412425e-01 5.07779062e-01 2.97666520e-01 1.25263289e-01 -6.78544879e-01 5.33791959e-01 -6.14125878e-02 -4.50777084e-01 -3.87242883e-01 -1.27935779e+00 2.42371500e-01 -3.08167011e-01 -3.73343751e-02 6.03636563e-01 -1.06417012e+00 -6.24726951e-01 4.27189440e-01 -1.23699689e+00 -9.29858983e-01 -1.66791812e-01 4.95567232e-01 -1.91128314e-01 -1.81974977e-01 -5.35872161e-01 -5.18460631e-01 -3.49128783e-01 -1.14624190e+00 3.48604351e-01 2.84700125e-01 -5.46844542e-01 -1.08142686e+00 3.83059382e-01 2.45016783e-01 6.08972549e-01 -9.60666686e-02 1.02442622e+00 -8.26928556e-01 -8.19555521e-02 6.75275400e-02 -4.71593112e-01 5.18515646e-01 -1.14046328e-01 3.40203851e-01 -1.00925684e+00 2.01718286e-02 -3.56366746e-02 -2.84963429e-01 1.25100124e+00 6.61036193e-01 1.35108578e+00 -5.56654692e-01 -3.27361763e-01 8.02613795e-01 1.25000286e+00 3.57899554e-02 4.23180908e-01 3.78202617e-01 7.53406763e-01 4.63063926e-01 1.84466064e-01 1.76719561e-01 3.71779233e-01 4.03411806e-01 -8.37632827e-03 3.92140687e-01 6.57925010e-02 -1.39905304e-01 6.67274654e-01 9.18935537e-01 -2.37674311e-01 -2.93058455e-01 -9.66653645e-01 1.99879795e-01 -1.83111107e+00 -8.50742280e-01 -9.25355926e-02 1.88140845e+00 1.41703081e+00 8.66533965e-02 -3.12451392e-01 -3.37403983e-01 5.50205350e-01 1.65757071e-02 -8.78488898e-01 -5.29206634e-01 -1.25092551e-01 2.27745384e-01 9.45287883e-01 6.26417518e-01 -6.83663547e-01 1.21983361e+00 7.55564117e+00 6.42500281e-01 -1.13094795e+00 3.07179451e-01 8.42874467e-01 -5.36647141e-01 -3.13298285e-01 -1.67980194e-01 -1.16158307e+00 3.59066516e-01 1.17836404e+00 -3.93523216e-01 6.73973799e-01 8.33424628e-01 6.41989559e-02 -1.71399638e-01 -1.50420487e+00 8.58057141e-01 -5.81405461e-02 -1.25767839e+00 -6.00789078e-02 -1.10785306e-01 1.10987580e+00 3.85855854e-01 4.88324225e-01 1.00506556e+00 8.63453805e-01 -1.30980372e+00 8.09814274e-01 7.43107915e-01 6.01901829e-01 -4.95653838e-01 6.63027763e-01 2.77358413e-01 -5.40989280e-01 -2.02540740e-01 -4.12885934e-01 -2.32752323e-01 -2.99409598e-01 5.68621814e-01 -6.32617295e-01 3.35328691e-02 8.29152644e-01 7.15563834e-01 -8.53669047e-01 6.72525764e-01 -1.96656704e-01 7.50440240e-01 -2.31866866e-01 1.03057306e-02 -2.74246991e-01 -1.25856787e-01 2.57423580e-01 1.22915542e+00 1.80060700e-01 -1.47277191e-01 -6.03020132e-01 1.14414787e+00 -6.59700394e-01 -1.20279670e-01 -5.53740442e-01 -2.50311524e-01 8.95257473e-01 9.39090371e-01 -1.85374081e-01 -1.34440005e-01 -2.74034023e-01 6.36948407e-01 1.08440483e+00 6.34071350e-01 -1.14568555e+00 -1.67975072e-02 8.28720927e-01 -2.35647857e-01 -5.90277240e-02 -1.26511350e-01 -8.47162008e-01 -1.19070005e+00 1.40353978e-01 -9.77797270e-01 -2.04135291e-02 -7.41019070e-01 -1.23153508e+00 2.07287490e-01 7.68469721e-02 -2.69476891e-01 -1.17348306e-01 -7.21220851e-01 -4.12916213e-01 8.74138176e-01 -1.62101340e+00 -3.92328024e-01 -3.06359321e-01 2.77877808e-01 3.60012472e-01 -1.56319857e-01 4.64304507e-01 2.66524076e-01 -1.34158695e+00 1.10661650e+00 2.21904397e-01 8.84144381e-02 1.06582403e+00 -9.97256279e-01 3.12031955e-01 6.99817717e-01 -1.86593235e-01 1.04979658e+00 5.43644905e-01 -7.23750710e-01 -1.03317797e+00 -1.13977528e+00 7.83447802e-01 -7.78910041e-01 6.94687545e-01 9.48334858e-02 -9.84710276e-01 1.20233560e+00 1.76170453e-01 -3.55291456e-01 5.99218011e-01 5.99800467e-01 -5.75869560e-01 -3.01856428e-01 -8.93046081e-01 8.56622696e-01 1.25049901e+00 -4.18596387e-01 -5.84575593e-01 5.10395365e-03 1.13277423e+00 -4.35780138e-01 -9.97083664e-01 4.95914668e-01 6.31293893e-01 -6.91124499e-01 6.94381118e-01 -9.59330440e-01 7.32790828e-01 3.87815446e-01 -7.94587582e-02 -1.68870401e+00 -3.98152411e-01 -4.34137940e-01 -3.37697893e-01 1.27390909e+00 1.10226810e+00 -9.00104761e-01 5.34363508e-01 1.08810043e+00 -9.04283375e-02 -6.29933298e-01 -7.63189793e-01 -5.74457467e-01 5.04141331e-01 -5.18104494e-01 2.98755854e-01 1.23376012e+00 -2.54107267e-01 4.44104522e-01 -1.27042696e-01 -1.98353782e-01 2.46643931e-01 -6.82862759e-01 6.74560130e-01 -1.25649500e+00 -2.38175347e-01 -9.70637381e-01 2.93363243e-01 -8.36730361e-01 6.18097961e-01 -9.43160176e-01 6.24737926e-02 -1.27105916e+00 2.74039716e-01 -3.71191829e-01 -3.30819070e-01 5.76603889e-01 -3.92944574e-01 -2.35315144e-01 1.89904645e-01 2.05975592e-01 -4.76839900e-01 5.47381639e-01 9.80482161e-01 -8.84185359e-02 -3.95749509e-01 -3.72868776e-01 -1.20979190e+00 7.54141092e-01 1.01057673e+00 -7.07417428e-01 -3.79612386e-01 -1.01347589e+00 1.01994431e+00 -2.66589165e-01 2.11732179e-01 -8.68277609e-01 2.34751567e-01 -1.36593506e-01 1.16870396e-01 -2.13866532e-01 3.42667475e-02 -7.42932022e-01 -7.11241066e-02 3.29752684e-01 -9.01412547e-01 4.36822444e-01 5.98497808e-01 5.05723953e-01 1.24805301e-01 -3.78168643e-01 5.88571370e-01 7.38050938e-02 -5.01477003e-01 1.02060951e-01 -4.49494012e-02 4.45286453e-01 7.07808018e-01 -8.84523541e-02 -7.00803041e-01 1.38998672e-01 -5.28152645e-01 3.82990301e-01 1.03048002e-02 5.35186470e-01 2.83297181e-01 -1.14347315e+00 -8.06368232e-01 -9.12135914e-02 -9.62732062e-02 -6.34476170e-02 1.97540015e-01 1.09236538e+00 -4.79908764e-01 4.67748612e-01 5.39560318e-02 -1.53531730e-01 -9.34578538e-01 -1.16723105e-02 8.53788257e-01 -6.64481699e-01 5.58671243e-02 1.15683007e+00 7.55734593e-02 -6.39862955e-01 3.62199724e-01 -6.17017746e-01 -1.59294307e-01 2.22208306e-01 1.02851592e-01 6.26361132e-01 9.83087942e-02 2.46241502e-02 -2.44310766e-01 3.66384029e-01 -4.92004693e-01 -1.63433149e-01 1.22680867e+00 -3.85167915e-03 -1.51037678e-01 7.17211306e-01 1.03698075e+00 -1.17138579e-01 -1.22148883e+00 -3.42831880e-01 1.41199648e-01 6.29656091e-02 3.58756602e-01 -1.13810980e+00 -1.18976533e+00 5.66080272e-01 3.87192160e-01 -1.52235866e-01 6.80320919e-01 -3.97482336e-01 2.30626836e-01 8.93483698e-01 -1.02191865e-01 -1.33315980e+00 -1.39671460e-01 9.96613741e-01 1.04044378e+00 -1.42340231e+00 1.67916611e-01 1.61138505e-01 -7.72962987e-01 6.66787803e-01 1.21770585e+00 -2.38153845e-01 8.69923830e-01 4.37906742e-01 -3.29110563e-01 5.05213290e-02 -1.18068039e+00 3.14359069e-01 2.08625972e-01 5.28597057e-01 8.23169887e-01 -4.94367592e-02 -4.03737277e-01 1.07791865e+00 -6.39496624e-01 1.05538726e-01 2.45257750e-01 4.36210215e-01 -2.04716414e-01 -5.79841912e-01 -2.16494631e-02 8.04373741e-01 -5.16319335e-01 -5.63643277e-01 -6.02474451e-01 1.03999829e+00 -4.71831188e-02 7.59245098e-01 2.92484552e-01 -6.06602907e-01 3.59546870e-01 3.23437810e-01 5.36470771e-01 -4.88240659e-01 -9.13302004e-01 -7.90010810e-01 3.64483982e-01 -5.63339174e-01 -1.71792224e-01 -3.13113391e-01 -9.25779402e-01 -6.27528667e-01 -6.36816740e-01 -2.51727849e-01 6.55172527e-01 1.03592455e+00 4.93676245e-01 8.35605264e-01 1.32151991e-01 -3.94332170e-01 -9.27079558e-01 -1.35748279e+00 -1.32890552e-01 1.40936300e-01 2.92512000e-01 -6.82323992e-01 -6.49513900e-01 -1.21284332e-02]	[10.660789489746094, 8.232978820800781]
faeaec64-63d5-4305-b335-8848d74410ad	compositional-sketch-search	2106.08009	null	https://arxiv.org/abs/2106.08009v1	https://arxiv.org/pdf/2106.08009v1.pdf	Compositional Sketch Search	We present an algorithm for searching image collections using free-hand sketches that describe the appearance and relative positions of multiple objects. Sketch based image retrieval (SBIR) methods predominantly match queries containing a single, dominant object invariant to its position within an image. Our work exploits drawings as a concise and intuitive representation for specifying entire scene compositions. We train a convolutional neural network (CNN) to encode masked visual features from sketched objects, pooling these into a spatial descriptor encoding the spatial relationships and appearances of objects in the composition. Training the CNN backbone as a Siamese network under triplet loss yields a metric search embedding for measuring compositional similarity which may be efficiently leveraged for visual search by applying product quantization.	['John Collomosse', 'Hailin Jin', 'Long Mai', 'Tu Bui', 'Alexander Black']	2021-06-15	null	null	null	null	['sketch-based-image-retrieval']	['computer-vision']	[ 3.26830029e-01 -5.21694481e-01 -4.91982907e-01 -2.42035940e-01 -7.65499115e-01 -8.96279275e-01 9.09853876e-01 2.57774681e-01 -1.60255462e-01 5.20559102e-02 1.93520218e-01 7.15751797e-02 -2.72203952e-01 -7.09140420e-01 -7.04830110e-01 -4.77225065e-01 -1.20843187e-01 4.91156071e-01 1.21637478e-01 -2.32917387e-02 7.40077376e-01 1.17547667e+00 -1.57083619e+00 4.60897595e-01 -1.18942313e-01 1.46743345e+00 2.70016402e-01 6.66158497e-01 -2.77521640e-01 5.26796341e-01 -6.56622529e-01 -5.19713521e-01 4.45870101e-01 2.95394640e-02 -6.74475431e-01 1.46290794e-01 1.19243908e+00 -6.81053281e-01 -7.66830146e-01 9.02897358e-01 1.87087238e-01 1.51792601e-01 8.61385345e-01 -1.54861641e+00 -1.25244868e+00 4.42429781e-02 -3.31401527e-01 4.34710458e-03 5.92574835e-01 2.95157973e-02 1.59914231e+00 -1.09042728e+00 1.01128876e+00 1.31215012e+00 4.32802528e-01 1.13711424e-01 -1.40088427e+00 -5.12601912e-01 -1.41109571e-01 1.97637483e-01 -1.83803940e+00 -4.04825866e-01 1.14218056e+00 -4.41804320e-01 8.34611237e-01 4.11193073e-01 6.72107339e-01 8.42835128e-01 -2.10976794e-01 9.54870760e-01 3.34874928e-01 -2.41445646e-01 3.78005579e-02 -2.59762228e-01 -2.10468233e-01 1.09799230e+00 -4.86402847e-02 7.97362067e-03 -8.74777794e-01 -3.85402083e-01 1.31052089e+00 5.81415176e-01 4.44220565e-02 -8.66055846e-01 -1.38032830e+00 9.37097549e-01 9.13733602e-01 1.61073685e-01 -2.82314241e-01 8.04177284e-01 3.29439640e-01 3.90716493e-01 5.97435981e-03 7.88053513e-01 -4.81202565e-02 1.38147742e-01 -1.11558044e+00 7.19218850e-01 4.81462628e-01 1.37666118e+00 1.05492163e+00 -4.31586444e-01 -5.02709925e-01 8.58561873e-01 5.43861948e-02 5.73781490e-01 1.65976778e-01 -1.41880929e+00 2.57361501e-01 7.00058520e-01 1.50521368e-01 -1.39162052e+00 3.67663205e-01 2.36624807e-01 -4.11592960e-01 2.14621320e-01 -3.17995921e-02 9.27894175e-01 -7.25808680e-01 1.50896072e+00 -1.74277931e-01 -1.87441409e-01 -4.34289396e-01 9.82511044e-01 6.85380697e-01 3.27485710e-01 -1.08916163e-02 6.79782748e-01 1.40430737e+00 -9.61883545e-01 -2.72828221e-01 1.88784674e-01 1.79928392e-01 -9.60352600e-01 1.09993720e+00 1.12221956e-01 -1.47426319e+00 -4.98446196e-01 -1.28461063e+00 -6.45883560e-01 -8.41819346e-01 3.14740986e-01 5.12644887e-01 2.54719973e-01 -1.16060674e+00 7.74305165e-01 -3.76691639e-01 -1.89791635e-01 5.93220592e-01 3.23910892e-01 -5.87304235e-01 -1.76056519e-01 -6.35049164e-01 5.71962893e-01 1.34133995e-02 -2.69014120e-01 -8.91882360e-01 -7.97000587e-01 -9.09775436e-01 2.79019952e-01 -9.30985436e-03 -5.81787646e-01 1.08494949e+00 -7.89698899e-01 -1.23517573e+00 1.04417551e+00 -3.06712717e-01 -7.13173151e-02 2.65441179e-01 3.09461623e-01 -1.61731631e-01 6.81845427e-01 3.91287118e-01 1.14335716e+00 1.28180742e+00 -1.17073214e+00 -4.32076842e-01 -2.66500860e-01 9.42042768e-02 -2.76582595e-03 -7.65317604e-02 2.36905545e-01 -6.97963417e-01 -9.96175230e-01 3.40022504e-01 -7.33770370e-01 2.82147139e-01 1.16605282e+00 -3.29199165e-01 -4.60260093e-01 1.13269305e+00 -2.30430931e-01 9.93888319e-01 -2.15319967e+00 1.16129689e-01 4.27752376e-01 2.03376025e-01 -6.72152340e-02 -5.94986081e-01 8.59041154e-01 6.32936954e-02 2.32555330e-01 -4.39018616e-03 -3.74858022e-01 4.06915218e-01 2.09541515e-01 -7.15614438e-01 3.91222686e-01 6.22947931e-01 1.40503073e+00 -1.05477881e+00 -5.46692789e-01 2.26393849e-01 5.78868926e-01 -3.58160436e-01 3.25630069e-01 -2.03265831e-01 -2.63004601e-01 -3.79035026e-01 9.96457160e-01 5.79014599e-01 -3.51029843e-01 -1.88547894e-01 -6.11630976e-01 1.45324290e-01 2.15937093e-01 -9.12162364e-01 2.08800411e+00 -3.08798611e-01 8.89023125e-01 -1.02253057e-01 -7.84190297e-01 1.03475940e+00 1.71573821e-03 3.99783462e-01 -8.04114759e-01 -3.02491456e-01 1.77991122e-01 -7.92502284e-01 -1.52891949e-01 7.55922019e-01 3.16059768e-01 -2.39909589e-01 6.60329700e-01 8.16509351e-02 -5.06156385e-01 1.39417769e-02 3.20537627e-01 9.83482242e-01 5.65632712e-03 -4.74243499e-02 -5.45384996e-02 3.26098382e-01 -9.88389850e-02 -3.77069771e-01 8.06744695e-01 -1.56839900e-02 8.29369247e-01 1.64606392e-01 -9.23828363e-01 -1.51380897e+00 -1.66918683e+00 -4.15271968e-02 1.27272916e+00 3.44979107e-01 -4.96922463e-01 -7.44248480e-02 -2.89087117e-01 5.32411695e-01 5.82847334e-02 -7.43986666e-01 -9.17115286e-02 -5.41232049e-01 5.08039176e-01 6.02488816e-01 6.63697243e-01 2.13459715e-01 -1.16227901e+00 -7.42191792e-01 -3.63026932e-02 1.44150749e-01 -8.64028931e-01 -1.25096822e+00 5.27230725e-02 -4.54809815e-01 -1.15104330e+00 -9.52186286e-01 -1.18199289e+00 7.74053633e-01 6.00053608e-01 1.15258360e+00 2.30588451e-01 -8.62932146e-01 6.91681147e-01 -1.26309276e-01 8.75581652e-02 -8.99382606e-02 -8.40485767e-02 -2.54558116e-01 -7.81599490e-04 2.21998349e-01 -5.19137561e-01 -9.18977976e-01 2.67126739e-01 -1.13082242e+00 -3.01651537e-01 3.86720210e-01 7.97348917e-01 8.53320897e-01 -5.98938107e-01 -2.32495800e-01 -1.80990681e-01 7.65904963e-01 6.51825219e-02 -5.66734850e-01 7.34970570e-01 -1.71542391e-01 4.95329142e-01 3.47695529e-01 -5.64098299e-01 -1.05423011e-01 7.99807385e-02 5.24406195e-01 -1.24794030e+00 6.45876750e-02 1.20520875e-01 1.68312266e-01 -3.74028325e-01 2.83501893e-01 4.98785645e-01 3.12639810e-02 -4.12136763e-01 8.72155190e-01 3.19570214e-01 8.41472328e-01 -8.41462553e-01 7.09311604e-01 6.57350123e-01 3.36940646e-01 -6.82296455e-01 -4.11290854e-01 -6.89924419e-01 -8.75252664e-01 2.06515431e-01 8.18745553e-01 -7.07674861e-01 -1.02183533e+00 -1.55992478e-01 -1.40275502e+00 3.80144976e-02 -3.27785879e-01 -1.78593874e-01 -7.65465200e-01 2.57532388e-01 -4.22922373e-01 -5.12839437e-01 -5.10642707e-01 -1.12916017e+00 2.05318952e+00 -2.47833714e-01 -3.97799999e-01 -6.74021125e-01 -1.46552091e-02 -1.09602094e-01 4.16229397e-01 1.37545943e-01 1.07329631e+00 -2.55290985e-01 -1.17972350e+00 -5.55489898e-01 -6.69754446e-01 -1.89722002e-01 1.36447415e-01 2.89809048e-01 -7.33016849e-01 -3.44404340e-01 -8.44895065e-01 -4.44231659e-01 8.74639750e-01 7.61733502e-02 1.59725952e+00 -5.56444585e-01 -3.31009448e-01 7.94228852e-01 1.67615438e+00 -4.23369780e-02 4.36088353e-01 1.22194529e-01 6.18358433e-01 4.07740891e-01 5.46412431e-02 3.88969690e-01 9.82592925e-02 8.59715879e-01 2.95179933e-01 1.56332746e-01 -4.19382602e-01 -8.32096398e-01 -1.44322485e-01 3.18549693e-01 4.05946225e-01 6.28737658e-02 -5.37021518e-01 4.98890191e-01 -1.46999502e+00 -1.26985836e+00 7.52661049e-01 2.11395478e+00 6.62921727e-01 -3.63677442e-01 2.30398048e-02 -1.68274596e-01 5.76791167e-01 4.97807890e-01 -5.17460465e-01 -4.63690311e-01 -2.52439082e-01 4.51187074e-01 5.65901637e-01 3.58405471e-01 -1.16989553e+00 8.43800187e-01 7.05963802e+00 8.82306159e-01 -1.23579669e+00 -1.67881146e-01 3.10652584e-01 -2.17953786e-01 -4.86677259e-01 -1.44147128e-01 -3.37153524e-01 2.43294373e-01 2.83077627e-01 -5.15464060e-02 8.97488356e-01 7.31261849e-01 -3.39919627e-01 2.80845493e-01 -1.46341288e+00 1.43140638e+00 2.17042312e-01 -1.94988108e+00 7.21187174e-01 -6.27795085e-02 7.24282146e-01 5.27549395e-03 4.88809019e-01 -3.70145172e-01 2.81927288e-01 -1.24639726e+00 1.14230728e+00 6.78838193e-01 1.25640237e+00 -5.28606236e-01 -8.21054280e-02 -4.15788919e-01 -1.52998376e+00 -1.61516234e-01 -4.64636683e-01 2.49478996e-01 -2.83834368e-01 -5.13853610e-01 -6.90881550e-01 1.69288933e-01 5.93188941e-01 7.01829910e-01 -6.74322546e-01 8.31400275e-01 9.06705409e-02 -2.98683316e-01 -1.79838106e-01 -1.81896955e-01 4.78867501e-01 -3.21368724e-01 3.14716011e-01 1.16969991e+00 3.16600740e-01 -2.61825174e-01 1.59326106e-01 1.44625664e+00 -3.48447025e-01 -6.91767335e-02 -9.36424911e-01 -3.19558084e-01 8.43412876e-01 1.19863439e+00 -5.74333549e-01 -4.86580104e-01 -2.20103100e-01 1.40687943e+00 4.68556255e-01 4.49005365e-01 -3.16972315e-01 -7.33277142e-01 1.03419363e+00 8.34966078e-02 8.80296528e-01 -4.77255911e-01 -1.62790075e-01 -8.18360507e-01 1.18928909e-01 -5.84370255e-01 1.36669904e-01 -1.26486933e+00 -1.48104584e+00 3.45477939e-01 -8.21614861e-02 -1.31069052e+00 -9.73057300e-02 -8.84378791e-01 -5.34873962e-01 1.03290534e+00 -1.23269486e+00 -1.52766037e+00 -2.76667893e-01 6.72499537e-01 3.97809744e-01 -1.97158083e-01 1.11079013e+00 1.07860327e-01 -4.64382321e-02 8.29782724e-01 8.51958469e-02 3.60809743e-01 6.06089056e-01 -1.14949214e+00 6.04740322e-01 8.47962648e-02 7.79423594e-01 1.03412676e+00 1.89391851e-01 -1.53316423e-01 -1.79053664e+00 -9.23958659e-01 8.68479133e-01 -5.91621876e-01 8.17372501e-01 -5.15769124e-01 -5.61804414e-01 4.51381207e-01 6.52767941e-02 2.15577006e-01 2.95367658e-01 -4.01049495e-01 -1.24030697e+00 -1.25860438e-01 -1.00877392e+00 8.57026160e-01 1.03101516e+00 -1.65926123e+00 -3.08358431e-01 3.06156695e-01 6.43136680e-01 -6.05369471e-02 -9.51875269e-01 -1.51883513e-01 1.03482485e+00 -5.39030194e-01 1.56911278e+00 -7.71334171e-01 5.23938000e-01 -2.81150401e-01 -5.95883548e-01 -8.11889887e-01 -3.33327204e-01 -5.24729669e-01 1.69635862e-01 4.43143219e-01 1.20513029e-01 1.12762107e-02 7.86972821e-01 6.72238708e-01 3.16843152e-01 -5.24401307e-01 -8.30668092e-01 -7.65997887e-01 -5.66854328e-02 -6.24701455e-02 9.76923883e-01 7.94650853e-01 -1.28941759e-01 -8.60506669e-02 7.22303391e-02 4.91910167e-02 5.66795468e-01 6.32612944e-01 5.98648787e-01 -1.04442859e+00 1.75543204e-01 -1.15600312e+00 -8.31995845e-01 -1.28249705e+00 2.85236955e-01 -1.13273132e+00 -1.29855335e-01 -1.10750055e+00 2.13001326e-01 -3.62192929e-01 -3.14423680e-01 3.39643776e-01 2.85923690e-01 6.40559614e-01 3.60845864e-01 4.58330572e-01 -7.80799568e-01 3.94148856e-01 1.18312490e+00 -8.08149219e-01 2.57458806e-01 -5.57372630e-01 -1.57893077e-01 1.28139916e-04 3.68766308e-01 -1.68719113e-01 -2.44108230e-01 -7.88329005e-01 1.67157874e-01 -4.04334217e-02 8.82930040e-01 -6.21915877e-01 4.60497260e-01 -2.06953306e-02 5.63298106e-01 -7.06639588e-01 6.75609946e-01 -9.29291606e-01 1.68096140e-01 2.52024174e-01 -9.64871883e-01 6.40015483e-01 -1.16327800e-01 4.88250345e-01 -3.93628240e-01 -2.79597282e-01 3.15722734e-01 -2.39724010e-01 -6.86875105e-01 7.91749477e-01 1.76689178e-01 -2.31891796e-01 5.05339801e-01 -4.12586600e-01 -1.24081694e-01 -4.76355553e-01 -4.18021142e-01 -2.01717988e-01 8.43402028e-01 5.18981099e-01 1.05965281e+00 -1.82148826e+00 -2.52595425e-01 5.24631202e-01 5.58481574e-01 -4.78426546e-01 -2.28875712e-01 7.15814345e-03 -8.88406038e-01 7.32618511e-01 -4.58691299e-01 -5.14872611e-01 -1.10581708e+00 9.07418907e-01 3.16225648e-01 3.56979072e-01 -4.12059188e-01 9.39998567e-01 9.15247574e-02 -2.63979554e-01 4.52865869e-01 -3.44349802e-01 4.61131185e-01 3.31792720e-02 4.70003158e-01 2.37764463e-01 -1.07601486e-01 -6.08142018e-01 -2.82056451e-01 9.54062462e-01 1.57286033e-01 -2.97843456e-01 1.03920901e+00 7.03516323e-03 -3.23831886e-01 2.97392458e-01 2.08297157e+00 -2.88461030e-01 -1.30983424e+00 -5.12023151e-01 1.92408636e-01 -7.33662367e-01 -9.51528102e-02 -4.06945616e-01 -8.62247705e-01 9.61559594e-01 3.73874068e-01 5.49097639e-03 6.61756933e-01 3.15387458e-01 5.65606475e-01 7.98299789e-01 4.19914335e-01 -6.82203233e-01 6.08660758e-01 8.76774117e-02 1.42810917e+00 -1.18616772e+00 8.94185230e-02 3.48622911e-02 -3.54705065e-01 1.56529188e+00 -1.55444354e-01 -5.77702105e-01 7.71187603e-01 1.72069483e-02 -1.13037214e-01 -6.71085179e-01 -5.09645581e-01 -1.41494334e-01 7.42573023e-01 4.45607901e-01 1.13414563e-01 1.73145697e-01 2.26447195e-01 -1.96588382e-01 -1.17488697e-01 -4.07895952e-01 -2.35695094e-01 8.91937137e-01 -2.80904770e-01 -1.07226884e+00 -1.62066948e-02 3.60516071e-01 -1.23119168e-01 -2.31099606e-01 -5.84242344e-01 4.68426138e-01 -1.06019035e-01 3.97191435e-01 5.44273794e-01 -1.11858010e-01 2.75884122e-01 3.40907909e-02 7.80492127e-01 -3.49184304e-01 -2.99998283e-01 -2.58374244e-01 -5.12104392e-01 -9.31828678e-01 -3.30826163e-01 -4.83062625e-01 -7.33523667e-01 -1.10943824e-01 1.43945754e-01 -2.69306332e-01 8.28051627e-01 2.79478788e-01 3.38059157e-01 -1.36422589e-01 6.74058497e-01 -1.05763018e+00 -6.32200539e-01 -3.68537039e-01 -6.54235542e-01 9.14013505e-01 5.90749085e-01 -6.01964116e-01 -1.06531590e-01 1.39131889e-01]	[11.640660285949707, 0.5519562363624573]
142ecea9-0db3-4a48-a17b-628549f8a990	multivariate-pair-trading-by-volatility-model	2106.09132	null	https://arxiv.org/abs/2106.09132v1	https://arxiv.org/pdf/2106.09132v1.pdf	Multivariate Pair Trading by Volatility & Model Adaption Trade-off	Pair trading is one of the most discussed topics among financial researches. Despite a growing base of work, portfolio management for multivariate time series is rarely discussed. On the other hand, most researches focus on refining strategy rules instead of finding the optimal portfolio weight. In this paper, we brought up a simple yet profitable strategy called Volatility & Model Adaption Trade-off (VMAT) to leverage the issues. Experiment studies show its superior profit performance over baselines.	['Tianyang Xie', 'Chenyanzi Yu']	2021-06-11	null	null	null	null	['pair-trading']	['time-series']	[-4.72742468e-01 -3.61981720e-01 -1.85304210e-01 -3.90304655e-01 -4.57973704e-02 -6.85167074e-01 5.82495570e-01 -1.48612261e-01 -2.08251134e-01 7.92955875e-01 -2.43613094e-01 -5.46159923e-01 -5.09657919e-01 -8.36286783e-01 -4.41200696e-02 -6.99279130e-01 8.84918571e-02 3.58809859e-01 3.62531304e-01 -2.89713621e-01 8.83052886e-01 3.59782428e-01 -1.16212094e+00 -9.76842418e-02 7.78218091e-01 1.24666667e+00 -1.57865986e-01 8.07609558e-02 -5.25698245e-01 6.22495711e-01 -5.61609983e-01 -9.12769914e-01 9.05601025e-01 -6.03898346e-01 6.18196912e-02 -2.64443785e-01 -4.30905342e-01 -3.04132760e-01 7.58052841e-02 1.24828672e+00 3.81154329e-01 8.44119936e-02 4.70189095e-01 -1.33639991e+00 -4.54374820e-01 1.00347233e+00 -1.23660374e+00 9.47387218e-01 -4.42451805e-01 8.22011847e-03 1.02574110e+00 -5.72876036e-01 1.01867348e-01 9.17935252e-01 3.79868120e-01 1.41561985e-01 -7.93138564e-01 -1.09237897e+00 2.41736054e-01 2.08778724e-01 -8.68233800e-01 2.62568463e-02 9.92934346e-01 -1.58533558e-01 8.69596660e-01 4.23748791e-01 7.92084396e-01 4.30091351e-01 8.19928408e-01 2.71003783e-01 1.53007758e+00 -1.69818103e-01 2.35137358e-01 4.20940429e-01 2.35319242e-01 -8.90032947e-02 8.03291917e-01 2.65342951e-01 -2.69696444e-01 -3.08691114e-01 8.24645162e-01 2.09369659e-01 -7.91156814e-02 -1.29842639e-01 -8.99768651e-01 8.78504872e-01 -2.22098529e-01 4.39472646e-01 -3.25693280e-01 -2.49090150e-01 3.00169617e-01 8.15954566e-01 5.68405747e-01 5.74433386e-01 -5.36000729e-01 -6.04345322e-01 -8.60840499e-01 4.77760881e-01 9.23350930e-01 5.66360950e-01 1.64865255e-02 3.41569096e-01 -3.13971122e-03 2.78715044e-01 3.64279300e-01 4.01088536e-01 7.22600877e-01 -8.74238729e-01 6.12771809e-01 5.94098747e-01 5.31267673e-02 -1.11332321e+00 -9.06865001e-02 -8.04184675e-01 -5.52915037e-01 4.15429890e-01 3.49700391e-01 -4.36028719e-01 -1.42328352e-01 1.19092119e+00 1.12525813e-01 1.96300462e-01 1.52313616e-02 6.07846260e-01 -1.30181096e-03 5.01978159e-01 -3.41573417e-01 -7.36802816e-01 8.95674169e-01 -9.80394900e-01 -8.38461339e-01 2.93890297e-01 6.98495656e-02 -9.42702532e-01 5.30357003e-01 7.09928095e-01 -1.21477520e+00 -9.71161649e-02 -9.29209709e-01 8.14173698e-01 -3.57596934e-01 -6.54149473e-01 9.26626384e-01 9.11674261e-01 -8.30498219e-01 9.96705115e-01 -6.82805359e-01 1.36901885e-01 9.73899588e-02 3.24024767e-01 3.53780895e-01 8.02384019e-01 -1.12007725e+00 8.52848172e-01 4.61836040e-01 -1.01127513e-01 -1.98873177e-01 -7.23263800e-01 4.79827495e-03 6.35999292e-02 6.19998217e-01 -3.14994633e-01 1.48285007e+00 -9.01576161e-01 -1.84803689e+00 2.51887172e-01 4.40314740e-01 -1.05585444e+00 8.56761336e-01 -1.68396562e-01 -7.60843337e-01 -2.10383713e-01 -2.57600516e-01 -2.03094959e-01 8.32752585e-01 -5.99071026e-01 -9.40211058e-01 -2.34457776e-01 -1.85746476e-01 2.38237873e-01 -6.10234797e-01 4.38483238e-01 2.63036042e-01 -1.12937975e+00 1.03773288e-01 -6.26077235e-01 -2.71539181e-01 -6.57199204e-01 -8.55926871e-02 5.43472953e-02 7.00748563e-01 -5.77464342e-01 1.78272605e+00 -1.82810104e+00 -2.68476129e-01 4.97319818e-01 -2.73886137e-02 1.63788840e-01 4.81288821e-01 5.43394983e-01 -1.87544301e-01 4.73371893e-01 -1.02083623e-01 2.84422100e-01 1.30779026e-02 -1.35267437e-01 -9.05200422e-01 1.75209060e-01 -1.78793073e-01 8.17018926e-01 -5.11939943e-01 -2.71372914e-01 5.77054806e-02 -7.88503811e-02 -4.25349742e-01 2.28034720e-01 -1.14493214e-01 2.77593046e-01 -6.52538478e-01 8.64535332e-01 7.21788645e-01 -1.65394172e-01 2.51400322e-01 1.61554124e-02 -5.63246012e-01 -1.47973314e-01 -1.52148211e+00 4.54122484e-01 8.77967775e-02 3.29184264e-01 -4.39245194e-01 -8.77718508e-01 1.29458404e+00 2.24621519e-01 6.74430370e-01 -8.16020489e-01 3.67416620e-01 5.70675910e-01 4.52576041e-01 -2.64652759e-01 4.70800668e-01 -4.62598234e-01 1.74314156e-01 7.22220778e-01 -5.12003064e-01 9.75883231e-02 9.40533280e-02 -4.60573614e-01 8.25670004e-01 1.55663952e-01 5.96204221e-01 -3.82998079e-01 4.32343692e-01 -8.06988925e-02 7.40376651e-01 3.42484951e-01 -4.60857749e-01 2.13830009e-01 7.55009949e-01 -2.29652017e-01 -8.09056163e-01 -8.28435004e-01 -4.98681925e-02 6.97604299e-01 8.76194537e-02 -7.77222738e-02 -6.18949294e-01 -3.91001374e-01 3.80480111e-01 7.51478791e-01 -5.21736145e-01 1.84468403e-01 -7.93059170e-01 -9.58330035e-01 1.52348608e-01 4.40471202e-01 8.31017554e-01 -1.08003306e+00 -9.15347874e-01 4.72230494e-01 5.61765432e-01 -3.50461692e-01 -5.35486460e-01 1.03342570e-01 -1.33870029e+00 -9.44518566e-01 -8.26181054e-01 -1.40652537e-01 1.13263123e-01 3.82349640e-01 1.05629838e+00 -2.84966201e-01 8.97048637e-02 -1.18216217e-01 -3.81937087e-01 -9.89214599e-01 1.53938308e-01 -1.50469273e-01 -9.80802812e-04 2.51149505e-01 3.37072015e-01 -6.87407494e-01 -6.95010662e-01 4.50909346e-01 -6.73450530e-01 -3.98638725e-01 7.10171580e-01 4.64316577e-01 7.06463873e-01 3.50844771e-01 8.97264361e-01 -9.14886475e-01 1.15690100e+00 -5.73499680e-01 -1.15451455e+00 5.23481309e-01 -1.31117761e+00 2.19032407e-01 3.28198642e-01 -5.21419168e-01 -1.34492385e+00 -8.03168416e-01 6.10030413e-01 -4.72047508e-01 4.31509078e-01 5.30832767e-01 1.74031146e-02 -1.17218204e-01 -2.44073659e-01 1.68255761e-01 6.33491278e-02 -7.01453924e-01 -7.47765601e-02 3.91938806e-01 1.65468395e-01 -3.30723971e-01 7.28423834e-01 2.97129005e-01 3.80868353e-02 -2.32569173e-01 -5.66107452e-01 -1.66275039e-01 -4.48728539e-02 -1.88084513e-01 5.78933358e-01 -3.28420132e-01 -8.05933297e-01 3.91406924e-01 -5.49283564e-01 1.12683550e-01 -1.93976223e-01 6.55249298e-01 -4.52477813e-01 1.09218225e-01 -3.88840467e-01 -1.21839964e+00 -6.15976751e-01 -7.54408002e-01 6.40519038e-02 5.71905375e-01 -1.40397355e-01 -1.08372295e+00 4.72684115e-01 6.09557033e-02 8.97433043e-01 5.95656574e-01 3.85664016e-01 -1.18135452e+00 -6.84011400e-01 -1.23026542e-01 -9.12646949e-02 2.03135699e-01 1.58685803e-01 2.89966494e-01 -5.23415744e-01 -6.86979070e-02 8.65985513e-01 4.23919439e-01 6.70422852e-01 4.46164846e-01 8.72639477e-01 -2.92524517e-01 -1.69739038e-01 7.36623287e-01 1.46483910e+00 1.06409550e+00 3.48320574e-01 1.02874196e+00 3.81518453e-02 7.11205542e-01 1.17337370e+00 7.80165315e-01 -1.15306579e-01 3.05000246e-01 3.23147774e-01 5.75615108e-01 6.65883660e-01 7.26938248e-04 2.92695552e-01 9.50026453e-01 -2.81369358e-01 -5.13240844e-02 -8.40986371e-01 1.23073515e-02 -1.87644434e+00 -1.26912487e+00 -2.15844009e-02 2.09653044e+00 5.86980343e-01 8.41393113e-01 3.64497364e-01 1.80768684e-01 6.82426870e-01 1.61759824e-01 -6.93984926e-01 -4.55478638e-01 -3.08111370e-01 1.52120546e-01 7.63424933e-01 6.47802353e-02 -8.90029252e-01 3.13803345e-01 6.11213875e+00 8.41767192e-01 -1.30988586e+00 -6.54209182e-02 9.03482080e-01 -3.91677111e-01 -6.41354740e-01 3.18189979e-01 -9.87405121e-01 7.91737556e-01 9.91800427e-01 -1.06665552e+00 1.28394574e-01 9.35852528e-01 2.35927567e-01 4.24742959e-02 -5.92410564e-01 8.72831583e-01 -2.16529638e-01 -1.14599764e+00 -7.67006800e-02 4.41360444e-01 6.67619169e-01 -4.58993107e-01 3.17145675e-01 2.57893682e-01 -5.59685715e-02 -7.81901777e-01 7.45465159e-01 1.03905261e+00 -8.72565657e-02 -1.17759192e+00 9.98832464e-01 1.27629206e-01 -1.09430981e+00 -2.55037069e-01 -3.21238190e-01 -2.38201573e-01 3.62463415e-01 5.47733247e-01 -1.72346458e-01 5.65297186e-01 7.38737226e-01 5.56771040e-01 -5.29788017e-01 1.36578929e+00 2.12891698e-01 7.15041697e-01 -1.45514131e-01 -3.15498352e-01 1.85015321e-01 -8.47710252e-01 5.65974474e-01 6.17097199e-01 6.27112448e-01 3.73651892e-01 -2.15500042e-01 7.18236625e-01 2.69277662e-01 2.99658507e-01 -5.82652271e-01 -3.23983669e-01 4.40557659e-01 1.09757531e+00 -9.67221022e-01 -1.39041394e-01 -4.37818378e-01 2.05585688e-01 -1.87527642e-01 -6.04474060e-02 -1.04509389e+00 -3.34112048e-01 3.53206217e-01 1.45199493e-01 4.88042682e-01 -6.56527877e-02 -7.87242353e-01 -9.90517378e-01 1.66959062e-01 -9.13678348e-01 5.54809093e-01 -1.24469206e-01 -1.42943323e+00 4.93358850e-01 2.95573652e-01 -1.59903812e+00 -1.78060278e-01 -3.96674991e-01 -1.10708511e+00 8.24789405e-01 -1.50439835e+00 -3.56598973e-01 1.72921434e-01 2.32762948e-01 5.26475012e-01 -7.87451625e-01 3.10122848e-01 -5.76326624e-02 -7.86133051e-01 5.86108863e-01 5.25849521e-01 -4.20239270e-01 5.45211375e-01 -1.22311759e+00 3.31974715e-01 6.91448689e-01 -1.12126179e-01 7.98339486e-01 9.86653864e-01 -9.42710400e-01 -1.13673496e+00 -5.30495942e-01 5.89979291e-01 -1.44140258e-01 1.28279841e+00 2.79507995e-01 -1.12156641e+00 4.48528498e-01 6.82783723e-01 -4.67856020e-01 7.66469657e-01 -4.46844846e-01 -9.26333945e-03 -5.23700774e-01 -1.18952954e+00 7.04745412e-01 4.37547117e-01 2.25779250e-01 -6.53604567e-01 -2.99195141e-01 8.45903635e-01 -1.00196548e-01 -9.31885123e-01 5.20079195e-01 6.88226163e-01 -1.39004517e+00 7.68922150e-01 -4.03329879e-01 -7.85817206e-02 9.79033411e-02 -1.48263782e-01 -8.72403800e-01 2.09133979e-02 -1.19852531e+00 -4.60918725e-01 1.56354678e+00 4.22523677e-01 -1.24667013e+00 8.06744754e-01 7.16914773e-01 3.26785207e-01 -1.08483183e+00 -7.26309001e-01 -1.12311280e+00 9.11718905e-02 -3.60030010e-02 1.06661081e+00 8.32580984e-01 -1.72660738e-01 -5.97000942e-02 -3.61821234e-01 -4.63635921e-01 8.86341393e-01 7.17994094e-01 4.78235096e-01 -1.08203709e+00 -7.46248722e-01 -1.10753083e+00 1.12156548e-01 -1.47469610e-01 -3.85297269e-01 -2.91239381e-01 -6.39993072e-01 -8.02136064e-01 1.27050644e-02 -1.60856515e-01 -9.47943747e-01 -8.87993425e-02 -1.50154769e-01 -2.31022388e-01 3.06641519e-01 3.63607436e-01 -2.78741866e-01 3.51250678e-01 1.09666145e+00 1.00238301e-01 -3.65714818e-01 5.03709972e-01 -8.93794239e-01 6.78898692e-01 1.63471889e+00 -4.59767789e-01 -6.13292217e-01 6.95997328e-02 2.73345441e-01 5.11512816e-01 -1.45638600e-01 -7.51060724e-01 1.78466141e-01 -9.59870100e-01 -3.78555842e-02 -8.76302898e-01 -1.40258953e-01 -8.30015540e-01 6.22475266e-01 7.19010234e-01 -1.55114949e-01 9.05595899e-01 1.21192774e-03 7.05986917e-01 -4.28737432e-01 -3.92482132e-01 4.76043612e-01 -2.01625913e-01 -5.97280622e-01 3.52070361e-01 1.81627497e-01 1.64040029e-01 1.58203244e+00 -2.89105117e-01 -2.91369557e-01 -8.07448328e-02 -2.96154857e-01 2.25527421e-01 2.93272197e-01 5.16070008e-01 5.81698537e-01 -1.23053980e+00 -5.76356649e-01 7.86606073e-02 -3.28381389e-01 -6.35154784e-01 1.07395248e-02 8.64537418e-01 -5.59010923e-01 5.82289577e-01 -3.82681251e-01 9.36332494e-02 -1.11132669e+00 5.67292452e-01 4.54958141e-01 -6.80250227e-01 -4.73936141e-01 6.87100172e-01 -1.11393161e-01 2.80684143e-01 1.18794583e-01 -1.49727926e-01 -5.76622248e-01 3.43738079e-01 6.61200464e-01 9.31524158e-01 -1.61829248e-01 -5.43816686e-02 -2.49024644e-01 5.98579884e-01 -1.73063263e-01 -4.32024807e-01 1.53788400e+00 6.34366944e-02 -3.64552081e-01 7.64444113e-01 7.31472373e-01 5.72644509e-02 -1.12121773e+00 1.08615667e-01 7.87542999e-01 -8.16880584e-01 -3.49789292e-01 -4.69149917e-01 -1.15927219e+00 6.05743587e-01 5.63758373e-01 6.61785483e-01 1.05947447e+00 -6.55108035e-01 8.68074298e-01 2.88640290e-01 6.08177423e-01 -1.19854426e+00 6.74678758e-02 3.38772565e-01 9.47671235e-01 -1.06159616e+00 2.59691663e-02 -2.27922928e-02 -8.25747490e-01 1.11032641e+00 7.13506699e-01 -5.48871577e-01 1.34557903e+00 5.79248726e-01 2.34895170e-01 -1.16937876e-01 -9.03496385e-01 2.96101749e-01 9.24626961e-02 2.17830658e-01 4.42901313e-01 1.90702945e-01 -8.31417620e-01 1.08647537e+00 -4.45520222e-01 -2.18971565e-01 3.61634612e-01 1.07038605e+00 -7.38272548e-01 -1.44348073e+00 -4.26629514e-01 9.09805536e-01 -9.30806041e-01 2.04464104e-02 -3.12631190e-01 1.01985180e+00 -2.69504309e-01 6.86703444e-01 2.88744047e-02 -4.06842798e-01 5.47955811e-01 -1.03480227e-01 7.55168423e-02 -2.69958347e-01 -1.16199923e+00 3.08321834e-01 -2.53001481e-01 -4.41941470e-01 -3.32624584e-01 -1.09132898e+00 -1.02736473e+00 -4.12779748e-01 -4.02244240e-01 3.95192266e-01 3.27219576e-01 7.36152828e-01 2.94816643e-02 6.51643991e-01 1.00579810e+00 -2.86968261e-01 -1.32358730e+00 -7.42035985e-01 -1.13994825e+00 -8.35666340e-03 -1.94663242e-01 -8.32787454e-01 -4.49164271e-01 -4.20174539e-01]	[4.700611591339111, 4.03399133682251]
3ef62ba5-e1fe-487a-8ddb-88658da579bb	sg-translate-together-uplifting-singapores	null	null	https://aclanthology.org/2022.amta-upg.28	https://aclanthology.org/2022.amta-upg.28.pdf	SG Translate Together - Uplifting Singapore’s translation standards with the community through technology	The Singapore’s Ministry of Communications and Information (MCI) has officially launched the SG Translate Together (SGTT) web portal on 27 June 2022, with the aim of partnering its citizens to improve translation standards in Singapore. This web portal houses the Singapore Government’s first neural machine translation (MT) engine, known as SG Translate, which was jointly developed by MCI and the Institute for Infocomm Research (I2R), Agency for Science, Technology and Research (A*STAR). Adapted using localised translation data, SG Translate is able to generate translations that are attuned to Singapore’s context and supports Singapore’s four (4) official languages – English (Singapore), Chinese (Singapore), Bahasa Melayu (Singapore) and Tamil (Singapore). Upon completion of development, MCI allowed all Government agencies to use SG Translate for their daily operations. This presentation will briefly cover the methodologies adopted and showcase SG Translate’s capability to translate content involving local culture, everyday life and government policies and schemes. This presentation will also showcase MCI’s sustainable approach for the continual training of the SG Translate MT engine through citizenry participation.	['Nabilah Binte Md Johan', 'Tarun Kumar Vangani', 'Ding Yang', 'Zheng Weihua', 'Wu Kui', 'Aw Ai Ti', 'Sarina Mohamed Rasol', 'Gayathri Ayathorai', 'Foo Yong Xiang', 'Siti Amirah', 'Gowri Kanagarajah', 'Adeline Sim', 'Lee Siew Li']	null	null	null	null	amta-2022-9	['culture']	['speech']	[ 2.42525131e-01 1.85683742e-01 -3.24869096e-01 -2.27554604e-01 -1.46729195e+00 -9.62751269e-01 1.00077748e+00 -1.26045108e-01 -4.97342318e-01 9.31373894e-01 1.02584243e+00 -1.22629833e+00 3.83321941e-01 -6.80850446e-01 -3.55255127e-01 -3.63302648e-01 6.38744771e-01 3.97191286e-01 -6.65668726e-01 -4.12409186e-01 2.08654881e-01 3.35648924e-01 -6.79899454e-01 6.30491376e-01 1.21580029e+00 4.36206460e-01 3.09644222e-01 7.85642028e-01 1.37461200e-01 1.96920961e-01 -3.67947727e-01 -7.71592021e-01 6.97751880e-01 -9.04804945e-01 -9.47650492e-01 -3.87859374e-01 6.51253238e-02 -2.77399868e-01 -1.42854780e-01 8.63385201e-01 8.01601768e-01 -1.68677881e-01 1.43632188e-01 -5.62458694e-01 -1.16816473e+00 5.67021608e-01 1.13369122e-01 2.32623786e-01 4.58515584e-01 3.09154123e-01 9.73083317e-01 -1.25995219e+00 6.70662105e-01 8.46371770e-01 7.35227287e-01 3.42150301e-01 -6.42196894e-01 -6.33235455e-01 -1.41345546e-01 1.24775492e-01 -1.07337797e+00 -7.13124812e-01 3.84286284e-01 -4.29960527e-02 1.54588628e+00 7.95426428e-01 6.92111969e-01 1.26321399e+00 3.06813359e-01 6.51378214e-01 1.71654868e+00 -8.71484637e-01 5.80163933e-02 1.76480845e-01 -5.38661361e-01 1.61614761e-01 -2.98482385e-02 1.67448997e-01 -3.94593447e-01 -1.38075605e-01 4.58201945e-01 -4.61276412e-01 -1.33011723e-02 5.58643281e-01 -1.66049206e+00 9.92160141e-01 6.66466057e-02 6.26729608e-01 -6.44815087e-01 -2.92428404e-01 2.47942820e-01 6.05001271e-01 5.85481524e-01 2.94351667e-01 -7.04618096e-01 -9.44712698e-01 -6.34787202e-01 -1.60172358e-01 9.01694119e-01 1.12603891e+00 1.94245398e-01 1.13542050e-01 1.77885890e-01 9.42171991e-01 3.89764369e-01 8.55098009e-01 3.91780406e-01 -7.85712719e-01 9.43876147e-01 4.41254109e-01 1.92860663e-01 -6.91918850e-01 1.56718064e-02 -6.10718668e-01 -5.93704522e-01 -5.01022220e-01 1.73045516e-01 -6.12924814e-01 -6.67191923e-01 1.06741655e+00 -9.76166204e-02 -5.51674366e-01 3.64439011e-01 9.20740902e-01 6.44366145e-01 9.45846558e-01 -1.96148470e-01 -3.56880039e-01 9.16152954e-01 -9.49804008e-01 -5.81440449e-01 -3.79092425e-01 9.92866814e-01 -1.27339220e+00 9.96322572e-01 1.78122714e-01 -1.37444246e+00 -2.83277065e-01 -9.13039505e-01 2.22677793e-02 -5.36611855e-01 6.43154904e-02 3.94060880e-01 9.81155455e-01 -1.25264454e+00 9.09435600e-02 -9.09549832e-01 -1.05042434e+00 1.10635683e-02 4.52850014e-01 -3.67618114e-01 -3.46761972e-01 -1.43518424e+00 1.44012308e+00 3.03567380e-01 5.09802997e-01 -2.41075248e-01 -4.78531122e-02 -7.14900434e-01 -4.98046130e-01 -4.14608359e-01 -5.29387951e-01 9.78770137e-01 -1.10771561e+00 -1.66758299e+00 8.30434620e-01 -9.66044888e-02 -9.97859910e-02 4.85025167e-01 -6.38389587e-02 -1.11600184e+00 -2.44354829e-01 1.05036281e-01 1.97379649e-01 -9.16901082e-02 -6.08105779e-01 -9.56102550e-01 -2.20222384e-01 -1.12946935e-01 4.23395455e-01 -2.29509383e-01 1.00768483e+00 -1.02666482e-01 -6.46639764e-01 1.44567847e-01 -1.05916452e+00 -4.86976020e-02 -7.95657098e-01 -1.70624152e-01 1.70282349e-01 6.12748444e-01 -1.66204202e+00 1.30888677e+00 -1.63709497e+00 7.60245696e-02 2.05910861e-01 -4.28879619e-01 5.04877508e-01 -3.51645291e-01 1.28412068e+00 2.58800983e-01 1.52697980e-01 -3.22778136e-01 -7.31979534e-02 1.04993479e-02 5.76774716e-01 -8.34149271e-02 5.29102445e-01 7.20203519e-02 1.43660223e+00 -1.03795421e+00 1.58786967e-01 5.42041287e-02 3.79012257e-01 -7.23978281e-02 -3.70381683e-01 5.52990556e-01 6.23881876e-01 -2.71627128e-01 9.89745021e-01 5.61474741e-01 1.58083498e-01 4.78761017e-01 3.97233456e-01 -6.38073146e-01 8.33208382e-01 -5.00418007e-01 1.47249115e+00 -5.36426723e-01 7.90436804e-01 2.11361408e-01 -5.48645020e-01 1.09943223e+00 5.98414183e-01 1.00733049e-01 -1.14918661e+00 -1.49431631e-01 1.15996301e+00 4.03953280e-04 -3.43922496e-01 5.76718748e-01 -1.90115854e-01 -9.14411023e-02 7.90755630e-01 -3.41678947e-01 -9.58491340e-02 -2.53679723e-01 -1.75412640e-01 8.58033657e-01 2.85143495e-01 2.08779395e-01 -3.52277517e-01 4.92769420e-01 2.46681720e-01 5.22812366e-01 3.12378168e-01 -3.75852972e-01 4.17847544e-01 -3.68540078e-01 -5.79287469e-01 -1.50849581e+00 -7.75299966e-01 8.29895586e-02 7.71474779e-01 -6.39332891e-01 -2.13770792e-01 -6.86558366e-01 -8.13350439e-01 -6.44197643e-01 1.14831495e+00 -1.05479084e-01 1.68040007e-01 -9.97583389e-01 -6.07779860e-01 6.17066383e-01 1.62653178e-01 7.60307968e-01 -1.05140162e+00 -3.23289722e-01 4.29885119e-01 -8.71715188e-01 -1.07093501e+00 -7.87000477e-01 -2.73100566e-03 -6.80974066e-01 -4.20224458e-01 -8.98833632e-01 -9.92546022e-01 5.28032660e-01 2.09810138e-01 5.99647105e-01 -3.54047805e-01 4.26706463e-01 2.53045738e-01 -5.80394566e-01 -3.05523396e-01 -1.00667644e+00 3.80510956e-01 7.90435299e-02 -3.60369742e-01 6.39718235e-01 -5.17387033e-01 -5.53356588e-01 5.02982855e-01 -6.32224798e-01 3.95077527e-01 7.68153906e-01 5.74646235e-01 -1.09535279e-02 -4.25203264e-01 7.90835023e-01 -3.74482155e-01 1.03221262e+00 -2.84420788e-01 -2.62762010e-01 3.70655388e-01 -7.82442570e-01 -6.24466538e-01 4.45780545e-01 7.09027350e-02 -9.69693124e-01 -2.98526615e-01 -2.62700498e-01 6.09160602e-01 6.77728355e-02 8.99698079e-01 -2.79677093e-01 -2.44066641e-01 4.77092505e-01 7.99041092e-01 -9.57889184e-02 -2.83450931e-01 3.24338257e-01 1.55073762e+00 4.49102551e-01 -3.58556956e-01 9.30081606e-01 5.60731627e-03 -4.99800831e-01 -5.89528263e-01 -2.72544533e-01 -3.23771060e-01 -8.51610720e-01 -3.28253597e-01 7.91007638e-01 -1.08478606e+00 -8.28351602e-02 5.11141598e-01 -1.27724409e+00 -3.30859542e-01 1.18341573e-01 9.02005672e-01 -4.79099005e-01 1.32953972e-01 -8.72693717e-01 -4.65117902e-01 -7.04280436e-01 -1.07765794e+00 7.35322416e-01 4.77517620e-02 -4.02702212e-01 -1.13693810e+00 9.30713639e-02 8.64226103e-01 7.97190666e-01 4.59947079e-01 9.49043036e-01 -7.24324942e-01 -1.95011884e-01 -2.74524957e-01 -2.65778691e-01 8.26814830e-01 5.15549719e-01 -2.70044297e-01 -3.37781221e-01 -3.79068166e-01 1.12516321e-01 3.08841411e-02 -2.99316309e-02 -1.54450655e-01 -3.61365914e-01 -1.12436080e+00 1.07556969e-01 1.79441467e-01 1.18561196e+00 5.53167939e-01 6.90796614e-01 9.40927446e-01 3.10156077e-01 4.54867989e-01 5.58062136e-01 2.75064111e-02 6.21586382e-01 7.96486497e-01 -1.07538767e-01 -3.17834079e-01 -1.77139193e-01 -2.95365900e-01 9.41380680e-01 1.98236978e+00 -6.12196505e-01 2.54595429e-01 -1.01998484e+00 4.74848449e-01 -1.70803821e+00 -6.80194259e-01 -3.97399455e-01 2.30168557e+00 8.31630230e-01 -1.27360821e-01 5.33671230e-02 -3.32442343e-01 4.54825610e-01 -2.27994040e-01 3.32878716e-02 -1.33016586e+00 -4.21019047e-01 3.42114329e-01 7.26956964e-01 5.84834516e-01 -3.23858559e-01 1.03649223e+00 6.45644283e+00 5.12879252e-01 -1.05907178e+00 4.04856712e-01 7.16946661e-01 -1.40872858e-02 -6.06401742e-01 1.82937741e-01 -5.30506670e-01 3.60763371e-01 1.53088081e+00 -7.03262031e-01 9.72917974e-01 2.55506247e-01 8.68407369e-01 4.62841481e-01 -3.70720983e-01 3.46970618e-01 1.05814777e-01 -1.60024738e+00 -7.48517662e-02 7.15019107e-01 1.20606816e+00 6.57846034e-01 -7.87453027e-04 3.65603000e-01 2.93833494e-01 -9.28532064e-01 8.00333321e-01 2.77223915e-01 9.37373519e-01 -1.08802831e+00 9.43634748e-01 4.60270405e-01 -6.42918766e-01 -1.43034175e-01 -1.69079036e-01 -3.05379272e-01 3.39208454e-01 2.08276808e-01 -9.49403763e-01 8.23645115e-01 3.24874043e-01 2.65043676e-01 -3.17892879e-01 3.99522871e-01 -4.22614634e-01 9.96102333e-01 -1.47614464e-01 -7.79967532e-02 8.83200884e-01 -7.32356668e-01 6.67140961e-01 1.49740279e+00 7.09453762e-01 3.23362678e-01 -1.69124715e-02 2.42890194e-01 1.43843994e-01 3.11140537e-01 -4.08170581e-01 -3.93113256e-01 4.19401109e-01 8.40943873e-01 -4.36244160e-01 -1.90567896e-01 -7.16841578e-01 1.32537103e+00 -2.51929730e-01 5.42079091e-01 -5.84939718e-01 -3.88329595e-01 6.52500808e-01 -1.44084349e-01 -1.94678102e-02 -8.38246047e-01 -4.20542687e-01 -8.54293644e-01 6.56961724e-02 -1.70944655e+00 -2.79188771e-02 -7.86820054e-01 -7.00665653e-01 8.22124839e-01 -2.18917266e-01 -1.00462747e+00 -3.76930535e-01 -4.21401531e-01 -3.75584841e-01 1.56942713e+00 -9.88390446e-01 -1.93618941e+00 5.71746171e-01 3.23252708e-01 7.20473528e-01 -2.41676882e-01 1.14371991e+00 1.59359515e-01 -2.92923033e-01 6.31736517e-01 4.59243834e-01 9.69328135e-02 3.04602325e-01 -7.62854218e-01 1.15668511e+00 9.16380465e-01 1.36151642e-01 9.36591327e-01 6.66400433e-01 -7.65716195e-01 -1.70009851e+00 -1.06750667e+00 2.20824027e+00 -5.69604874e-01 8.86028826e-01 -5.23038507e-01 -5.93223050e-02 9.31293905e-01 6.54134870e-01 -8.19405973e-01 9.42880869e-01 -4.04601455e-01 5.96675649e-03 2.53375679e-01 -1.12314284e+00 8.82502794e-01 9.69886839e-01 -8.59025240e-01 -4.89484102e-01 7.64611959e-01 7.52327323e-01 -3.37888002e-01 -9.95703459e-01 9.80859101e-02 7.62305498e-01 -5.53500652e-01 3.95774752e-01 -6.84467554e-01 4.10519302e-01 -2.77076423e-01 -6.22366786e-01 -1.37471282e+00 -3.87266040e-01 -1.21211767e+00 1.01464324e-01 1.01055586e+00 8.52589726e-01 -1.29849625e+00 4.45349574e-01 5.83550870e-01 -5.84401190e-01 -7.14175880e-01 -1.31599057e+00 -9.02909279e-01 3.45158190e-01 -6.01247370e-01 7.50281811e-01 1.18647814e+00 1.02234580e-01 1.04351297e-01 -3.18842798e-01 7.71464258e-02 6.23348542e-03 -2.85786033e-01 7.30052590e-01 -2.15781391e-01 -2.75037348e-01 -5.63028574e-01 -2.97096014e-01 -8.88232470e-01 -4.16829228e-01 -1.08134806e+00 -3.99300694e-01 -1.98691928e+00 -7.36238286e-02 1.03160078e-02 -6.75123334e-02 5.01889527e-01 1.18788980e-01 3.95621091e-01 3.43938202e-01 4.91370112e-01 3.77927572e-01 2.29188219e-01 1.45355701e+00 -8.89815316e-02 -1.34863570e-01 2.16339350e-01 -1.21856070e+00 -1.24996692e-01 9.80903864e-01 -1.58944383e-01 2.09357426e-01 -7.82777250e-01 4.01195973e-01 -1.71557114e-01 -9.85511020e-02 -5.01860678e-01 2.19992936e-01 -4.62827146e-01 1.99950948e-01 -4.48671967e-01 -4.83893156e-02 -7.01299250e-01 6.79319501e-01 5.99265933e-01 -2.42366180e-01 7.27034092e-01 2.47611851e-01 -3.06097001e-01 -2.15084150e-01 2.71059513e-01 3.28190327e-01 -2.25548699e-01 -1.08983912e-01 -1.42407879e-01 -5.83625734e-01 -5.58423281e-01 7.09705651e-01 -6.78015351e-01 -4.74937588e-01 -7.44699955e-01 -4.24744904e-01 1.03221498e-01 5.71237206e-01 6.00442410e-01 4.09796298e-01 -1.49818087e+00 -1.42337239e+00 2.42632478e-01 3.09483185e-02 -8.17255318e-01 -3.17816466e-01 1.13954997e+00 -7.32921064e-01 8.03805649e-01 3.87161635e-02 -6.35284260e-02 -7.85898566e-01 2.42435690e-02 2.09458973e-02 -3.61207187e-01 -6.01334870e-01 3.98929417e-01 -5.91907561e-01 -1.15123963e+00 -5.47473431e-01 -8.32266137e-02 2.61592567e-01 -4.29725200e-01 4.80579048e-01 4.05208617e-01 2.77269363e-01 -1.12063968e+00 -9.96723250e-02 2.00727224e-01 -2.46542729e-02 -7.05915928e-01 1.38174915e+00 -3.63918960e-01 -2.76806802e-01 2.63471067e-01 1.28317404e+00 3.43641013e-01 -5.91264188e-01 1.10287257e-01 2.75311619e-01 -2.84138829e-01 -1.55847564e-01 -1.58340263e+00 -6.24497533e-01 3.62741947e-01 3.35426390e-01 -2.67819554e-01 1.15337551e+00 -2.81520426e-01 1.25655746e+00 7.44423494e-02 6.30538106e-01 -1.32034719e+00 -8.83802474e-01 7.45707572e-01 1.08719051e+00 -6.05018318e-01 -2.58319229e-01 2.30486542e-02 -5.40507853e-01 1.26729548e+00 -3.93636167e-01 4.58159536e-01 1.47155881e-01 -5.77927567e-02 7.32905269e-01 3.45875472e-01 -5.54522634e-01 2.45589688e-01 2.70722747e-01 9.43603575e-01 7.10394919e-01 6.06369317e-01 -1.04099846e+00 1.45474404e-01 -7.19187319e-01 5.60646765e-02 3.52860212e-01 1.09105909e+00 -2.42850333e-01 -1.65929925e+00 -3.38431239e-01 3.19712400e-01 -3.64314198e-01 -5.97653031e-01 -9.30789351e-01 8.56407225e-01 2.06296146e-02 1.13397717e+00 -3.49313289e-01 -4.54501957e-01 4.14780140e-01 -5.88519946e-02 3.04887891e-01 -3.55078757e-01 -1.19034970e+00 1.55950770e-01 8.70523274e-01 -3.05520058e-01 -9.95491818e-02 -1.00893128e+00 -6.65882945e-01 -6.51852965e-01 -8.05469528e-02 4.88557369e-01 1.41637850e+00 1.02676487e+00 4.20087367e-01 -2.70774990e-01 9.30111587e-01 -3.82250518e-01 -4.53693092e-01 -1.06325626e+00 -6.48127720e-02 -2.60088682e-01 -2.53766894e-01 5.89723289e-01 -1.56075850e-01 -1.07777588e-01]	[11.485623359680176, 10.419071197509766]
5ec57a14-949e-4617-bc96-9e3d48054a43	residual-dense-network-for-image-super	1802.08797	null	http://arxiv.org/abs/1802.08797v2	http://arxiv.org/pdf/1802.08797v2.pdf	Residual Dense Network for Image Super-Resolution	A very deep convolutional neural network (CNN) has recently achieved great success for image super-resolution (SR) and offered hierarchical features as well. However, most deep CNN based SR models do not make full use of the hierarchical features from the original low-resolution (LR) images, thereby achieving relatively-low performance. In this paper, we propose a novel residual dense network (RDN) to address this problem in image SR. We fully exploit the hierarchical features from all the convolutional layers. Specifically, we propose residual dense block (RDB) to extract abundant local features via dense connected convolutional layers. RDB further allows direct connections from the state of preceding RDB to all the layers of current RDB, leading to a contiguous memory (CM) mechanism. Local feature fusion in RDB is then used to adaptively learn more effective features from preceding and current local features and stabilizes the training of wider network. After fully obtaining dense local features, we use global feature fusion to jointly and adaptively learn global hierarchical features in a holistic way. Extensive experiments on benchmark datasets with different degradation models show that our RDN achieves favorable performance against state-of-the-art methods.	['Yulun Zhang', 'Yu Kong', 'Yapeng Tian', 'Bineng Zhong', 'Yun Fu']	2018-02-24	residual-dense-network-for-image-super-1	http://openaccess.thecvf.com/content_cvpr_2018/html/Zhang_Residual_Dense_Network_CVPR_2018_paper.html	http://openaccess.thecvf.com/content_cvpr_2018/papers/Zhang_Residual_Dense_Network_CVPR_2018_paper.pdf	cvpr-2018-6	['color-image-denoising']	['computer-vision']	[ 1.50672391e-01 -1.71705544e-01 -3.02283943e-01 -3.14555049e-01 -7.08581448e-01 2.25219011e-01 3.00587535e-01 -3.50731283e-01 -1.46173224e-01 7.08957434e-01 5.91958106e-01 2.47176394e-01 -8.44146609e-02 -9.69148040e-01 -7.06413031e-01 -8.41368556e-01 -3.72182727e-02 -3.71740252e-01 5.98341823e-01 -3.40232909e-01 1.27855539e-01 7.86900043e-01 -1.64358270e+00 6.24771297e-01 7.56952822e-01 1.21797955e+00 5.31244338e-01 3.16614151e-01 5.80386585e-03 1.15319026e+00 -3.88117760e-01 3.79902542e-01 2.25853935e-01 -2.93825448e-01 -8.37576330e-01 4.81237695e-02 5.19113898e-01 -7.75819302e-01 -8.20854008e-01 9.55713153e-01 6.19352043e-01 3.93558830e-01 3.32862921e-02 -4.52578306e-01 -8.92337322e-01 6.63258076e-01 -8.23854387e-01 7.89958119e-01 2.71200873e-02 1.30966261e-01 7.95569599e-01 -1.07850730e+00 5.42188287e-01 1.41371882e+00 4.14556026e-01 4.92616802e-01 -1.14944756e+00 -8.24623525e-01 1.83307126e-01 3.36271495e-01 -1.48443520e+00 -5.15999258e-01 8.02737713e-01 3.00823357e-02 8.39977145e-01 2.32058223e-02 3.24631095e-01 9.00613368e-01 1.37236252e-01 5.30251920e-01 1.19283855e+00 -6.43235594e-02 -1.58656798e-02 -1.86521038e-01 2.35480487e-01 6.23173237e-01 -4.07538339e-02 1.68468058e-01 -8.51778746e-01 2.20859379e-01 1.36235309e+00 1.76524669e-01 -5.71192861e-01 4.71572243e-02 -1.00475216e+00 7.14465618e-01 1.21720624e+00 6.46296203e-01 -5.72320223e-01 1.46460682e-01 1.85747385e-01 2.74459153e-01 3.57754588e-01 7.29506761e-02 -3.09646249e-01 3.95438164e-01 -7.28318751e-01 -8.64886194e-02 -3.60242054e-02 6.06245399e-01 1.08936620e+00 2.19902650e-01 -4.63000417e-01 1.14383090e+00 -1.07673720e-01 -4.12788466e-02 4.98514205e-01 -1.17495966e+00 3.67104411e-01 6.90131247e-01 -3.28592286e-02 -1.18980432e+00 -2.86687374e-01 -7.79604197e-01 -1.56855321e+00 2.26215690e-01 -1.64437041e-01 3.76535624e-01 -9.88176227e-01 1.44264543e+00 2.21415251e-01 3.12499553e-01 1.78282320e-01 1.13081038e+00 1.13846624e+00 8.80161345e-01 -1.10793740e-01 -2.37972051e-01 1.13514006e+00 -9.99312639e-01 -5.49342394e-01 -1.10608704e-01 1.38385057e-01 -4.39541638e-01 9.94759083e-01 3.41580838e-01 -1.17166305e+00 -1.22675884e+00 -1.15315974e+00 -5.50758123e-01 -2.17049923e-02 1.69134751e-01 4.09779638e-01 -6.78690597e-02 -1.35056126e+00 7.96171606e-01 -6.94382429e-01 1.03311539e-01 8.43988180e-01 4.23547715e-01 -3.42969358e-01 -4.55496371e-01 -1.27086735e+00 4.95197713e-01 3.75227392e-01 4.43666995e-01 -1.09082472e+00 -6.12763226e-01 -9.74382758e-01 1.38068959e-01 1.96071953e-01 -5.02425790e-01 7.44501173e-01 -8.50585341e-01 -1.40705740e+00 4.12834108e-01 -2.34083369e-01 -5.07554233e-01 3.44139896e-02 -2.79167414e-01 -4.17375594e-01 4.62053925e-01 -2.50010118e-02 7.50405729e-01 1.09989977e+00 -1.31360817e+00 -6.32258117e-01 -4.26578850e-01 8.02455023e-02 1.98165074e-01 -5.67704320e-01 9.54333544e-02 -3.51705253e-01 -6.67332649e-01 4.29260969e-01 -3.21052819e-01 -3.68128151e-01 -7.47475773e-02 -1.50528729e-01 -2.43331775e-01 9.36191022e-01 -7.24851370e-01 1.22767901e+00 -2.32489538e+00 2.70506114e-01 -1.61792487e-01 5.44330478e-01 3.52244884e-01 -2.50809103e-01 -2.07580373e-01 -1.68559104e-01 -2.88043749e-02 -1.10387780e-01 -3.58193666e-01 -6.80851161e-01 2.72192478e-01 -3.38342577e-01 3.15105170e-01 4.30898935e-01 8.92316282e-01 -6.29181564e-01 -4.87974018e-01 2.78133243e-01 9.12421703e-01 -2.95832038e-01 2.43997827e-01 1.30340517e-01 8.02584529e-01 -5.08776546e-01 4.31328505e-01 8.99069190e-01 -5.08055985e-01 -1.29870921e-01 -6.65451646e-01 -3.08927566e-01 2.24936411e-01 -1.06209207e+00 1.70764661e+00 -6.90196812e-01 5.21957099e-01 -5.65751754e-02 -9.54300165e-01 1.27302814e+00 -1.51634201e-01 3.07342321e-01 -1.12288380e+00 -8.42719246e-03 6.04968816e-02 -2.90041715e-01 -1.80233300e-01 5.36723971e-01 2.68081613e-02 2.66516238e-01 1.78451240e-02 5.30853346e-02 5.10255158e-01 -3.25474292e-01 2.12125421e-01 9.66982961e-01 -5.63799739e-02 1.97726905e-01 -1.10512473e-01 9.28401649e-01 -4.29534793e-01 7.66079903e-01 5.92397153e-01 -5.20431548e-02 8.64363194e-01 1.13524094e-01 -7.17508852e-01 -1.03180325e+00 -9.25151765e-01 -2.59150177e-01 9.04241741e-01 3.89807254e-01 -2.87584722e-01 -3.17427665e-01 -3.71432096e-01 -3.86031508e-01 1.81379497e-01 -7.84949422e-01 -3.12231272e-01 -9.17621315e-01 -6.79998517e-01 1.77726418e-01 6.86552346e-01 1.22485721e+00 -1.30077708e+00 -4.79003400e-01 2.97311723e-01 -2.41406247e-01 -1.28348577e+00 -2.14450598e-01 -1.53598655e-02 -1.09462941e+00 -5.81010103e-01 -6.99666440e-01 -7.65294790e-01 4.27236319e-01 7.22456992e-01 8.79107594e-01 3.32214892e-01 -4.15836781e-01 -2.72276133e-01 -3.74914110e-01 4.77034301e-01 -2.53376603e-01 -1.84224602e-02 -1.48720741e-01 2.55201101e-01 7.79929087e-02 -7.62773693e-01 -9.20124114e-01 2.90499508e-01 -9.66024697e-01 1.80914834e-01 8.61151338e-01 9.07131672e-01 8.79473925e-01 4.90724236e-01 6.12657726e-01 -5.18457055e-01 8.79440010e-02 -3.32411170e-01 -2.95924723e-01 -1.49805754e-01 -3.29168081e-01 1.03389286e-01 6.68933272e-01 -2.76488572e-01 -1.24416983e+00 -2.82216728e-01 -1.70454562e-01 -6.19669616e-01 -1.85629621e-01 1.82269856e-01 -3.84251684e-01 -1.95777953e-01 4.89440203e-01 5.59925914e-01 -2.02518821e-01 -8.27369034e-01 1.80420995e-01 7.18511939e-01 6.84207618e-01 -3.41579497e-01 7.78011143e-01 5.78457177e-01 -3.87736820e-02 -7.07108736e-01 -1.18553841e+00 -2.58787781e-01 -7.07016110e-01 3.43052372e-02 9.50160265e-01 -1.38079441e+00 -4.08386022e-01 7.26690471e-01 -8.68751824e-01 -3.65794808e-01 -3.96307230e-01 1.73945189e-01 -3.57621729e-01 3.72128516e-01 -9.30935979e-01 -3.99799734e-01 -4.91984099e-01 -1.16150832e+00 1.10160625e+00 6.15464926e-01 4.47863936e-01 -5.14829099e-01 -2.79841393e-01 3.96403611e-01 7.82417119e-01 3.08177888e-01 6.76397085e-01 5.80098033e-02 -8.72425556e-01 2.21841648e-01 -8.54778171e-01 7.63520360e-01 3.46285522e-01 -3.69771659e-01 -9.14017737e-01 -6.10446215e-01 8.99592042e-02 -3.65800947e-01 1.47756219e+00 3.30693513e-01 1.57712412e+00 -3.73758167e-01 -1.07707351e-01 8.56680870e-01 1.74456334e+00 -3.93369913e-01 9.75233316e-01 5.05684137e-01 1.01857448e+00 3.71874005e-01 6.81231737e-01 3.73754531e-01 3.73634070e-01 7.55317032e-01 5.38609624e-01 -3.85461092e-01 -6.22548759e-01 6.32359013e-02 5.71942389e-01 6.44325018e-01 -2.71424323e-01 3.93299788e-01 -4.35011983e-01 5.66946149e-01 -1.62172079e+00 -8.57110083e-01 1.69147179e-02 2.01735139e+00 1.04385400e+00 2.05006108e-01 -1.86426625e-01 4.33693938e-02 7.81282604e-01 6.21191144e-01 -5.66382885e-01 -7.75551051e-02 -4.00714964e-01 4.04283226e-01 2.77716339e-01 2.47657120e-01 -9.36002195e-01 1.00210857e+00 5.34259176e+00 1.10607183e+00 -1.13560975e+00 2.93516487e-01 8.88317049e-01 -5.78913689e-02 -1.16188020e-01 -1.55519381e-01 -1.06325376e+00 3.29906076e-01 7.61380851e-01 3.06183487e-01 3.98335695e-01 5.90487599e-01 2.14285448e-01 5.71382651e-03 -5.91147363e-01 9.05554414e-01 -7.35819489e-02 -1.42439246e+00 4.11654830e-01 5.08115888e-02 9.84646678e-01 1.62811413e-01 2.85636604e-01 2.42508739e-01 2.59827107e-01 -1.14261663e+00 3.36698294e-01 5.94141841e-01 1.04047310e+00 -1.09900141e+00 6.79570079e-01 1.17342792e-01 -1.43305039e+00 -6.01799190e-01 -9.16050255e-01 1.90026149e-01 -1.33233130e-01 8.22096348e-01 -4.21451069e-02 7.93660045e-01 1.22565329e+00 1.10013306e+00 -7.54098535e-01 5.05402148e-01 -1.52554080e-01 3.20532680e-01 1.45699039e-01 8.45745564e-01 2.52120405e-01 1.95627823e-01 4.46067154e-01 9.40990984e-01 9.51209441e-02 3.80048990e-01 1.37901604e-01 8.38573933e-01 -2.06981853e-01 -1.88780203e-01 -2.28514522e-01 4.29285079e-01 3.19529831e-01 1.43055904e+00 -3.58110309e-01 -4.47269320e-01 -5.05259097e-01 1.08656073e+00 6.47869289e-01 3.61776799e-01 -4.58723992e-01 -1.34228408e-01 7.84426451e-01 2.11789966e-01 6.37949526e-01 -3.28606665e-01 -4.98257019e-02 -1.19917083e+00 -9.26045552e-02 -9.30603504e-01 4.57396835e-01 -8.31611872e-01 -1.10377824e+00 9.47474718e-01 -3.33134681e-01 -1.28572416e+00 2.34481096e-01 -1.05184183e-01 -4.13629115e-01 1.05204856e+00 -2.13185263e+00 -1.17638338e+00 -8.65420461e-01 1.02397978e+00 8.16458166e-01 -9.76240486e-02 4.30027246e-01 1.97331861e-01 -7.65394270e-01 4.39435333e-01 -4.48048040e-02 1.38866752e-01 6.49996519e-01 -1.00957847e+00 2.35378221e-01 1.03300822e+00 -2.27832273e-01 6.63700521e-01 3.23128879e-01 -5.87327123e-01 -1.18161464e+00 -1.41505766e+00 3.06918323e-01 7.39763975e-02 4.70764965e-01 -1.09395042e-01 -1.33891690e+00 3.35758179e-01 -3.08754272e-03 5.69382906e-01 1.32331714e-01 -1.65424481e-01 -5.90916514e-01 -6.18871152e-01 -1.02374232e+00 2.57842958e-01 1.12420368e+00 -5.65263033e-01 -3.75258178e-01 -6.39551952e-02 1.20215881e+00 -2.59216964e-01 -1.13216436e+00 7.19438910e-01 2.31290817e-01 -1.13951778e+00 1.27108562e+00 -2.16923371e-01 7.30492413e-01 -4.93692875e-01 -3.24415058e-01 -9.34699237e-01 -7.64981806e-01 -2.98542619e-01 -4.29951996e-01 1.16993725e+00 -1.35693401e-01 -4.68524724e-01 3.99265409e-01 1.93112120e-02 -2.51648366e-01 -8.85575235e-01 -7.97333717e-01 -6.61985993e-01 -1.44523263e-01 8.58681574e-02 6.36453152e-01 7.60249019e-01 -7.36347735e-01 2.61157691e-01 -5.14693320e-01 5.34384310e-01 8.93696308e-01 3.91786277e-01 4.72494543e-01 -1.04424691e+00 -2.00204492e-01 -2.40672603e-01 -4.18825239e-01 -1.24455369e+00 3.28560844e-02 -6.61570013e-01 -1.89569313e-02 -1.52841640e+00 6.38282776e-01 -5.29531598e-01 -7.63718724e-01 5.54963052e-01 -3.61552715e-01 8.22363019e-01 2.26866882e-02 4.84827340e-01 -7.20680773e-01 7.07754850e-01 1.60778809e+00 6.66455412e-03 -3.13845217e-01 -3.30798447e-01 -8.31562996e-01 4.96746182e-01 7.78388023e-01 -2.08278790e-01 -1.48695886e-01 -4.28651690e-01 -1.01193249e-01 3.42268273e-02 6.62843347e-01 -1.17210650e+00 1.48266941e-01 4.88178059e-02 9.23277795e-01 -7.05331564e-01 2.74305969e-01 -5.17781019e-01 -6.88702241e-02 2.31174216e-01 -3.14901769e-01 -4.91792589e-01 1.15231641e-01 5.56405842e-01 -4.40289259e-01 2.46683761e-01 1.30280328e+00 -2.41925210e-01 -1.02008712e+00 6.16465807e-01 4.78560068e-02 -4.35628235e-01 5.86319506e-01 -1.84399396e-01 -3.58855575e-01 -4.23681326e-02 -8.95042360e-01 1.84139505e-01 5.00851989e-01 5.72171748e-01 1.06757462e+00 -1.34722567e+00 -9.27180767e-01 3.04639578e-01 -1.28791034e-01 4.30028230e-01 8.62922847e-01 7.87912130e-01 -1.32895634e-01 1.96068004e-01 -4.85802591e-01 -6.26125932e-01 -1.18382895e+00 7.62069106e-01 3.96977603e-01 -4.24661130e-01 -1.14458656e+00 7.81265140e-01 5.33393562e-01 1.08150877e-01 8.53303522e-02 -7.78163671e-02 -7.07014501e-01 -1.12669475e-01 1.09040523e+00 3.99175197e-01 -8.17462355e-02 -8.35038245e-01 -1.61105007e-01 7.36285090e-01 -5.11456251e-01 2.35234916e-01 1.73872280e+00 -5.77023923e-01 -4.10557240e-01 1.54434457e-01 1.25452828e+00 -2.87507683e-01 -1.58856761e+00 -7.62761414e-01 -3.54538113e-01 -6.86889946e-01 5.35905898e-01 -4.52021539e-01 -1.59293365e+00 8.68655384e-01 7.11369276e-01 -1.93360791e-01 1.73202002e+00 1.52610019e-01 1.08858860e+00 5.64069785e-02 4.30285275e-01 -7.36930072e-01 5.39552152e-01 2.82027483e-01 1.04235530e+00 -1.11644006e+00 1.32762700e-01 -3.14085931e-01 -4.15602773e-01 1.14184678e+00 8.81558836e-01 -3.84165049e-01 4.89088207e-01 1.04007691e-01 -2.68650532e-01 -1.15234889e-01 -7.05077231e-01 -2.69121319e-01 2.49491751e-01 4.16714609e-01 4.44505811e-02 -2.18887076e-01 -4.30038609e-02 6.62444293e-01 2.43627399e-01 1.10522807e-01 5.71105719e-01 6.32106066e-01 -8.41246665e-01 -9.17404056e-01 -1.82691172e-01 2.86096275e-01 -4.85294223e-01 -2.14956626e-01 1.12216212e-01 4.03530151e-01 1.96198672e-01 8.23640704e-01 7.60349482e-02 -4.87706065e-01 9.91728157e-02 -6.18690372e-01 3.51332486e-01 -5.59875369e-01 -3.74272853e-01 7.67204165e-02 -2.98581064e-01 -1.06748116e+00 -5.85829496e-01 -4.36719626e-01 -1.30090034e+00 -1.95190892e-01 -1.55658916e-01 -1.65018484e-01 2.83415228e-01 6.97974205e-01 4.26221609e-01 7.84189582e-01 1.10176563e+00 -1.04183948e+00 -3.65252197e-01 -1.05177248e+00 -5.43504596e-01 -2.88373325e-02 7.93510199e-01 -5.48925817e-01 -3.38485867e-01 -1.29871741e-01]	[11.00069808959961, -1.989983081817627]
a2d1d375-8e76-4f6c-8c28-31a2adff3b27	a-unified-framework-of-predicting-binary	1910.05996	null	https://arxiv.org/abs/1910.05996v1	https://arxiv.org/pdf/1910.05996v1.pdf	A unified framework of predicting binary interestingness of images based on discriminant correlation analysis and multiple kernel learning	In the modern content-based image retrieval systems, there is an increasingly interest in constructing a computationally effective model to predict the interestingness of images since the measure of image interestingness could improve the human-centered search satisfaction and the user experience in different applications. In this paper, we propose a unified framework to predict the binary interestingness of images based on discriminant correlation analysis (DCA) and multiple kernel learning (MKL) techniques. More specially, on the one hand, to reduce feature redundancy in describing the interestingness cues of images, the DCA or multi-set discriminant correlation analysis (MDCA) technique is adopted to fuse multiple feature sets of the same type for individual cues by taking into account the class structure among the samples involved to describe the three classical interestingness cues, unusualness,aesthetics as well as general preferences, with three sets of compact and representative features; on the other hand, to make good use of the heterogeneity from the three sets of high-level features for describing the interestingness cues, the SimpleMKL method is employed to enhance the generalization ability of the built model for the task of the binary interestingness classification. Experimental results on the publicly-released interestingness prediction data set have demonstrated the rationality and effectiveness of the proposed framework in the binary prediction of image interestingness where we have conducted several groups of comparative studies across different interestingness feature combinations, different interestingness cues, as well as different feature types for the three interestingness cues.	['Yuxiang Yang', 'Maohui Li', 'Liting Wang', 'Longtao Zhang', 'Qiang Sun']	2019-10-14	null	null	null	null	['content-based-image-retrieval']	['computer-vision']	[ 5.83192594e-02 -3.30852330e-01 -4.89309192e-01 -4.12973255e-01 -6.39726043e-01 -2.47060269e-01 5.05728424e-01 5.73733807e-01 -3.79577786e-01 2.89738744e-01 1.59617841e-01 1.69403642e-01 -9.36294079e-01 -4.69517857e-01 -1.40780494e-01 -9.71831143e-01 -1.92237601e-01 -4.11880054e-02 4.73165214e-02 -2.25081947e-02 6.06718540e-01 6.47020102e-01 -2.14845228e+00 3.61612350e-01 1.17692089e+00 1.65755296e+00 3.70991319e-01 1.45340949e-01 1.46402093e-02 5.40039897e-01 2.60143466e-02 -1.32811114e-01 3.69732618e-01 -2.39602983e-01 -5.76975644e-01 2.55463272e-01 1.40892744e-01 6.30746409e-02 2.25385472e-01 8.19467783e-01 1.89908400e-01 3.03333580e-01 8.18178356e-01 -1.22664201e+00 -5.25309324e-01 -1.10413611e-01 -5.81083775e-01 2.47353345e-01 3.68248492e-01 4.10142392e-02 1.21699822e+00 -9.51040566e-01 5.84438145e-01 1.07564938e+00 5.26044965e-02 -1.79436013e-01 -7.78463960e-01 -5.12258232e-01 1.15068018e-01 8.29756200e-01 -1.43632615e+00 -1.94213465e-01 9.65752602e-01 -4.20859456e-01 3.21724832e-01 5.52180886e-01 7.43891656e-01 5.76116502e-01 4.70772177e-01 8.04967582e-01 1.48691523e+00 -4.10039574e-01 2.11426482e-01 7.40859032e-01 2.82072783e-01 6.35645688e-01 -1.05926236e-02 -1.27752677e-01 -2.74274468e-01 -2.44956255e-01 2.87062407e-01 3.47706944e-01 -2.34929457e-01 -5.75617850e-01 -9.54255700e-01 8.70422900e-01 6.22046471e-01 6.50659740e-01 -5.12753725e-01 -6.05489433e-01 2.93870598e-01 2.30867997e-01 5.43102741e-01 5.88625312e-01 -2.53427923e-01 2.10440382e-01 -6.95127428e-01 4.47066352e-02 3.51448447e-01 5.80596328e-01 1.10526693e+00 -3.05224270e-01 -4.35857147e-01 8.32122087e-01 8.66273344e-02 2.48524502e-01 8.53144169e-01 -4.56165552e-01 6.73667863e-02 1.04328465e+00 -3.23582083e-01 -1.60770106e+00 -2.90711850e-01 -3.96449506e-01 -8.95011425e-01 6.94874078e-02 7.56037384e-02 5.59561014e-01 -5.31045437e-01 1.28353333e+00 1.94966882e-01 -4.93345261e-02 -8.11240152e-02 9.73727524e-01 8.83612216e-01 5.24461389e-01 1.36600584e-01 -3.68236929e-01 1.33223987e+00 -6.15329444e-01 -3.19980770e-01 2.60359287e-01 3.87101203e-01 -8.19911420e-01 1.22513390e+00 5.85273385e-01 -6.41289890e-01 -6.90727651e-01 -8.60926032e-01 2.90864438e-01 -6.83086097e-01 4.18329328e-01 7.68564761e-01 4.37164098e-01 -8.01500082e-01 4.34881151e-01 -1.02454953e-01 -3.19512814e-01 2.03432485e-01 2.99360961e-01 -6.40555561e-01 -2.09564880e-01 -8.65905464e-01 8.63565564e-01 5.40386438e-01 -9.57729965e-02 -2.06526548e-01 -4.47542965e-01 -5.68997204e-01 3.40753615e-01 3.51072252e-01 -2.51852393e-01 1.72292650e-01 -1.50449014e+00 -1.00805557e+00 7.13742375e-01 -5.52282743e-02 -1.36128411e-01 8.44043195e-02 2.77890295e-01 -5.95225811e-01 4.65303421e-01 1.39669374e-01 5.65469563e-01 8.02905023e-01 -1.06801581e+00 -5.33372760e-01 -5.19977570e-01 -2.19101101e-01 1.92512840e-01 -6.92202568e-01 -6.34663403e-02 -2.84635365e-01 -5.54742754e-01 3.79808366e-01 -8.18540037e-01 -6.75999150e-02 -3.13460529e-02 -1.30142331e-01 -3.10114264e-01 7.72314608e-01 -6.06186867e-01 1.12722540e+00 -2.31670046e+00 2.14035571e-01 8.31452727e-01 6.88479021e-02 4.27503102e-02 -1.93783984e-01 1.28476635e-01 -2.28074849e-01 -5.87654002e-02 -4.49307971e-02 3.15038711e-01 -2.97358394e-01 6.66305497e-02 1.85096890e-01 3.19507599e-01 4.86008614e-01 5.54143369e-01 -7.32954264e-01 -8.53312254e-01 4.27759647e-01 1.59278080e-01 -4.95325923e-01 8.99035111e-02 1.57043308e-01 3.77193362e-01 -7.02664256e-01 7.95251489e-01 6.01545453e-01 -3.97369042e-02 -2.06295162e-01 -3.84243250e-01 -1.81926340e-02 -4.23376024e-01 -1.37912476e+00 1.03089380e+00 -4.65391099e-01 3.26558918e-01 -4.04480457e-01 -9.74798441e-01 1.07422662e+00 5.43942042e-02 8.90551448e-01 -1.04551613e+00 5.03404737e-02 3.35737437e-01 3.28264348e-02 -8.15587282e-01 4.11386877e-01 5.09168580e-02 3.37434523e-02 1.62979260e-01 1.25716746e-01 3.11258018e-01 1.64730117e-01 -2.72745527e-02 6.63780630e-01 -4.25959110e-01 7.57035017e-01 -5.17236531e-01 8.92732561e-01 -3.05587769e-01 1.89367265e-01 4.42891687e-01 -5.09856082e-02 4.01391327e-01 2.66218692e-01 -5.28699934e-01 -6.49408817e-01 -7.58398652e-01 -3.43643457e-01 8.96155953e-01 4.43065405e-01 -1.62883550e-01 -2.21891850e-01 -5.77810764e-01 -5.87226935e-02 4.66701567e-01 -5.56655884e-01 -2.66272098e-01 -4.90415879e-02 -6.95003033e-01 -3.26129049e-02 4.17662486e-02 6.34007394e-01 -1.14962471e+00 -4.51410472e-01 -2.93836594e-01 -2.02154577e-01 -7.54208446e-01 -1.58292294e-01 2.43688509e-01 -6.89115942e-01 -1.15412426e+00 -6.50942087e-01 -6.76374316e-01 6.37846053e-01 6.42825603e-01 6.83073342e-01 2.04828545e-01 -6.23017192e-01 4.97546583e-01 -7.13111997e-01 -1.89840600e-01 -7.66573176e-02 -2.00538993e-01 -7.97728524e-02 5.00162721e-01 3.37620676e-01 -2.55448580e-01 -5.39807379e-01 2.83928841e-01 -1.05378592e+00 -3.79182875e-01 1.01902330e+00 8.44070435e-01 6.56026185e-01 4.75497365e-01 5.73202133e-01 -3.25637132e-01 7.62295723e-01 -6.87169552e-01 -3.33485693e-01 3.61951411e-01 -8.44289243e-01 1.37726173e-01 5.75467110e-01 -4.62124079e-01 -1.02309549e+00 -6.05384745e-02 1.64877161e-01 -2.25476861e-01 -2.05070913e-01 5.77313364e-01 -3.87375951e-01 -4.83889073e-01 3.99714977e-01 5.45290053e-01 1.88144848e-01 -3.16369653e-01 2.46133417e-01 6.17452621e-01 4.74803336e-02 -4.05534238e-01 5.11444569e-01 2.31332898e-01 2.86796242e-01 -1.02092195e+00 -4.34439719e-01 -1.09785807e+00 -6.53609872e-01 -3.41849387e-01 6.36957169e-01 -7.27165699e-01 -7.46003926e-01 1.66033715e-01 -7.26363242e-01 7.70304441e-01 -1.72820941e-01 5.91287911e-01 -3.71296734e-01 8.05466771e-01 -1.25662610e-02 -7.97409475e-01 -3.24617594e-01 -1.47043633e+00 8.80498588e-01 4.41867143e-01 -1.08660422e-02 -6.62428141e-01 -2.32328057e-01 3.09143692e-01 2.09734559e-01 9.71210822e-02 1.32774186e+00 -9.31096017e-01 -7.76607394e-01 -4.81611311e-01 -3.68765682e-01 4.60668296e-01 2.53250122e-01 -1.47992402e-01 -6.84264541e-01 -1.60394851e-02 1.95790157e-01 -2.20168754e-01 9.55488026e-01 4.45336550e-01 1.35716343e+00 -4.79229659e-01 -1.77420363e-01 1.84907600e-01 1.51915872e+00 2.54175365e-01 7.13448763e-01 4.98121470e-01 3.08636099e-01 7.33475268e-01 1.08545721e+00 6.75592661e-01 9.10180286e-02 7.48148799e-01 4.14658129e-01 -1.65465951e-01 4.27153051e-01 1.84921622e-01 5.47178909e-02 3.97865951e-01 -2.64301121e-01 4.01980467e-02 -3.82196784e-01 2.84825861e-01 -1.65225148e+00 -1.18449771e+00 2.50553098e-02 2.43650293e+00 2.95188218e-01 -1.22112669e-01 1.92954510e-01 2.66448915e-01 6.06473207e-01 1.74700707e-01 -1.54991075e-01 -4.71633196e-01 -3.49095881e-01 -2.03769773e-01 1.56631440e-01 7.60296658e-02 -1.06002963e+00 4.57799286e-01 4.78268671e+00 1.33222413e+00 -1.24614799e+00 -2.99884379e-01 8.78101587e-01 1.14203736e-01 -4.57601219e-01 -1.38771860e-02 -5.58313668e-01 5.63747287e-01 2.21943513e-01 -5.37814200e-03 2.94510901e-01 1.15554535e+00 3.16472203e-01 -6.30955219e-01 -6.06267214e-01 9.72910166e-01 3.74881208e-01 -9.82297122e-01 3.60435665e-01 1.53692320e-01 6.42130017e-01 -6.61110342e-01 4.37888294e-01 -1.17405236e-01 -6.13460541e-01 -6.91132545e-01 3.23609143e-01 7.43960321e-01 2.95379758e-01 -8.41683805e-01 8.42083097e-01 4.04758811e-01 -1.02158237e+00 -4.97271925e-01 -3.92609060e-01 9.97274667e-02 -4.56019491e-01 6.22154534e-01 -8.26006770e-01 7.08728492e-01 7.02957332e-01 7.19049573e-01 -1.12710583e+00 1.33155906e+00 4.28199440e-01 -3.35643115e-03 -1.63929969e-01 -2.45353460e-01 5.11069834e-01 -3.96686375e-01 5.66875577e-01 8.67738843e-01 4.33249772e-01 1.95582137e-01 4.44863141e-02 5.67040443e-01 3.08320493e-01 8.30615938e-01 -5.92498302e-01 1.32002398e-01 9.11946520e-02 1.62061155e+00 -8.89395654e-01 -1.94312483e-01 -2.20391631e-01 7.82602251e-01 -1.50782173e-03 -2.18800027e-02 -4.81939644e-01 -1.98681012e-01 1.85750410e-01 1.90865263e-01 2.55364686e-01 3.90872806e-02 -2.17219427e-01 -9.86196518e-01 3.11384141e-01 -9.18491125e-01 6.30009532e-01 -5.85091591e-01 -1.33346629e+00 7.21689939e-01 1.62129849e-01 -1.69057953e+00 -3.98859046e-02 -5.46920061e-01 -6.53105438e-01 7.09702075e-01 -1.55621445e+00 -1.28202868e+00 -4.14908111e-01 8.19746435e-01 4.91916329e-01 -4.52467471e-01 7.14922667e-01 6.87604845e-02 -3.20301145e-01 3.13986003e-01 3.27828854e-01 -3.85443032e-01 6.37753546e-01 -9.05472755e-01 -7.81775296e-01 6.33474052e-01 1.88005909e-01 5.55904388e-01 4.35453266e-01 -4.04739648e-01 -1.02983320e+00 -6.54864311e-01 8.77433717e-01 1.24740385e-01 4.45822656e-01 1.10190921e-01 -8.13619137e-01 -1.14709102e-01 -6.64331242e-02 -1.43948302e-01 8.57358158e-01 1.65906474e-01 -2.71693885e-01 -3.08690131e-01 -1.20259047e+00 5.21078110e-01 1.43849298e-01 -2.66009927e-01 -2.57730365e-01 2.18901098e-01 2.44581595e-01 4.04276341e-01 -9.50905621e-01 3.99627805e-01 7.58267105e-01 -1.27783084e+00 1.12801337e+00 -2.52932698e-01 4.34409559e-01 -1.30552471e-01 -3.20387810e-01 -9.99239206e-01 -5.92666447e-01 1.83900431e-01 3.74791205e-01 1.25304711e+00 2.91766346e-01 -4.06167358e-01 4.98740494e-01 5.61079264e-01 1.00223243e-01 -1.10074806e+00 -1.02342403e+00 -6.17263854e-01 -4.27592516e-01 -1.28943548e-01 3.91818225e-01 5.86855829e-01 3.32761228e-01 5.63394763e-02 -4.52620894e-01 -5.54383919e-02 4.36654955e-01 4.71523911e-01 3.76052380e-01 -1.40700316e+00 -2.97798634e-01 -5.76367795e-01 -7.88681209e-01 -4.09513950e-01 5.59615307e-02 -8.34576249e-01 -1.45254388e-01 -9.19840097e-01 5.70138216e-01 -5.03658414e-01 -6.69706583e-01 1.17382303e-01 -2.94754952e-01 1.12396009e-01 3.95540655e-01 4.17155176e-01 -7.66898870e-01 6.05895340e-01 1.17453587e+00 -3.46550643e-01 -3.61572206e-01 3.77817541e-01 -6.97959363e-01 7.07491159e-01 4.37922210e-01 -1.66889027e-01 -4.47293788e-01 2.84440070e-01 -8.87733232e-03 7.16017634e-02 6.19657397e-01 -1.07971227e+00 4.53284979e-02 -3.17401201e-01 6.84483886e-01 -4.36291158e-01 4.17521119e-01 -1.14599419e+00 -7.52678141e-03 4.08863604e-01 -4.75295544e-01 -1.69887319e-01 -2.99766362e-02 7.21050560e-01 -5.01798511e-01 -3.49709958e-01 7.27554679e-01 -2.64920682e-01 -1.41342390e+00 2.35905707e-01 -2.40863964e-01 -3.91484976e-01 1.20838559e+00 -4.90235388e-01 1.12798981e-01 -3.08699012e-01 -6.08449280e-01 -1.09210327e-01 2.37304479e-01 5.42330980e-01 9.27432597e-01 -1.32637000e+00 -4.16220248e-01 1.80366442e-01 6.54412031e-01 -8.00321400e-01 6.38205647e-01 9.63838220e-01 1.28399478e-02 4.66597438e-01 -6.19817138e-01 -5.88781595e-01 -1.56289744e+00 8.64355445e-01 -3.50232050e-02 -4.07282501e-01 -1.77371606e-01 3.57746214e-01 2.70447791e-01 9.50567201e-02 7.27281421e-02 -7.59105012e-02 -8.25148344e-01 4.88481522e-01 3.29812318e-01 5.87572694e-01 1.16290092e-01 -9.05464172e-01 -4.04929072e-01 8.85305166e-01 4.31482978e-02 4.46551532e-01 1.17598021e+00 -3.64629507e-01 -1.87452406e-01 2.56182224e-01 1.39314568e+00 -2.14727312e-01 -5.59813678e-01 -2.62070537e-01 2.28590846e-01 -6.80705428e-01 2.01829076e-01 -7.18930721e-01 -8.39255750e-01 7.67005920e-01 9.50621963e-01 1.55868217e-01 1.56007540e+00 1.14821633e-02 3.09146553e-01 3.40511799e-01 4.06953663e-01 -1.05289900e+00 2.74256140e-01 -6.94482625e-02 9.59217608e-01 -1.60514033e+00 9.41187814e-02 -5.42538941e-01 -1.05093193e+00 1.31179655e+00 3.33567768e-01 -2.29896512e-02 7.92838335e-01 -4.08829749e-01 -2.82277703e-01 -1.68650165e-01 -3.45261246e-01 -3.64007205e-01 9.13340688e-01 4.16286647e-01 2.16884941e-01 5.12517318e-02 -6.57989681e-01 8.18729162e-01 3.54365557e-01 -1.60286024e-01 2.02624023e-01 3.56299788e-01 -7.46874809e-01 -8.74160290e-01 -3.24840546e-01 6.06876850e-01 -1.20285742e-01 -1.27369434e-01 -2.66543418e-01 7.00446248e-01 4.86019045e-01 9.72234249e-01 -1.68272704e-01 -4.59846228e-01 2.94794664e-02 -2.31901616e-01 2.10403830e-01 -1.97732463e-01 -5.01208246e-01 1.13661021e-01 -3.03500205e-01 -5.28588414e-01 -5.51624119e-01 -4.99168009e-01 -6.34315848e-01 4.65268977e-02 -4.94760633e-01 2.00678870e-01 4.45538551e-01 1.16926861e+00 3.07941169e-01 -1.23514682e-01 1.09921372e+00 -7.47657359e-01 -3.57094377e-01 -7.76347280e-01 -8.50239933e-01 6.01713419e-01 5.96637614e-02 -6.92508876e-01 -3.05916995e-01 -1.63849115e-01]	[10.691283226013184, 0.9112085700035095]
297127f2-bb46-4dad-b2b3-2cb2e6a43ac8	infinite-dimensional-optimization-and	2205.15368	null	https://arxiv.org/abs/2205.15368v1	https://arxiv.org/pdf/2205.15368v1.pdf	Infinite-dimensional optimization and Bayesian nonparametric learning of stochastic differential equations	The paper has two major themes. The first part of the paper establishes certain general results for infinite-dimensional optimization problems on Hilbert spaces. These results cover the classical representer theorem and many of its variants as special cases and offer a wider scope of applications. The second part of the paper then develops a systematic approach for learning the drift function of a stochastic differential equation by integrating the results of the first part with Bayesian hierarchical framework. Importantly, our Baysian approach incorporates low-cost sparse learning through proper use of shrinkage priors while allowing proper quantification of uncertainty through posterior distributions. Several examples at the end illustrate the accuracy of our learning scheme.	['Jinpu Zhou', 'Riten Mitra', 'Arnab Ganguly']	2022-05-30	null	null	null	null	['sparse-learning']	['methodology']	[-2.12157801e-01 2.91155457e-01 3.40206847e-02 -6.70672208e-02 -1.02172089e+00 -3.64706516e-01 3.84772003e-01 -2.82510459e-01 -1.72631681e-01 1.12391245e+00 1.40653387e-01 1.35378480e-01 -5.34160554e-01 -3.47653449e-01 -3.85056466e-01 -1.16578794e+00 -3.71209145e-01 2.16411501e-01 -1.53721690e-01 -1.15119100e-01 3.97847265e-01 4.09572840e-01 -1.06465697e+00 -4.87367630e-01 6.67829275e-01 1.22737384e+00 -1.21176742e-01 7.78856039e-01 1.82776392e-01 6.76897109e-01 -4.42472100e-01 -2.13346153e-01 4.88545656e-01 -2.61908114e-01 -5.77885032e-01 3.69096339e-01 1.72893703e-01 -3.91215652e-01 -3.10692221e-01 1.32648540e+00 6.46145225e-01 3.59656155e-01 1.11525297e+00 -1.16335511e+00 -2.62570232e-01 5.15624881e-01 -6.07321858e-01 2.74873197e-01 8.30924660e-02 -1.19513616e-01 9.65874195e-01 -1.16313076e+00 7.15460032e-02 1.10665488e+00 1.09714603e+00 1.21715792e-01 -1.32465279e+00 -1.40789300e-01 -9.24168155e-03 -3.56321260e-02 -1.50713277e+00 -4.44410652e-01 8.74446929e-01 -8.56261253e-01 5.81913233e-01 1.11448511e-01 6.47044420e-01 7.00676084e-01 4.07626808e-01 9.59514320e-01 1.14197469e+00 -4.60429966e-01 6.02993488e-01 3.38164926e-01 3.02311480e-01 8.11991870e-01 4.68836427e-01 1.88983932e-01 -2.17844099e-01 -3.86697173e-01 8.82252574e-01 -1.85660228e-01 -3.12139213e-01 -8.18051159e-01 -1.01080394e+00 1.26620626e+00 -2.99975067e-01 3.28918397e-01 -3.73790592e-01 4.12371010e-02 3.15570176e-01 4.85360101e-02 5.44372737e-01 1.15995362e-01 -1.20010585e-01 -9.15120393e-02 -1.21674812e+00 3.46693754e-01 1.28432870e+00 1.16542721e+00 4.07406986e-01 4.45047587e-01 -1.36266485e-01 4.97790307e-01 6.01549864e-01 5.91764569e-01 1.64453626e-01 -1.37751114e+00 1.93323165e-01 -4.40632313e-01 3.75906408e-01 -6.94457293e-01 -3.35877448e-01 -7.14030564e-01 -1.09819615e+00 2.84255058e-01 4.65065122e-01 -6.23956084e-01 -1.70027807e-01 1.49438858e+00 3.86327893e-01 3.49109054e-01 2.46323153e-01 7.57845700e-01 2.14887112e-01 6.37404323e-01 -3.90350759e-01 -6.22657001e-01 9.19316590e-01 -6.21025383e-01 -1.01423883e+00 3.95881772e-01 2.57899046e-01 -6.41860247e-01 4.25675809e-01 7.09356070e-01 -1.20871806e+00 -2.01694667e-01 -1.16887963e+00 3.69865179e-01 7.06437826e-02 1.50562912e-01 6.33295536e-01 7.42443502e-01 -1.05494547e+00 9.81151044e-01 -7.77743459e-01 -1.05799161e-01 1.93438768e-01 2.99472392e-01 2.22362816e-01 2.66456693e-01 -1.05868435e+00 9.12824035e-01 3.93261351e-02 3.49340588e-01 -8.49852383e-01 -9.44595993e-01 -9.83393013e-01 -2.03331411e-01 -7.21120238e-02 -7.17449248e-01 1.46056104e+00 -4.61256564e-01 -1.91735089e+00 5.76523781e-01 -2.26223227e-02 -3.88070643e-01 8.12463999e-01 -3.38508248e-01 9.23820511e-02 1.40377596e-01 5.22222891e-02 -1.59594595e-01 1.12351847e+00 -8.74431729e-01 -4.51222360e-01 -3.69015634e-01 -1.55889839e-01 8.34903121e-02 2.45807227e-03 -2.07780749e-01 1.74540460e-01 -8.65349650e-01 4.03821170e-02 -5.71199954e-01 -4.53294426e-01 -1.24816433e-01 -2.88884580e-01 -2.41478205e-01 5.10614991e-01 -5.58960676e-01 1.31734729e+00 -2.07326388e+00 7.08441794e-01 2.94092745e-01 1.54578596e-01 -2.55752951e-01 4.22885507e-01 6.31552577e-01 -1.78663656e-01 -1.92467183e-01 -6.39349818e-01 -4.70310867e-01 3.56938601e-01 1.81852266e-01 -4.85424101e-01 1.17711854e+00 8.19672197e-02 4.39062208e-01 -7.42083073e-01 -4.77981746e-01 3.32841158e-01 4.82035697e-01 -2.76146680e-01 -1.97914634e-02 1.53175324e-01 3.51959765e-01 -5.57409644e-01 4.49452639e-01 5.63803554e-01 -1.40959054e-01 -2.08121255e-01 -1.92147836e-01 -2.69643247e-01 -2.26719186e-01 -1.91607070e+00 1.70673275e+00 -3.20264429e-01 5.87583780e-01 7.62696087e-01 -1.59045470e+00 5.54589033e-01 5.84989190e-01 8.02383661e-01 2.04788625e-01 2.07781598e-01 3.65736842e-01 -4.37905490e-01 -5.56088328e-01 2.56927073e-01 -8.69288683e-01 -9.32648331e-02 3.86320472e-01 2.71259099e-01 -5.53642392e-01 8.99749249e-02 2.11466134e-01 7.24808455e-01 1.82674572e-01 6.76651120e-01 -8.26451600e-01 7.93058753e-01 -3.03116709e-01 5.06073594e-01 7.38874555e-01 -4.51333016e-01 3.68863404e-01 4.92499262e-01 -1.48100749e-01 -1.11314499e+00 -9.84599054e-01 -8.66741538e-01 5.08958995e-01 -1.40164852e-01 -1.05700284e-01 -6.25825942e-01 -2.25921571e-01 1.63398713e-01 7.05744147e-01 -8.72171283e-01 2.00309321e-01 -2.22530603e-01 -9.93010938e-01 3.51919770e-01 5.48052371e-01 3.28857213e-01 -2.49462441e-01 -4.74834323e-01 8.45434517e-02 4.81832884e-02 -7.03389883e-01 -2.71236897e-01 3.29730749e-01 -1.08394372e+00 -1.00473762e+00 -1.32397652e+00 -4.65069473e-01 3.34253341e-01 -2.31090575e-01 8.29936743e-01 -6.55180454e-01 -3.44588786e-01 1.12905943e+00 1.40076861e-01 -4.04393256e-01 -3.89814675e-02 -3.20842266e-01 3.16874266e-01 3.63101065e-03 1.84508353e-01 -6.00184619e-01 -2.88723230e-01 1.67563688e-02 -4.78212774e-01 -7.85924196e-01 3.50808084e-01 1.04565501e+00 6.75222993e-01 1.64407864e-01 5.72386265e-01 -5.07493794e-01 6.79308057e-01 -7.01269627e-01 -1.15150476e+00 -1.17145106e-01 -5.69638491e-01 3.19373667e-01 2.79717118e-01 -2.46246502e-01 -1.16101253e+00 1.13191232e-01 -4.47982028e-02 -2.87692666e-01 2.42558971e-01 3.24820220e-01 -3.47541831e-02 -1.32176980e-01 5.60267329e-01 1.57010108e-01 1.35027796e-01 -4.67675567e-01 2.94221610e-01 5.07621646e-01 3.77734900e-01 -7.26279199e-01 7.16165602e-01 5.26833951e-01 5.85780442e-01 -1.19054055e+00 -1.20729387e+00 -7.07970023e-01 -9.02929783e-01 -1.46543115e-01 6.86721325e-01 -8.17657411e-01 -7.84942269e-01 3.99800479e-01 -1.11085677e+00 -1.98029801e-01 -8.50109935e-01 8.61922860e-01 -1.11183846e+00 7.03112066e-01 -7.04233348e-01 -1.44622934e+00 -1.45457402e-01 -9.93059278e-01 1.09646153e+00 1.39904805e-02 -1.51771009e-01 -1.63689482e+00 6.27971172e-01 -1.69487119e-01 2.79231668e-01 2.89744318e-01 3.79397899e-01 -5.29097676e-01 -2.81679332e-01 -3.06986481e-01 1.79843128e-01 7.36941457e-01 -2.14200988e-01 -1.49373278e-01 -9.25163984e-01 -3.67727816e-01 9.25757766e-01 -1.91066325e-01 7.15943098e-01 8.41413736e-01 7.03430772e-01 -2.70582348e-01 -1.22335337e-01 6.59564614e-01 1.78867173e+00 -2.39464819e-01 2.88700402e-01 7.33793378e-02 2.42378861e-01 5.29648483e-01 5.98760843e-01 9.84150529e-01 1.51679777e-02 4.55929399e-01 2.49298915e-01 4.20900941e-01 2.63898611e-01 3.80864441e-01 3.72634053e-01 1.01164329e+00 -9.66154039e-02 3.19151580e-01 -4.82125223e-01 4.91315484e-01 -1.93914580e+00 -1.16425848e+00 -2.94917166e-01 2.25607204e+00 7.69772530e-01 -4.25075322e-01 2.92816579e-01 3.08505148e-01 6.54101431e-01 6.23708218e-02 -5.53631246e-01 -1.96894974e-01 -2.64439374e-01 2.57695973e-01 6.56099796e-01 8.91766071e-01 -1.27273309e+00 1.66833133e-01 8.52039528e+00 7.72625685e-01 -3.89365077e-01 1.43393457e-01 2.78350443e-01 5.13263717e-02 -8.73635188e-02 -2.08054543e-01 -9.87478435e-01 3.92679513e-01 8.81495714e-01 -4.36654866e-01 1.14589989e-01 9.07473087e-01 3.30343723e-01 -2.23469421e-01 -1.01046836e+00 1.10692859e+00 1.63430721e-01 -1.05198240e+00 -6.16387665e-01 2.28524625e-01 9.62893844e-01 -8.10774490e-02 6.99783266e-02 9.67836529e-02 3.48082542e-01 -8.54833603e-01 6.65016055e-01 1.02224016e+00 4.30680215e-01 -8.04412484e-01 9.62278366e-01 3.59736145e-01 -9.93792176e-01 -8.51825178e-02 -6.30263090e-01 -3.40091288e-01 4.96886849e-01 1.04563046e+00 -1.49282813e-01 5.46246946e-01 4.13218319e-01 9.34377730e-01 9.01016742e-02 1.39748383e+00 -1.25237331e-01 6.29650772e-01 -5.70711613e-01 -5.98032307e-03 2.94576347e-01 -6.48335874e-01 1.01256478e+00 1.32848716e+00 5.27022064e-01 2.08726734e-01 3.31969000e-02 7.80027151e-01 4.06560838e-01 8.96748155e-02 -7.07994044e-01 1.84686005e-01 1.09695181e-01 1.11509991e+00 -2.14919284e-01 -4.30788517e-01 -5.44745862e-01 6.06672525e-01 -7.10383356e-02 4.39731836e-01 -5.64696670e-01 -6.56302989e-01 4.22312260e-01 -1.34785116e-01 5.48304498e-01 -3.08341295e-01 -4.41550881e-01 -1.41710484e+00 -1.43601567e-01 -4.40265507e-01 5.61028123e-01 -2.58366823e-01 -1.52377355e+00 -1.02255844e-01 2.64854610e-01 -1.03871679e+00 -4.22403991e-01 -8.44553471e-01 -4.66002017e-01 1.07288957e+00 -1.21334076e+00 -5.13237476e-01 1.56258062e-01 5.91247559e-01 4.54942167e-01 -3.82208586e-01 8.44802499e-01 3.97879966e-02 -6.97711289e-01 8.79288316e-02 7.79527068e-01 -9.27835703e-02 2.48293072e-01 -1.65768099e+00 -3.42433929e-01 8.84158731e-01 -1.84078649e-01 4.75510985e-01 1.16008413e+00 -4.35845792e-01 -1.27872658e+00 -6.87736213e-01 4.75513577e-01 -3.72957349e-01 1.10776114e+00 8.44286829e-02 -7.27927446e-01 7.18321204e-01 3.08430314e-01 -1.45176619e-01 7.02755749e-01 9.59860906e-03 1.51204795e-01 7.11802542e-02 -1.29726827e+00 1.26577675e-01 3.57132852e-01 -2.94697464e-01 -8.98995101e-01 4.66064483e-01 2.01461762e-01 -2.28488177e-01 -1.26753545e+00 2.95730293e-01 5.31480193e-01 -8.37744474e-01 9.86053348e-01 -4.85166579e-01 1.56452321e-02 -1.63207218e-01 -4.30305660e-01 -1.07927465e+00 -3.36332470e-01 -1.18041658e+00 -6.51475668e-01 9.99710917e-01 6.79201335e-02 -6.12104118e-01 5.95287502e-01 4.96478498e-01 -8.60691220e-02 -7.96404958e-01 -1.13538885e+00 -9.42326367e-01 6.30086124e-01 -3.34022582e-01 -1.28378868e-01 6.43278420e-01 3.42935145e-01 1.58820584e-01 -4.96378899e-01 7.30816498e-02 1.47319424e+00 -1.02177925e-01 3.44833732e-01 -1.34400022e+00 -5.38456559e-01 -5.40571988e-01 -4.75876331e-01 -1.12639642e+00 2.58210540e-01 -7.38747954e-01 1.94106892e-01 -1.25183105e+00 1.10859081e-01 -6.02254421e-02 -1.25902295e-01 -3.57376993e-01 1.52579516e-01 -1.40209138e-01 -1.91880465e-01 1.33415312e-01 -5.09983182e-01 7.36136079e-01 9.09333348e-01 2.03517824e-03 1.58661380e-01 6.36818111e-01 -4.47644591e-01 1.11763573e+00 7.70156503e-01 -2.48596832e-01 -4.59095180e-01 9.39302519e-02 8.22126418e-02 2.04837590e-01 4.20043558e-01 -7.99723804e-01 2.42679715e-01 1.12591526e-02 1.81014344e-01 -5.77091575e-01 4.18100506e-01 -6.56673908e-01 -5.73857129e-02 3.55692595e-01 -2.96756089e-01 -9.63397175e-02 7.17579871e-02 9.90637779e-01 -2.15036094e-01 -8.93886805e-01 1.20955586e+00 -1.20753258e-01 -4.06900227e-01 1.06004052e-01 -5.09687185e-01 4.11015540e-01 1.05420089e+00 -6.06874675e-02 5.08771062e-01 -4.77233648e-01 -1.21441674e+00 1.03589155e-01 8.59772116e-02 -4.96695429e-01 4.58205640e-01 -1.28626633e+00 -7.45964468e-01 -8.87301713e-02 -4.71159756e-01 -8.28819126e-02 1.53706938e-01 1.31340516e+00 -2.08530515e-01 6.10473633e-01 1.90981463e-01 -6.95304751e-01 -8.65590930e-01 5.30826867e-01 6.41529262e-01 -1.98424786e-01 -7.61463165e-01 8.72658372e-01 -3.50815803e-03 -6.37206482e-03 6.03289068e-01 -1.27109379e-01 -9.95076895e-02 6.75580055e-02 6.15947723e-01 9.36905026e-01 -3.13329130e-01 -6.63040578e-01 -2.09097415e-01 8.05184245e-01 4.76923615e-01 -6.60747766e-01 1.36173415e+00 -5.94512701e-01 -1.58174008e-01 1.07446623e+00 1.26107085e+00 -1.09832577e-01 -1.54785728e+00 -3.99912715e-01 6.25479966e-02 -1.48452818e-01 2.49519050e-01 -3.57282281e-01 -8.92208457e-01 9.37132239e-01 4.00197268e-01 2.40307599e-01 9.89625990e-01 -1.07920587e-01 2.36457378e-01 4.07825530e-01 4.29398477e-01 -1.25801325e+00 -9.78274047e-02 5.02334416e-01 1.00393653e+00 -1.08362949e+00 3.10966820e-01 -1.61959082e-01 -4.59169388e-01 1.31570089e+00 -1.98982745e-01 -5.57823896e-01 1.18334699e+00 4.68424022e-01 -3.45079392e-01 -1.30924150e-01 -3.93111080e-01 -4.15874273e-02 2.91451067e-01 7.29254007e-01 5.03319263e-01 -2.43537486e-01 -4.85119820e-01 7.08660007e-01 -9.23986956e-02 -1.50577396e-01 6.27545595e-01 8.72377098e-01 -5.42713284e-01 -7.78649807e-01 -5.03135264e-01 1.42741650e-01 -6.98686957e-01 3.37104872e-02 2.93900222e-01 6.88672900e-01 -4.16716397e-01 7.47457266e-01 -2.59425133e-01 3.57813627e-01 1.41992837e-01 2.78505266e-01 7.64156222e-01 -4.12611216e-01 -3.99656631e-02 4.78839815e-01 -2.01940298e-01 -4.95459795e-01 -6.24570906e-01 -1.44865966e+00 -7.79894829e-01 -2.54434288e-01 -3.17062020e-01 5.11644244e-01 6.18560076e-01 9.31443989e-01 -2.08251446e-01 4.26484466e-01 5.31160593e-01 -9.76627886e-01 -1.53922284e+00 -9.54296768e-01 -1.48693609e+00 -3.24304461e-01 7.39546716e-01 -7.63014674e-01 -7.94000983e-01 3.05932295e-02]	[6.976744651794434, 4.144832134246826]
65a22bce-72ef-4128-8330-3d5bd6539f1e	unimodal-concentrated-loss-fully-adaptive	2204.00309	null	https://arxiv.org/abs/2204.00309v1	https://arxiv.org/pdf/2204.00309v1.pdf	Unimodal-Concentrated Loss: Fully Adaptive Label Distribution Learning for Ordinal Regression	Learning from a label distribution has achieved promising results on ordinal regression tasks such as facial age and head pose estimation wherein, the concept of adaptive label distribution learning (ALDL) has drawn lots of attention recently for its superiority in theory. However, compared with the methods assuming fixed form label distribution, ALDL methods have not achieved better performance. We argue that existing ALDL algorithms do not fully exploit the intrinsic properties of ordinal regression. In this paper, we emphatically summarize that learning an adaptive label distribution on ordinal regression tasks should follow three principles. First, the probability corresponding to the ground-truth should be the highest in label distribution. Second, the probabilities of neighboring labels should decrease with the increase of distance away from the ground-truth, i.e., the distribution is unimodal. Third, the label distribution should vary with samples changing, and even be distinct for different instances with the same label, due to the different levels of difficulty and ambiguity. Under the premise of these principles, we propose a novel loss function for fully adaptive label distribution learning, namely unimodal-concentrated loss. Specifically, the unimodal loss derived from the learning to rank strategy constrains the distribution to be unimodal. Furthermore, the estimation error and the variance of the predicted distribution for a specific sample are integrated into the proposed concentrated loss to make the predicted distribution maximize at the ground-truth and vary according to the predicting uncertainty. Extensive experimental results on typical ordinal regression tasks including age and head pose estimation, show the superiority of our proposed unimodal-concentrated loss compared with existing loss functions.	['ShiLiang Pu', 'Chunmao Wang', 'Jingwei Yan', 'Pengju Yang', 'Yachun Li', 'Zhaoliang Yao', 'Jingjing Wang', 'Qiang Li']	2022-04-01	null	http://openaccess.thecvf.com//content/CVPR2022/html/Li_Unimodal-Concentrated_Loss_Fully_Adaptive_Label_Distribution_Learning_for_Ordinal_Regression_CVPR_2022_paper.html	http://openaccess.thecvf.com//content/CVPR2022/papers/Li_Unimodal-Concentrated_Loss_Fully_Adaptive_Label_Distribution_Learning_for_Ordinal_Regression_CVPR_2022_paper.pdf	cvpr-2022-1	['head-pose-estimation']	['computer-vision']	[ 4.81716841e-02 1.72456279e-01 -3.95992339e-01 -7.85144269e-01 -6.70577884e-01 -2.18334675e-01 1.37405038e-01 1.74189970e-01 -2.74410099e-01 8.75651002e-01 1.56850249e-01 3.41515481e-01 -4.54363614e-01 -4.12523955e-01 -6.00606263e-01 -1.03386927e+00 1.08133316e-01 5.04898965e-01 -3.09219062e-02 8.47623646e-02 1.54834136e-01 1.64779991e-01 -1.57073390e+00 -4.81716469e-02 1.05690217e+00 1.27653623e+00 -6.11797646e-02 -2.30516791e-01 -2.99418755e-02 3.64552438e-01 -4.88451421e-01 -3.24156553e-01 1.99900001e-01 -2.42493451e-01 -4.25545305e-01 1.06509797e-01 6.89290404e-01 -2.69032240e-01 2.23308820e-02 1.18478334e+00 6.02699876e-01 9.86158997e-02 1.10402441e+00 -1.49303877e+00 -5.38191199e-01 4.52456117e-01 -9.94793177e-01 -3.92953217e-01 2.68788517e-01 -2.36099243e-01 1.25566483e+00 -8.47207665e-01 2.14480028e-01 1.36445022e+00 7.36204147e-01 6.08270109e-01 -1.26627684e+00 -1.07827508e+00 5.84313691e-01 1.92044005e-01 -1.66383338e+00 -8.78757536e-02 9.06396806e-01 -4.46352273e-01 -1.32113472e-01 -1.22092068e-01 3.57084930e-01 8.53142560e-01 9.61755589e-02 9.07729328e-01 1.30397975e+00 -2.89104313e-01 1.21641681e-01 1.90729201e-01 1.14810606e-02 8.16514015e-01 2.20802501e-01 1.47223650e-02 -6.12986982e-01 -8.77342299e-02 4.07125652e-01 -1.64156765e-01 -2.14506999e-01 -5.94683111e-01 -7.12550879e-01 8.17404449e-01 3.36168319e-01 -6.49352893e-02 -5.53914644e-02 7.70863444e-02 5.03421724e-01 2.88013685e-02 6.20129764e-01 2.24024579e-02 -4.86734986e-01 1.96770251e-01 -7.75376439e-01 3.17076445e-01 3.92466158e-01 7.93137968e-01 8.11462045e-01 -2.43344143e-01 -4.32907581e-01 1.12310791e+00 5.35263419e-01 5.35834253e-01 4.09677088e-01 -8.63934696e-01 3.89107078e-01 4.93656158e-01 3.43673974e-02 -1.22140574e+00 -5.12136936e-01 -4.19752300e-01 -9.01879728e-01 2.01914415e-01 7.46278286e-01 -2.48556137e-01 -6.81336701e-01 2.29785466e+00 4.17471677e-01 9.81414691e-02 -3.38897079e-01 8.29776824e-01 6.85592473e-01 4.80142862e-01 1.45928487e-01 -6.07288063e-01 1.05015934e+00 -4.24279273e-01 -8.18424642e-01 -8.40091705e-02 2.95089066e-01 -4.92176592e-01 1.26436269e+00 4.54197466e-01 -8.08434784e-01 -5.35965621e-01 -8.93594086e-01 2.69870520e-01 1.13699742e-01 1.94370747e-01 4.99975562e-01 5.91569543e-01 -5.81157565e-01 4.42336470e-01 -3.05906743e-01 -5.16188964e-02 3.24742168e-01 4.33833838e-01 -1.79214716e-01 1.45468200e-02 -1.29695058e+00 4.59403694e-01 3.96243095e-01 1.78936288e-01 -4.26407546e-01 -6.69656217e-01 -8.10688913e-01 -7.45820627e-02 2.66034037e-01 -3.89516443e-01 1.11446559e+00 -1.04690123e+00 -1.51384377e+00 7.70771265e-01 -1.93644702e-01 -8.02538097e-02 5.75720429e-01 -2.44446039e-01 -1.04880698e-01 -2.21526623e-01 2.52648532e-01 7.94055283e-01 1.07187569e+00 -1.27344584e+00 -7.29973197e-01 -5.42948544e-01 -1.71665609e-01 3.31638873e-01 -4.56523985e-01 -5.29460490e-01 -1.30438834e-01 -5.74218631e-01 3.12298566e-01 -8.49347949e-01 1.01188980e-01 2.09351912e-01 -3.55493635e-01 -7.38507450e-01 6.58603907e-01 -2.78530806e-01 1.42361975e+00 -2.36495614e+00 2.50590481e-02 4.08847272e-01 3.08442228e-02 -2.78732866e-01 -1.21797188e-04 -7.79376104e-02 -1.09372139e-01 -1.28499404e-01 -2.13595480e-01 -1.57883629e-01 1.67574599e-01 1.04081579e-01 -1.92360282e-01 6.57479227e-01 5.52852713e-02 3.93841416e-01 -1.00375855e+00 -7.63466120e-01 -7.31604546e-02 2.07806557e-01 -5.06366372e-01 1.03190809e-01 -2.07431525e-01 5.89175761e-01 -4.68043715e-01 6.14542484e-01 9.03338253e-01 1.17849763e-02 4.09710268e-03 -3.78942728e-01 1.03127033e-01 -2.37990439e-01 -1.42510366e+00 1.21820807e+00 -3.27133864e-01 2.79904604e-01 2.63856929e-02 -9.60338354e-01 1.14641452e+00 1.06097102e-01 6.83708191e-01 -5.56741536e-01 -1.92294698e-02 3.10475230e-01 -1.58967301e-01 -2.50130862e-01 7.75431842e-02 -4.92676109e-01 -2.48099297e-01 1.71595544e-01 -1.81191251e-01 -2.65807919e-02 -6.38467744e-02 -3.33618730e-01 3.20404291e-01 1.45990521e-01 2.75153577e-01 -1.92078307e-01 7.19243228e-01 -5.88391185e-01 1.00692344e+00 5.22271037e-01 -4.73851830e-01 4.86429393e-01 7.05827951e-01 -1.29934534e-01 -6.41021430e-01 -1.24730921e+00 -6.93956614e-01 1.32037520e+00 4.52139169e-01 -1.05551161e-01 -5.90667069e-01 -9.10892725e-01 2.23338500e-01 5.72087765e-01 -6.70305789e-01 -2.20798418e-01 -4.17525381e-01 -7.09131539e-01 2.84640253e-01 3.96910667e-01 6.08026206e-01 -8.66481721e-01 -6.02474101e-02 -6.83972687e-02 -1.77303508e-01 -7.58559465e-01 -6.37218714e-01 5.15150242e-02 -6.59132123e-01 -1.05166054e+00 -7.18745589e-01 -9.32029068e-01 9.10951138e-01 -1.82118282e-01 9.13659275e-01 -2.82437831e-01 -3.02051436e-02 3.62925380e-01 -1.77545741e-01 -4.17980909e-01 1.05183274e-01 3.83516401e-02 1.56185031e-01 2.85112560e-01 2.95494050e-01 -5.09569466e-01 -6.89374983e-01 5.03379881e-01 -6.57491088e-01 -3.19295615e-01 4.78320539e-01 1.06509459e+00 7.82819211e-01 3.33755702e-01 9.80953574e-01 -9.54403460e-01 6.92679465e-01 -5.62671602e-01 -4.36960638e-01 2.69618273e-01 -8.22326303e-01 1.99042708e-01 6.53025270e-01 -5.91795802e-01 -1.04033303e+00 5.89953437e-02 -7.18414113e-02 -2.08376080e-01 2.04573199e-02 4.06664222e-01 -4.06409591e-01 1.85273156e-01 3.66486639e-01 -8.65714774e-02 1.67314693e-01 -1.47319064e-01 3.67177427e-01 6.43728137e-01 2.70056039e-01 -9.06930447e-01 6.59259558e-01 4.67409253e-01 4.31483656e-01 -4.86118078e-01 -1.25589776e+00 -3.43868911e-01 -5.10706842e-01 -3.82270485e-01 7.24613428e-01 -5.57463229e-01 -9.07362938e-01 7.00248361e-01 -7.70719886e-01 -8.31535645e-03 -2.39016548e-01 5.31413376e-01 -5.97623765e-01 4.97460961e-01 -2.39022791e-01 -8.95135105e-01 -2.73242414e-01 -1.00017416e+00 1.14278555e+00 4.30029273e-01 -2.84675777e-01 -1.06277585e+00 -1.31390408e-01 3.41693610e-02 8.51893239e-03 2.86138326e-01 1.19653904e+00 -4.59791601e-01 -1.23223193e-01 -2.52823055e-01 -2.64169097e-01 4.17071611e-01 4.08987373e-01 -1.20721711e-02 -7.96524882e-01 -3.68381351e-01 -1.68905839e-01 -6.76651120e-01 6.59542203e-01 6.92679226e-01 1.49079609e+00 -2.46867523e-01 -2.28729725e-01 5.06016374e-01 1.07068348e+00 8.34437534e-02 3.53151232e-01 1.48773819e-01 5.77953279e-01 9.01969910e-01 1.07735884e+00 7.03961253e-01 2.63418794e-01 7.57461250e-01 5.78904390e-01 1.70953482e-01 1.95319727e-01 -5.11322677e-01 2.62305886e-01 4.58183378e-01 2.22250536e-01 8.44661370e-02 -4.70111132e-01 2.30955586e-01 -1.88328993e+00 -7.65390933e-01 1.20769426e-01 2.55505824e+00 1.03077650e+00 2.07935572e-01 2.43681997e-01 -3.02496981e-02 8.86186540e-01 2.89325655e-01 -8.60855222e-01 -1.90170899e-01 -2.30139215e-02 -1.61086559e-01 3.50303382e-01 5.18143952e-01 -1.09932542e+00 7.05796063e-01 5.74473095e+00 1.29948580e+00 -1.13947725e+00 -1.22177862e-01 6.99164450e-01 1.39281536e-02 -3.44416410e-01 -2.78379977e-01 -1.09834433e+00 8.79437983e-01 3.04187089e-01 -2.43347451e-01 1.54490784e-01 9.70703483e-01 1.92263186e-01 -1.23210615e-02 -1.15887988e+00 1.08186316e+00 7.66331621e-04 -4.69299883e-01 -1.10160187e-02 -6.65617436e-02 6.35024846e-01 -6.16007090e-01 5.73485434e-01 4.80141759e-01 -1.02527123e-02 -1.12851846e+00 7.58773386e-01 5.60501456e-01 9.60720122e-01 -1.00642264e+00 7.74254501e-01 4.72122878e-01 -1.23584199e+00 -1.87031686e-01 -3.63975823e-01 2.02567860e-01 8.06179792e-02 6.78583860e-01 -6.10530317e-01 1.93778887e-01 5.88671207e-01 6.99335515e-01 -4.09649044e-01 8.34902346e-01 -4.19174194e-01 3.86469513e-01 -3.44590247e-01 -8.50344002e-02 1.33286575e-02 -3.50686610e-01 4.06167924e-01 8.51279736e-01 1.30982444e-01 -2.53289968e-01 2.94361562e-01 5.38891852e-01 -7.45221525e-02 3.86561662e-01 -3.69220823e-01 3.67654651e-01 7.64005899e-01 9.47635889e-01 -1.71902731e-01 1.54547006e-01 -3.47800285e-01 6.34594381e-01 5.26586413e-01 2.25073799e-01 -9.42741692e-01 -2.18481496e-01 4.59502727e-01 3.05974603e-01 -1.67253558e-02 1.89789191e-01 -4.28870380e-01 -5.14578521e-01 1.15997881e-01 -6.14263415e-01 5.82842648e-01 -4.56735402e-01 -1.72745836e+00 2.98632026e-01 3.24634731e-01 -1.32793665e+00 -1.81765929e-01 -5.71455359e-01 -4.83536601e-01 6.79321468e-01 -1.28848183e+00 -1.00721276e+00 -3.06384683e-01 4.92517292e-01 3.61999810e-01 -1.29198357e-01 4.44502383e-01 3.66819203e-01 -6.05666995e-01 1.16820765e+00 2.44815737e-01 -1.73544465e-03 1.06588793e+00 -1.28056979e+00 -6.46804273e-01 1.28654689e-01 -3.44087541e-01 5.85941255e-01 6.20763183e-01 -5.12064397e-01 -6.76152587e-01 -1.08554614e+00 6.78528726e-01 -5.20595945e-02 5.29296219e-01 -1.79511920e-01 -9.06498015e-01 4.27079022e-01 -2.70122141e-01 -6.07580179e-03 6.70215726e-01 3.83847415e-01 -4.24998790e-01 -6.89799368e-01 -1.30927277e+00 5.27787328e-01 8.11617315e-01 -3.17779541e-01 -3.66472065e-01 4.08393294e-01 3.54293227e-01 -3.04572672e-01 -7.84864426e-01 9.55694258e-01 9.23929513e-01 -8.41254890e-01 7.74607301e-01 -3.62134218e-01 3.07315439e-01 -3.58430505e-01 -1.88476816e-01 -1.26152778e+00 -5.16438007e-01 -1.73563167e-01 -5.03298920e-03 1.45744753e+00 2.86955923e-01 -7.29712427e-01 8.56402040e-01 4.54303205e-01 6.17246032e-02 -1.20085144e+00 -1.12696803e+00 -8.00160766e-01 2.59123921e-01 -1.10058799e-01 4.65719342e-01 6.70469463e-01 -1.68148652e-01 4.83689129e-01 -5.02017498e-01 1.72095791e-01 8.07218194e-01 7.43219554e-02 5.09341657e-01 -1.45993102e+00 -1.83581293e-01 -5.10412514e-01 -3.81116688e-01 -1.13361466e+00 6.39684916e-01 -7.54172385e-01 2.64458150e-01 -1.32829094e+00 3.66035134e-01 -7.31114686e-01 -4.25954163e-01 3.13219339e-01 -3.75582963e-01 4.67615686e-02 -9.13230106e-02 3.83285612e-01 -6.35551274e-01 8.75319481e-01 1.20540047e+00 -1.56031460e-01 -1.01148210e-01 3.11028272e-01 -6.95583880e-01 1.01644409e+00 6.35360539e-01 -4.73558843e-01 -6.44938886e-01 7.17813298e-02 2.55168051e-01 -1.00339107e-01 -7.39176199e-02 -6.69385374e-01 -2.02373508e-03 -2.85726726e-01 4.18141723e-01 -6.20651841e-01 2.44315565e-01 -9.72683847e-01 -2.66791821e-01 2.87659854e-01 -4.28858310e-01 -3.09832692e-01 -1.39643639e-01 8.32902133e-01 -2.01689795e-01 -3.46418321e-01 1.11295235e+00 2.30860963e-01 -5.54427922e-01 4.73017186e-01 1.37299821e-01 3.86348814e-01 1.05563164e+00 -3.43707502e-01 -6.96366653e-02 -5.08162379e-01 -5.87622464e-01 4.65250134e-01 3.54341358e-01 2.62326300e-01 4.58799541e-01 -1.55639839e+00 -8.08498740e-01 6.89610690e-02 3.15640956e-01 1.68496132e-01 1.62416384e-01 8.34504306e-01 -1.31292166e-02 9.51350704e-02 -1.11851450e-02 -6.94364130e-01 -1.21428859e+00 3.63856047e-01 4.01223928e-01 -2.90896147e-01 -9.96135026e-02 8.69576395e-01 6.69441819e-01 -5.54725766e-01 5.79314291e-01 -5.30378595e-02 -4.15645987e-01 3.81295741e-01 2.68805534e-01 3.72866869e-01 -2.31584862e-01 -5.37117362e-01 -3.51987273e-01 9.31434512e-01 -1.41329601e-01 1.06875844e-01 1.09136331e+00 -1.78921849e-01 -1.94921374e-01 6.37234807e-01 1.15859759e+00 2.24167258e-01 -1.47016931e+00 -2.97489256e-01 -1.42763173e-02 -5.63003242e-01 -1.66001499e-01 -6.05411112e-01 -1.02222359e+00 5.71348250e-01 6.77152753e-01 1.81726161e-02 1.22339642e+00 5.10258600e-02 4.63912398e-01 -2.14550458e-03 4.64854509e-01 -1.29031897e+00 3.77688885e-01 4.56177473e-01 7.73081839e-01 -1.31973791e+00 1.61268696e-01 -4.77494508e-01 -7.76812375e-01 9.35983777e-01 9.83613431e-01 4.11603898e-02 7.27066278e-01 2.41179094e-02 -1.26933426e-01 7.37025887e-02 -3.06679517e-01 -6.81806356e-03 6.00075424e-01 6.80352390e-01 5.64516842e-01 1.12531178e-01 -6.54350877e-01 7.68124282e-01 -8.07338282e-02 -2.23874494e-01 -5.97133040e-02 3.88318509e-01 -5.87090671e-01 -1.06637990e+00 -4.21718985e-01 5.24535954e-01 -2.90428787e-01 1.81258872e-01 -2.44787499e-01 7.44902015e-01 5.15602529e-01 7.08806157e-01 1.27199680e-01 -2.34750703e-01 3.92933130e-01 9.23155982e-04 6.32764518e-01 -6.41872466e-01 1.78326398e-01 -6.57905415e-02 -2.89438307e-01 -1.50420442e-01 -2.78039664e-01 -7.98212349e-01 -1.43343687e+00 -5.10682352e-02 -3.94955933e-01 2.21268713e-01 2.95254976e-01 8.93665552e-01 -8.10244456e-02 1.91938370e-01 9.25989270e-01 -5.25072098e-01 -8.77541065e-01 -9.86481845e-01 -9.37624574e-01 5.60216129e-01 2.23649442e-01 -1.08166349e+00 -6.53479338e-01 -3.17528844e-01]	[9.769686698913574, 3.666600465774536]
71f09953-77a8-47c3-993a-585eb64d6b8c	3dmm-rf-convolutional-radiance-fields-for-3d	2209.07366	null	https://arxiv.org/abs/2209.07366v1	https://arxiv.org/pdf/2209.07366v1.pdf	3DMM-RF: Convolutional Radiance Fields for 3D Face Modeling	Facial 3D Morphable Models are a main computer vision subject with countless applications and have been highly optimized in the last two decades. The tremendous improvements of deep generative networks have created various possibilities for improving such models and have attracted wide interest. Moreover, the recent advances in neural radiance fields, are revolutionising novel-view synthesis of known scenes. In this work, we present a facial 3D Morphable Model, which exploits both of the above, and can accurately model a subject's identity, pose and expression and render it in arbitrary illumination. This is achieved by utilizing a powerful deep style-based generator to overcome two main weaknesses of neural radiance fields, their rigidity and rendering speed. We introduce a style-based generative network that synthesizes in one pass all and only the required rendering samples of a neural radiance field. We create a vast labelled synthetic dataset of facial renders, and train the network on these data, so that it can accurately model and generalize on facial identity, pose and appearance. Finally, we show that this model can accurately be fit to "in-the-wild" facial images of arbitrary pose and illumination, extract the facial characteristics, and be used to re-render the face in controllable conditions.	['Stefanos Zafeiriou', 'Alexandros Lattas', 'Baris Gecer', 'Stathis Galanakis']	2022-09-15	null	null	null	null	['3d-face-modeling']	['computer-vision']	[ 3.01818818e-01 7.81714693e-02 3.07952404e-01 -5.88092506e-01 -3.08223009e-01 -4.59734261e-01 6.96561098e-01 -8.93397808e-01 2.15350285e-01 6.00702167e-01 -1.33132905e-01 2.17025578e-01 2.30320916e-01 -9.42371190e-01 -5.97464800e-01 -7.21077144e-01 2.65482426e-01 3.41035694e-01 -2.39516973e-01 -4.88955319e-01 -1.78926483e-01 9.30686951e-01 -1.96311033e+00 3.33770402e-02 6.73075199e-01 1.10306513e+00 -2.12972537e-01 5.44594526e-01 3.40197347e-02 5.05940914e-01 -6.08883440e-01 -8.29289675e-01 3.95159990e-01 -7.84746587e-01 -4.05242652e-01 3.83275092e-01 8.11518192e-01 -4.99852300e-01 -3.46499026e-01 8.92431557e-01 7.37247825e-01 4.63855732e-03 6.36359632e-01 -1.27905834e+00 -8.75635147e-01 -2.50471622e-01 -5.22618771e-01 -4.62860346e-01 4.79899913e-01 3.43961194e-02 5.13112903e-01 -7.83058822e-01 7.50251770e-01 1.57863569e+00 6.01474583e-01 1.26053548e+00 -1.26140296e+00 -8.48050177e-01 -1.41514316e-01 -1.59702212e-01 -1.33021688e+00 -6.33432508e-01 1.16613364e+00 -2.55845010e-01 3.79829168e-01 3.95520121e-01 9.47998285e-01 1.47007358e+00 2.33888015e-01 4.04434562e-01 1.31969213e+00 -3.10287476e-01 9.42990780e-02 1.01524860e-01 -8.41701686e-01 1.02486527e+00 -2.83468515e-01 1.72020748e-01 -4.24092531e-01 -1.35597870e-01 1.15761030e+00 1.13234539e-02 -3.24787557e-01 -3.38622302e-01 -7.31885314e-01 7.39706039e-01 3.65807444e-01 -4.57196571e-02 -1.70178369e-01 2.09947065e-01 -5.17860800e-02 2.00644374e-01 7.63011813e-01 2.73153096e-01 -1.82918668e-01 1.96667954e-01 -8.99191499e-01 4.60428566e-01 7.04941213e-01 9.24605787e-01 1.02831864e+00 4.82034236e-01 8.60368833e-02 1.08357573e+00 2.07371309e-01 8.93847346e-01 4.70111579e-01 -1.05274284e+00 -4.16950248e-02 4.58786666e-01 -1.99260086e-01 -1.20421052e+00 -4.09445792e-01 -2.32201174e-01 -1.10998368e+00 5.95569551e-01 -3.64272203e-03 -3.65466960e-02 -1.11225963e+00 2.04099846e+00 6.05737507e-01 1.41723216e-01 -1.94135368e-01 7.31897235e-01 9.43888664e-01 5.69607437e-01 -1.53144196e-01 -7.57646747e-03 1.20032048e+00 -4.27470624e-01 -6.08626008e-01 -1.07588522e-01 1.00590929e-01 -9.59417105e-01 9.03921247e-01 3.54600817e-01 -1.26377463e+00 -8.00106466e-01 -9.78191257e-01 -1.40833586e-01 -2.52472818e-01 6.32083192e-02 8.05802345e-01 8.96087646e-01 -1.42042363e+00 5.86166441e-01 -5.55039525e-01 -1.50110379e-01 5.78106284e-01 3.39536637e-01 -5.03760934e-01 -6.45341352e-03 -1.06697023e+00 7.62236059e-01 -2.72323608e-01 3.53281826e-01 -9.27639425e-01 -7.04059362e-01 -8.72702122e-01 -2.10042179e-01 -1.03927493e-01 -1.10350227e+00 9.65306163e-01 -1.42088330e+00 -1.94642377e+00 1.27785552e+00 -9.74296182e-02 2.46037468e-01 5.95993280e-01 1.24061182e-01 -5.16671658e-01 1.67782363e-02 -1.13921195e-01 9.11760271e-01 1.24936497e+00 -1.36729085e+00 -8.66933167e-02 -5.99032819e-01 2.84147486e-02 7.99411684e-02 -2.52379119e-01 4.04368751e-02 -4.02522683e-01 -6.31563604e-01 8.31563994e-02 -1.06907916e+00 -1.33141369e-01 2.65499562e-01 -2.59039253e-01 1.05962262e-01 7.84756005e-01 -5.16272068e-01 5.10214925e-01 -1.99284029e+00 2.05056638e-01 2.17883945e-01 8.34379569e-02 2.63610244e-01 -3.43729317e-01 1.33737147e-01 -2.84493506e-01 1.33544892e-01 -1.92624599e-01 -6.07587874e-01 -1.34493873e-01 2.03542128e-01 -2.26481110e-01 4.00112897e-01 4.00163442e-01 8.78937602e-01 -6.56595170e-01 -3.06893736e-01 1.58446625e-01 1.11213422e+00 -7.60116637e-01 4.72807020e-01 -3.23104024e-01 9.13910508e-01 -4.10383761e-01 6.05169654e-01 1.01722801e+00 1.12644166e-01 -3.81116644e-02 -3.13939989e-01 2.27021784e-01 -2.85355091e-01 -1.01315355e+00 1.79445255e+00 -7.05217600e-01 6.50196612e-01 1.64315119e-01 -6.20873332e-01 1.36559653e+00 3.37964386e-01 5.89384496e-01 -8.99688482e-01 3.50598097e-01 2.22799316e-01 -3.45677197e-01 -4.66262937e-01 2.21898973e-01 -5.08486509e-01 1.08489849e-01 3.61480623e-01 2.42559046e-01 -7.43721187e-01 -2.41142392e-01 -1.99990392e-01 4.96947140e-01 4.35010761e-01 -1.58828720e-01 -6.91303089e-02 6.33303344e-01 -6.89300597e-01 3.65620166e-01 -1.51944607e-02 3.05574834e-01 1.03575075e+00 3.29486996e-01 -7.30906367e-01 -1.08850443e+00 -1.05907321e+00 -8.43968019e-02 7.72637010e-01 -7.33195320e-02 -1.19479194e-01 -1.19055688e+00 -3.10592443e-01 -2.38895297e-01 3.23584318e-01 -8.64730418e-01 -3.84599745e-01 -6.44511640e-01 -7.24587023e-01 5.63981175e-01 2.96639413e-01 6.44225001e-01 -1.06422246e+00 -3.99780452e-01 -1.05625533e-01 -7.21129552e-02 -1.17296100e+00 -2.12797970e-01 -3.26371342e-01 -6.27931952e-01 -9.23148692e-01 -8.54180515e-01 -7.72530496e-01 8.80014658e-01 1.62301678e-02 1.22262073e+00 2.74973810e-01 -5.56111932e-01 2.99305528e-01 -5.47165759e-02 -5.11527717e-01 -6.69990957e-01 -2.03544840e-01 -1.83436589e-03 5.90717554e-01 -1.31638840e-01 -9.84432518e-01 -6.61273062e-01 3.53881299e-01 -1.09938502e+00 2.77563125e-01 2.33638763e-01 6.00068688e-01 5.96738458e-01 -1.00315720e-01 5.14447391e-01 -9.77327287e-01 4.49498713e-01 -1.16279989e-01 -5.54922879e-01 4.85858954e-02 -2.15471745e-01 2.36362647e-02 6.98861182e-01 -3.65792066e-01 -1.28691649e+00 1.20923907e-01 -6.79844260e-01 -6.48258626e-01 -2.06283480e-01 -1.54216304e-01 -6.46308541e-01 -4.54474926e-01 7.87010550e-01 1.92484960e-01 1.66255742e-01 -4.28897589e-01 5.10566652e-01 3.87484252e-01 6.52646303e-01 -6.50400043e-01 1.12496388e+00 7.46937513e-01 4.71897721e-01 -8.58507872e-01 -7.40259230e-01 2.76607096e-01 -7.47505009e-01 -4.04356211e-01 7.59540141e-01 -7.77735114e-01 -7.60472417e-01 8.30273807e-01 -1.27100313e+00 -3.40981245e-01 -2.51858264e-01 -4.63480242e-02 -8.43713522e-01 -2.79111098e-02 -2.72391826e-01 -6.82091117e-01 -3.07632923e-01 -1.16156816e+00 1.36071801e+00 5.24581790e-01 -7.90708438e-02 -1.02032888e+00 1.50154859e-01 9.58366394e-02 4.94380504e-01 8.21440160e-01 8.18156600e-01 2.73646951e-01 -5.08087337e-01 -1.45215660e-01 -1.59099922e-01 5.55576265e-01 4.24093336e-01 2.94763476e-01 -1.40430558e+00 -1.70597896e-01 1.26015931e-01 -3.10870826e-01 4.50361758e-01 1.55960903e-01 1.34994841e+00 -2.24901065e-01 -8.61955136e-02 1.36729145e+00 1.29533184e+00 1.04666889e-01 8.61627817e-01 -1.63139403e-01 9.30653930e-01 7.37967670e-01 -3.07921115e-02 3.72262031e-01 2.00179681e-01 8.47035229e-01 6.83269143e-01 -4.81774122e-01 -4.48270470e-01 -4.49082464e-01 1.56281158e-01 5.55600464e-01 -6.13439560e-01 -7.13396519e-02 -3.27116340e-01 1.44734412e-01 -1.28148937e+00 -1.00975382e+00 2.12737486e-01 1.95003104e+00 9.04259145e-01 -4.14043248e-01 -2.75686141e-02 -5.25703505e-02 4.82551813e-01 1.62406892e-01 -5.80127537e-01 -6.76731229e-01 -3.01353097e-01 9.18707788e-01 1.67451818e-02 3.44734162e-01 -8.26435804e-01 1.05577564e+00 6.49156618e+00 7.24266469e-01 -1.48924553e+00 -7.10508525e-02 9.33216929e-01 6.73263073e-02 -6.97040379e-01 -3.21356207e-01 -6.83845222e-01 2.29780078e-01 7.67398059e-01 1.28061958e-02 5.82750857e-01 8.29437494e-01 8.35108608e-02 2.95529842e-01 -1.15138626e+00 1.22715962e+00 5.28119385e-01 -1.22368789e+00 5.46972275e-01 1.75744206e-01 1.02053249e+00 -5.60469508e-01 5.03559709e-01 -5.38400263e-02 8.23357105e-02 -1.46512520e+00 7.27517962e-01 7.42950559e-01 1.29468727e+00 -9.56672072e-01 2.87822753e-01 1.82816803e-01 -8.72448206e-01 2.96346962e-01 -4.96223032e-01 2.64293432e-01 2.16961000e-02 3.60954493e-01 -5.56270003e-01 4.66048896e-01 6.34210646e-01 5.58094144e-01 -5.55039108e-01 5.91945171e-01 -3.45236510e-01 1.09215729e-01 -1.17287733e-01 8.79432037e-02 8.81699380e-03 -4.87504721e-01 2.47629642e-01 7.86604464e-01 5.79940259e-01 8.41985568e-02 -2.75146067e-01 1.17367065e+00 -3.52092206e-01 1.40297130e-01 -6.68098271e-01 3.17566663e-01 -3.79626192e-02 1.51158476e+00 -3.54083985e-01 4.67295712e-03 -1.24812238e-01 1.20852423e+00 2.27166206e-01 2.96487868e-01 -9.79321301e-01 -1.55610412e-01 8.90712976e-01 2.97512650e-01 -4.79574427e-02 -1.78516537e-01 1.15632556e-01 -1.17408705e+00 -3.16467918e-02 -9.32025611e-01 -2.45046794e-01 -1.08466065e+00 -1.07607615e+00 1.15998316e+00 -2.46889949e-01 -1.09746659e+00 -3.63418162e-01 -7.16418803e-01 -5.15474260e-01 1.09252834e+00 -1.55707741e+00 -1.55382180e+00 -6.02968037e-01 9.43016171e-01 3.55023682e-01 -1.48519620e-01 1.16008568e+00 4.57133800e-01 -3.77752751e-01 8.44199240e-01 -2.42585033e-01 1.83472693e-01 6.61829352e-01 -7.87940681e-01 5.68588138e-01 4.34713572e-01 2.18543306e-01 4.26843077e-01 4.89429384e-01 -1.63814172e-01 -1.45406473e+00 -1.23262513e+00 6.11995220e-01 -5.45548201e-01 1.36475787e-01 -6.69737577e-01 -6.91175878e-01 5.43102384e-01 4.21866439e-02 1.00380041e-01 5.59605300e-01 -2.42667302e-01 -4.80497599e-01 -4.02798206e-01 -1.39536309e+00 7.62196422e-01 1.36407566e+00 -5.02778530e-01 -8.83641914e-02 3.86739492e-01 4.61559206e-01 -6.35789871e-01 -7.28148103e-01 4.40515250e-01 8.16455901e-01 -1.37684131e+00 1.05922627e+00 -5.87685466e-01 5.19538820e-01 -1.84863843e-02 -2.30718087e-02 -1.63521075e+00 -2.12130398e-01 -1.03751242e+00 2.71453500e-01 1.27884722e+00 9.24879089e-02 -6.58898592e-01 9.00002182e-01 5.73625267e-01 -1.35225952e-01 -7.81073749e-01 -7.90326655e-01 -4.93964791e-01 1.19126819e-01 -1.74575746e-01 1.13104141e+00 7.02135265e-01 -6.68477416e-01 1.10938370e-01 -5.95098734e-01 -1.77882597e-01 5.66448390e-01 1.29316956e-01 1.11031389e+00 -1.32653475e+00 -6.17493838e-02 -4.36327338e-01 -5.79802632e-01 -9.98560667e-01 5.87079465e-01 -8.49802136e-01 -1.60733059e-01 -1.05959451e+00 -9.78508778e-03 -6.09839320e-01 3.34877193e-01 3.55178088e-01 1.80283375e-02 8.08835089e-01 6.64587915e-02 -4.87603694e-02 1.87054694e-01 7.88347960e-01 1.78871334e+00 3.14757740e-03 2.35099904e-02 1.77050948e-01 -7.45274305e-01 8.56951892e-01 6.40132844e-01 -9.03825648e-03 -5.53330541e-01 -4.41337466e-01 4.54698622e-01 -4.64065261e-02 5.49786091e-01 -9.43905175e-01 -2.59306043e-01 -1.98216200e-01 7.80078292e-01 -7.62859285e-02 7.87305892e-01 -8.21685195e-01 6.46807849e-01 -4.98827659e-02 -1.45318210e-01 2.77286191e-02 1.83691189e-01 1.86949372e-01 -1.09209150e-01 1.57509491e-01 1.19055498e+00 -2.57412434e-01 -4.20488775e-01 8.32344115e-01 2.26081878e-01 -1.07189678e-01 8.66106212e-01 -3.38014454e-01 -1.15397498e-02 -6.01757109e-01 -6.03161991e-01 -4.38930869e-01 6.61771297e-01 5.65123320e-01 6.25686526e-01 -1.65007329e+00 -8.46930444e-01 8.58239055e-01 -7.99102634e-02 6.88875467e-02 3.83620679e-01 2.27454394e-01 -7.36719251e-01 -1.10574059e-01 -5.88846266e-01 -6.46267712e-01 -1.11319149e+00 4.21394587e-01 6.89720213e-01 2.45647684e-01 -4.15454626e-01 8.45392644e-01 5.17354012e-01 -5.28037608e-01 -2.05574587e-01 3.80389020e-02 -7.55804628e-02 -1.67664364e-01 5.67493916e-01 3.84148844e-02 1.58951148e-01 -1.06133258e+00 -3.85265946e-02 1.17998993e+00 4.92305189e-01 -5.22051677e-02 1.36247885e+00 3.11221462e-02 -2.67436445e-01 -3.31859253e-02 1.44108272e+00 1.27465859e-01 -1.29845035e+00 7.91398287e-02 -8.57853174e-01 -7.23323703e-01 -2.16770962e-01 -4.80714113e-01 -1.66940725e+00 1.08174908e+00 4.40737665e-01 1.17336154e-01 1.39233887e+00 -1.79228455e-01 8.40404332e-01 -1.50806084e-01 4.98703927e-01 -7.10623264e-01 1.48888692e-01 2.35305637e-01 1.10451365e+00 -8.09135556e-01 -2.03857556e-01 -6.37312949e-01 -3.05621296e-01 1.26007724e+00 5.55733204e-01 -1.59611762e-01 7.36267924e-01 2.61959970e-01 2.22075507e-01 -1.74440861e-01 -3.41445684e-01 1.54028639e-01 4.42725539e-01 9.55658674e-01 5.20556271e-01 7.98705500e-03 2.71828860e-01 1.77872539e-01 -7.83536971e-01 -1.43431455e-01 3.32254469e-01 3.39851141e-01 -1.00075349e-01 -1.25098610e+00 -3.82336915e-01 -3.78544419e-03 -2.63464421e-01 1.62412927e-01 -4.67601806e-01 7.10083604e-01 3.74244511e-01 6.53355241e-01 9.00040418e-02 -4.37554866e-01 4.61709857e-01 -6.73546223e-03 9.74291980e-01 -4.77882773e-01 -3.36499929e-01 -4.56183404e-02 -1.95795566e-01 -6.30870104e-01 -5.31125247e-01 -3.73152465e-01 -8.09792340e-01 -5.58752060e-01 -3.27519514e-02 -2.04976603e-01 8.24154139e-01 6.25490248e-01 3.88790727e-01 3.53680909e-01 1.17432725e+00 -1.37607896e+00 -3.22951913e-01 -6.60076916e-01 -6.35533929e-01 6.17411911e-01 2.36367255e-01 -7.94554651e-01 -3.59150767e-01 3.21293145e-01]	[12.760024070739746, -0.22108587622642517]
f2670a99-f923-4052-8226-6dbf944ad127	3d-neural-embedding-likelihood-for-robust-sim	2302.03744	null	https://arxiv.org/abs/2302.03744v2	https://arxiv.org/pdf/2302.03744v2.pdf	3D Neural Embedding Likelihood for Robust Probabilistic Inverse Graphics	The ability to perceive and understand 3D scenes is crucial for many applications in computer vision and robotics. Inverse graphics is an appealing approach to 3D scene understanding that aims to infer the 3D scene structure from 2D images. In this paper, we introduce probabilistic modeling to the inverse graphics framework to quantify uncertainty and achieve robustness in 6D pose estimation tasks. Specifically, we propose 3D Neural Embedding Likelihood (3DNEL) as a unified probabilistic model over RGB-D images, and develop efficient inference procedures on 3D scene descriptions. 3DNEL effectively combines learned neural embeddings from RGB with depth information to improve robustness in sim-to-real 6D object pose estimation from RGB-D images. Performance on the YCB-Video dataset is on par with state-of-the-art yet is much more robust in challenging regimes. In contrast to discriminative approaches, 3DNEL's probabilistic generative formulation jointly models multi-object scenes, quantifies uncertainty in a principled way, and handles object pose tracking under heavy occlusion. Finally, 3DNEL provides a principled framework for incorporating prior knowledge about the scene and objects, which allows natural extension to additional tasks like camera pose tracking from video.	['Vikash K. Mansinghka', 'Dileep George', 'Miguel Lázaro-Gredilla', 'Dan Gutfreund', 'Joshua B. Tenenbaum', 'Lirui Wang', 'Nishad Gothoskar', 'Guangyao Zhou']	2023-02-07	null	null	null	null	['pose-tracking', '6d-pose-estimation-1', '6d-pose-estimation']	['computer-vision', 'computer-vision', 'computer-vision']	[ 6.94664717e-02 1.40309677e-01 -5.39942682e-02 -4.29176420e-01 -7.14133203e-01 -7.01319575e-01 8.06689203e-01 -2.15566531e-01 -2.97091573e-01 9.75237116e-02 2.31851131e-01 -1.03106149e-01 -2.14078963e-01 -5.31263590e-01 -9.34920013e-01 -5.31986415e-01 2.28333488e-01 8.53168547e-01 2.13183120e-01 2.81301200e-01 3.24977517e-01 1.07691467e+00 -1.61225677e+00 -1.56300992e-01 1.18304051e-01 1.03514755e+00 3.71328920e-01 8.34323823e-01 -1.08632017e-02 3.69728416e-01 5.61792664e-02 -2.11216360e-01 2.52566397e-01 2.39509970e-01 -3.56095433e-01 6.71682596e-01 7.23540425e-01 -7.84356534e-01 -5.81348419e-01 9.25097287e-01 2.51784235e-01 -4.74070013e-02 1.05412257e+00 -1.32296360e+00 -6.29366279e-01 -2.98831850e-01 -7.17211843e-01 -6.59378767e-02 7.23154366e-01 1.20475568e-01 8.91663492e-01 -1.17674613e+00 5.82333684e-01 1.76299846e+00 6.99363232e-01 4.96589422e-01 -1.42458808e+00 3.54664437e-02 2.70130575e-01 -5.82747906e-03 -1.20056570e+00 -3.04091275e-01 9.29104507e-01 -6.36425078e-01 8.35410297e-01 8.92938375e-02 5.32679200e-01 1.01872408e+00 9.13637206e-02 1.11046863e+00 1.04078197e+00 -3.70731801e-01 1.99368760e-01 1.47240788e-01 3.16097774e-02 6.87645078e-01 4.62146312e-01 2.63842851e-01 -8.81497562e-01 -1.89418569e-01 1.11528420e+00 1.85738370e-01 -2.48417884e-01 -1.26009226e+00 -1.13946402e+00 7.49939442e-01 3.77385587e-01 -4.21882749e-01 -2.31932208e-01 5.15343308e-01 -2.78635025e-02 -5.00076175e-01 5.45581520e-01 1.03982031e-01 -4.62509930e-01 -1.72920167e-01 -3.45093518e-01 3.63839030e-01 7.64868140e-01 1.33838201e+00 8.01558018e-01 -2.41020918e-02 2.73652881e-01 3.12174112e-01 1.08403337e+00 8.85470510e-01 -1.84412763e-01 -1.64493465e+00 1.05811238e-01 2.93502808e-01 1.91119790e-01 -9.24776793e-01 -2.21586362e-01 -1.07850783e-01 -4.00113583e-01 6.31382704e-01 2.46047467e-01 5.18018186e-01 -9.88717914e-01 1.44944501e+00 4.92959857e-01 1.02716200e-02 -8.37046653e-02 1.09757698e+00 7.65226603e-01 5.11533380e-01 -4.20157284e-01 1.91777959e-01 1.30676901e+00 -4.05416727e-01 -3.53828073e-01 -5.32768190e-01 1.38052972e-02 -6.25430107e-01 7.49682546e-01 4.34460938e-01 -9.62163091e-01 -4.82455730e-01 -8.74853551e-01 -5.70186913e-01 -1.91329286e-01 -9.95584801e-02 6.83146715e-01 6.26203597e-01 -1.12995708e+00 3.56983900e-01 -1.16300035e+00 -4.74622935e-01 5.62639773e-01 2.41482556e-01 -5.82792699e-01 -4.86522317e-01 -4.34804708e-01 1.10854077e+00 4.58780348e-01 1.22998767e-01 -1.10432339e+00 -6.94240987e-01 -1.46349549e+00 -4.11618382e-01 3.89767379e-01 -1.05940580e+00 1.26382935e+00 1.86199084e-01 -1.53084564e+00 1.18546069e+00 -2.61939853e-01 -5.77232614e-02 3.76705229e-01 -6.37153506e-01 3.30157042e-01 2.24372298e-01 -2.00817864e-02 9.05873597e-01 1.02721083e+00 -1.67906749e+00 -1.44724339e-01 -8.91558468e-01 1.16958208e-01 5.18711805e-01 3.55826020e-01 -4.84536290e-01 -7.55312204e-01 -1.40356809e-01 8.13369036e-01 -9.59581733e-01 -2.45694548e-01 7.59956241e-01 -3.29917252e-01 -6.88218027e-02 1.06622612e+00 -2.98999041e-01 6.28174320e-02 -2.03338337e+00 4.95482445e-01 -1.31710827e-01 4.66461271e-01 -2.45540708e-01 1.36632577e-01 8.81797895e-02 2.13150457e-01 -1.99863166e-01 -1.94722041e-01 -9.07601535e-01 5.01768053e-01 5.45880556e-01 -3.32926631e-01 9.26363826e-01 4.90362704e-01 8.86140227e-01 -8.02345037e-01 -3.10204595e-01 9.08828914e-01 9.92722988e-01 -8.38591397e-01 4.02851731e-01 -4.06122267e-01 3.83397818e-01 -5.65467358e-01 8.08403611e-01 7.77522266e-01 -1.85979933e-01 -2.91474581e-01 -5.14733076e-01 2.97491122e-02 1.92128882e-01 -1.26513851e+00 2.25547409e+00 -2.61846215e-01 7.13137388e-01 1.31066188e-01 -6.11214936e-01 7.83637166e-01 -7.84303918e-02 3.47802430e-01 -3.13175619e-02 1.29198983e-01 -1.40219107e-01 -7.20942318e-01 -3.90467882e-01 6.01925671e-01 -1.68043375e-01 -2.35953718e-01 4.06509280e-01 2.78946996e-01 -1.24918520e+00 -3.95422339e-01 2.68374085e-01 7.01013267e-01 9.21028793e-01 3.49027544e-01 2.75339391e-02 3.41205820e-02 -7.73258135e-02 2.11757913e-01 6.81563199e-01 -1.99691266e-01 8.39661121e-01 1.35379136e-01 -3.23190272e-01 -1.06785989e+00 -1.72187376e+00 -3.48480850e-01 5.00880361e-01 4.03542399e-01 -2.11730555e-01 -2.48149320e-01 -3.76750499e-01 5.43216646e-01 9.29272950e-01 -4.96209532e-01 -8.39008391e-02 -2.00908616e-01 -4.58341718e-01 7.58490041e-02 5.66923022e-01 2.04773948e-01 -3.10991883e-01 -7.83766031e-01 -9.48305987e-03 1.19113594e-01 -1.42176616e+00 -1.71360970e-01 2.41249561e-01 -1.04439628e+00 -9.50913250e-01 -4.34338093e-01 -1.70654804e-01 4.44755793e-01 5.49556375e-01 1.17500472e+00 -7.32190251e-01 -6.55323029e-01 1.29225111e+00 -1.10157728e-01 -4.91118878e-01 -2.05046639e-01 -7.50543058e-01 4.16571587e-01 -2.88950741e-01 5.47563732e-01 -5.92238665e-01 -4.13008511e-01 2.16028646e-01 -7.02247798e-01 -7.39748776e-02 5.14327109e-01 4.10182327e-01 9.10161018e-01 -1.83264405e-01 -4.69180971e-01 -3.74472827e-01 -1.04132012e-01 -1.32035375e-01 -7.90663123e-01 -1.55712113e-01 -3.38299215e-01 2.46804699e-01 -1.20368317e-01 -2.60305882e-01 -1.06489468e+00 4.20942277e-01 1.62413009e-02 -1.02591550e+00 -5.31892776e-01 -1.57280471e-02 -4.78025705e-01 -8.48473236e-02 4.27625597e-01 1.19023710e-01 1.95307806e-01 -7.56048024e-01 8.21817517e-01 3.02347094e-01 7.17025042e-01 -6.97589874e-01 8.76420259e-01 9.85925555e-01 2.64691383e-01 -7.79850543e-01 -1.00379157e+00 -4.99992728e-01 -1.05798841e+00 -2.44079933e-01 1.09952152e+00 -1.18018055e+00 -9.21213508e-01 2.63194889e-01 -1.43171775e+00 -1.29673565e-02 -3.69457185e-01 7.94308662e-01 -1.19286060e+00 4.69211847e-01 -4.78412002e-01 -1.08015490e+00 4.13506389e-01 -1.24336827e+00 1.97260439e+00 6.80593699e-02 -1.44268557e-01 -1.00912166e+00 -3.59058082e-02 3.25029492e-01 -1.65846422e-01 4.98596370e-01 6.62360668e-01 8.63121077e-02 -1.17186546e+00 -2.12563783e-01 -3.75175893e-01 2.50911415e-01 5.50421663e-02 -6.63850829e-02 -1.36863279e+00 -3.85324024e-02 4.43190783e-01 -3.52295250e-01 9.18951809e-01 6.38285518e-01 9.37346399e-01 1.97031006e-01 -2.79092103e-01 7.31084824e-01 1.44280660e+00 -2.19911963e-01 3.12713504e-01 1.87181890e-01 9.43298459e-01 6.73017979e-01 6.44225180e-01 5.09389102e-01 5.24857819e-01 7.14418948e-01 1.00117266e+00 2.78222054e-01 -1.97524816e-01 -4.64558691e-01 3.23489726e-01 5.49099743e-01 1.97528541e-01 -1.41643167e-01 -8.17316949e-01 2.05254734e-01 -1.62851489e+00 -5.78858078e-01 -1.49555504e-01 1.99213696e+00 4.13848817e-01 2.33393043e-01 -1.64665654e-01 -3.06285662e-03 5.26377022e-01 2.53584623e-01 -8.90219331e-01 -1.93472765e-02 -1.07448161e-01 -2.66025066e-01 5.33378839e-01 6.93385243e-01 -9.15953100e-01 7.42757022e-01 6.64719963e+00 4.99748856e-01 -5.12303174e-01 -7.38333678e-03 2.89882720e-01 -3.16539072e-02 -5.65541863e-01 7.23395050e-02 -1.15370286e+00 -7.50097111e-02 3.50939065e-01 1.20455332e-01 2.40446687e-01 9.50248539e-01 -5.20017883e-03 -3.22684497e-01 -1.63054740e+00 1.44981658e+00 4.36506093e-01 -1.24594080e+00 1.53423041e-01 3.32009405e-01 4.25628662e-01 2.50442147e-01 3.43800694e-01 -4.92733195e-02 4.89648908e-01 -7.68824100e-01 1.07331777e+00 7.53903508e-01 6.71692669e-01 -4.81921107e-01 3.47062409e-01 6.58984423e-01 -8.70215058e-01 2.85611749e-01 -4.90013033e-01 -6.34788647e-02 5.93958795e-01 7.48461604e-01 -9.25552905e-01 3.92646283e-01 9.83108103e-01 1.04627669e+00 -4.01619107e-01 8.20010960e-01 -3.82704109e-01 -4.37612310e-02 -6.73298955e-01 1.53319687e-01 1.01421932e-02 -2.00348243e-01 9.54391897e-01 8.40142667e-01 2.76952416e-01 1.07205652e-01 1.05602272e-01 1.29782534e+00 1.45687237e-01 -7.26463675e-01 -1.04421556e+00 4.96372096e-02 3.06557506e-01 9.65368569e-01 -5.81821144e-01 2.25228630e-02 -2.07179353e-01 1.02675056e+00 2.48463944e-01 2.86181986e-01 -6.10834837e-01 1.41909018e-01 8.23157489e-01 -1.22699201e-01 4.34345961e-01 -9.39008474e-01 -2.41668969e-01 -1.20269239e+00 -6.89717010e-02 -2.85054058e-01 -2.86055207e-02 -1.47437179e+00 -1.32909548e+00 2.34002098e-01 4.88835514e-01 -1.28421962e+00 -3.77658725e-01 -1.48057473e+00 -2.58419327e-02 8.07631731e-01 -1.53895831e+00 -1.13884246e+00 -3.39952141e-01 4.40690190e-01 6.47144556e-01 3.86055470e-01 8.09668660e-01 -3.08657974e-01 -2.69406326e-02 -1.50697649e-01 4.18166537e-03 -2.17889830e-01 4.47312474e-01 -1.45020390e+00 5.13360381e-01 7.37579167e-01 5.72946846e-01 3.77372652e-01 8.83216381e-01 -4.90613341e-01 -2.10578895e+00 -8.07569683e-01 2.17808917e-01 -1.33888233e+00 4.82225239e-01 -7.01998711e-01 -5.53964257e-01 7.98157752e-01 -2.97135264e-01 3.13024461e-01 5.48228264e-01 -4.11701202e-02 -5.96019626e-01 2.73824930e-01 -1.10139692e+00 3.97678941e-01 1.12316370e+00 -9.71174002e-01 -8.46363425e-01 2.46328786e-01 1.03412080e+00 -8.20290625e-01 -9.14229155e-01 5.24165094e-01 5.95925808e-01 -1.04795456e+00 1.36817861e+00 -2.89365351e-01 4.42258008e-02 -5.71350455e-01 -9.83089268e-01 -1.03801882e+00 -2.17599839e-01 -4.03488398e-01 -6.27106786e-01 8.10873508e-01 -1.09868273e-01 -2.03308359e-01 1.03289020e+00 9.67971385e-01 -1.77435011e-01 -3.42802733e-01 -9.07554686e-01 -7.61946321e-01 -2.79204756e-01 -1.11453938e+00 1.01992518e-01 3.66927922e-01 -6.63485169e-01 1.20693207e-01 -1.64183050e-01 6.29915595e-01 1.28380024e+00 2.10992038e-01 1.02938414e+00 -1.27962363e+00 -4.03876215e-01 -3.59034091e-01 -8.53335083e-01 -1.72378600e+00 2.33770743e-01 -6.32337809e-01 3.23376685e-01 -1.52864206e+00 1.65786207e-01 -1.77634671e-01 1.31982729e-01 -1.19668923e-01 1.07853971e-01 3.75280440e-01 2.85150766e-01 1.45733193e-01 -5.93096852e-01 9.38114226e-01 1.15895283e+00 3.26977042e-03 6.65859804e-02 -1.09080762e-01 -4.68282491e-01 1.08739686e+00 1.78153411e-01 -3.38962972e-01 -3.38760942e-01 -8.44928324e-01 9.26434249e-02 -3.95509861e-02 9.33297932e-01 -6.03744566e-01 2.17463508e-01 -2.90515665e-02 7.44369745e-01 -1.21815407e+00 1.16472006e+00 -1.00637901e+00 -1.89487438e-03 1.03841208e-01 -5.27381375e-02 -1.70573980e-01 3.57698411e-01 1.12844419e+00 1.49683803e-01 -1.09867126e-01 5.01226723e-01 -2.57797062e-01 -1.03385258e+00 5.56578338e-01 -1.69789374e-01 -1.34205133e-01 7.73372293e-01 -5.67287028e-01 6.69574589e-02 -4.29234505e-01 -8.14623475e-01 4.91215736e-02 7.94709504e-01 3.81209582e-01 1.09905910e+00 -1.44348669e+00 -5.19102454e-01 3.70793790e-01 3.51310313e-01 5.61774492e-01 7.54799992e-02 5.18555760e-01 -5.59136868e-01 4.26094711e-01 8.52613822e-02 -1.53868258e+00 -9.07436252e-01 5.81257761e-01 1.47015393e-01 3.67615163e-01 -5.56847870e-01 1.13620687e+00 6.00292563e-01 -7.97053456e-01 3.09312344e-01 -5.00538528e-01 2.00156018e-01 -2.84893572e-01 4.14662898e-01 2.99623847e-01 -2.56129056e-01 -9.23044384e-01 -5.34862459e-01 9.87318635e-01 1.01245947e-01 -3.94049674e-01 1.38479221e+00 -5.83012760e-01 -4.11805930e-03 8.64473045e-01 1.39696860e+00 -2.88559169e-01 -2.06029463e+00 -4.19703692e-01 -4.40800861e-02 -8.24125528e-01 4.07128215e-01 -3.01853865e-01 -5.78088999e-01 1.23951602e+00 4.71362352e-01 -2.12355569e-01 6.85820758e-01 5.20808995e-01 1.28187954e-01 5.60379088e-01 6.01482272e-01 -5.33317029e-01 3.19123894e-01 4.97346669e-01 9.38775003e-01 -1.44821644e+00 4.52363372e-01 -4.88598168e-01 -4.37489569e-01 1.26571679e+00 4.17715728e-01 -2.47661516e-01 8.50886464e-01 2.56623596e-01 -2.63174385e-01 -3.58386248e-01 -5.06163001e-01 -2.32169122e-01 6.04871452e-01 8.89177084e-01 -1.28495097e-01 -9.65295359e-02 1.01259685e+00 6.79595619e-02 -6.04191842e-03 -3.21303755e-01 3.84910792e-01 8.62888157e-01 -5.73541462e-01 -7.22974360e-01 -6.37003601e-01 4.48745154e-02 9.67018455e-02 1.92842335e-01 -2.75613427e-01 9.94780362e-01 -9.07686260e-03 6.54196084e-01 2.75007874e-01 -1.30906671e-01 3.07368785e-01 -1.60300627e-01 1.14197803e+00 -8.07386458e-01 6.18268430e-01 4.08376485e-01 -2.49898165e-01 -9.30543125e-01 -7.38479137e-01 -9.75214958e-01 -9.92848992e-01 2.18067062e-03 -3.99861723e-01 -3.51887643e-01 1.10301793e+00 8.88649106e-01 2.33115554e-01 1.57006085e-01 3.59217197e-01 -1.68479967e+00 -4.31353658e-01 -4.98685718e-01 -8.24277520e-01 4.83904332e-02 6.07300043e-01 -1.06084013e+00 -6.53151810e-01 5.19169234e-02]	[7.811493873596191, -2.713285446166992]
ba2be0bd-2a2c-49a4-b295-e84f4f635fcb	flexmatch-boosting-semi-supervised-learning	2110.08263	null	https://arxiv.org/abs/2110.08263v3	https://arxiv.org/pdf/2110.08263v3.pdf	FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling	The recently proposed FixMatch achieved state-of-the-art results on most semi-supervised learning (SSL) benchmarks. However, like other modern SSL algorithms, FixMatch uses a pre-defined constant threshold for all classes to select unlabeled data that contribute to the training, thus failing to consider different learning status and learning difficulties of different classes. To address this issue, we propose Curriculum Pseudo Labeling (CPL), a curriculum learning approach to leverage unlabeled data according to the model's learning status. The core of CPL is to flexibly adjust thresholds for different classes at each time step to let pass informative unlabeled data and their pseudo labels. CPL does not introduce additional parameters or computations (forward or backward propagation). We apply CPL to FixMatch and call our improved algorithm FlexMatch. FlexMatch achieves state-of-the-art performance on a variety of SSL benchmarks, with especially strong performances when the labeled data are extremely limited or when the task is challenging. For example, FlexMatch achieves 13.96% and 18.96% error rate reduction over FixMatch on CIFAR-100 and STL-10 datasets respectively, when there are only 4 labels per class. CPL also significantly boosts the convergence speed, e.g., FlexMatch can use only 1/5 training time of FixMatch to achieve even better performance. Furthermore, we show that CPL can be easily adapted to other SSL algorithms and remarkably improve their performances. We open-source our code at https://github.com/TorchSSL/TorchSSL.	['Takahiro Shinozaki', 'Manabu Okumura', 'Jindong Wang', 'Hao Wu', 'Wenxin Hou', 'Yidong Wang', 'BoWen Zhang']	2021-10-15	null	http://proceedings.neurips.cc/paper/2021/hash/995693c15f439e3d189b06e89d145dd5-Abstract.html	http://proceedings.neurips.cc/paper/2021/file/995693c15f439e3d189b06e89d145dd5-Paper.pdf	neurips-2021-12	['semi-supervised-image-classification']	['computer-vision']	[ 3.47755700e-02 1.49895802e-01 -5.72318733e-01 -6.92774057e-01 -1.02012563e+00 -9.21601176e-01 4.26761061e-01 3.28965992e-01 -5.53791046e-01 7.33334184e-01 -1.07533596e-01 -5.04191399e-01 -4.61879633e-02 -5.42054415e-01 -7.82545686e-01 -7.15825617e-01 7.58208036e-02 6.34087622e-01 4.36694533e-01 2.14026004e-01 1.37338666e-02 1.76282987e-01 -1.47120011e+00 3.63046467e-01 7.77403474e-01 8.48480940e-01 3.53483297e-02 4.38843250e-01 -9.29296389e-02 6.36587381e-01 -2.47506842e-01 -1.68545648e-01 4.22735274e-01 -2.44690433e-01 -1.01186728e+00 1.31926248e-02 7.63512135e-01 1.98905040e-02 -2.40437135e-01 9.61021125e-01 2.73952425e-01 1.85498103e-01 6.17904842e-01 -1.28772461e+00 -3.25122237e-01 9.09191132e-01 -5.14909208e-01 9.22768787e-02 -1.49318829e-01 1.12552270e-01 1.05956447e+00 -1.05807066e+00 4.33180183e-01 1.16912174e+00 5.68528771e-01 7.45225966e-01 -1.42892528e+00 -1.00822866e+00 5.45346439e-01 2.26173908e-01 -1.29837739e+00 -1.43207133e-01 5.77224851e-01 -4.86450672e-01 4.42238420e-01 1.22046776e-01 3.11990052e-01 1.03347123e+00 -4.21219200e-01 1.19405305e+00 1.52199745e+00 -4.69103187e-01 4.24330443e-01 1.95604458e-01 8.80289674e-01 7.22886920e-01 1.00440003e-01 4.28737402e-02 -4.32380617e-01 -1.67231962e-01 4.19548362e-01 3.80521007e-02 -1.48089573e-01 -4.19232517e-01 -1.26977789e+00 8.81490827e-01 4.42951679e-01 -1.48165738e-02 1.99052274e-01 2.65553817e-02 3.81388038e-01 4.41723198e-01 2.77708352e-01 3.90522182e-01 -9.06099081e-01 1.54012010e-01 -1.00507939e+00 9.63391662e-02 6.52262032e-01 9.95723009e-01 9.67779040e-01 -2.20852882e-01 -4.38212454e-01 8.78311515e-01 1.93897516e-01 4.20772195e-01 3.31723034e-01 -1.09050560e+00 5.18017232e-01 5.53108573e-01 -1.47501543e-01 -2.55835801e-01 -3.61050069e-01 -7.75167704e-01 -6.82356417e-01 3.18818279e-02 6.27374291e-01 -1.90177426e-01 -1.33521366e+00 1.92385459e+00 3.10309350e-01 5.40546298e-01 3.01807281e-02 7.58206666e-01 7.26667523e-01 6.03301883e-01 2.27761671e-01 -9.84439999e-02 1.08995783e+00 -1.36415958e+00 -2.68111289e-01 -4.18756247e-01 9.72196996e-01 -6.17985547e-01 1.32482612e+00 5.78673601e-01 -7.89514422e-01 -5.43779910e-01 -1.02337539e+00 1.80589810e-01 -3.59398365e-01 1.91944897e-01 6.87958300e-01 5.83299696e-01 -7.67226696e-01 5.49154758e-01 -8.32225323e-01 1.64931305e-02 6.04100108e-01 4.86413151e-01 -1.07551984e-01 -4.44822788e-01 -1.05360889e+00 5.25434077e-01 6.84618771e-01 -2.58073866e-01 -1.02148438e+00 -9.97290254e-01 -7.08522916e-01 8.40429589e-02 6.12316072e-01 -2.40807086e-01 1.54998636e+00 -1.04294813e+00 -1.24951756e+00 8.17809701e-01 -3.34748067e-02 -4.70143586e-01 5.41794479e-01 -1.43341660e-01 -3.45631868e-01 -3.58653851e-02 -8.46517924e-03 1.00497699e+00 5.02078891e-01 -1.14803290e+00 -7.97379375e-01 -4.41617630e-02 -1.78208407e-02 1.13273017e-01 -4.65955466e-01 -2.46453539e-01 -6.99240625e-01 -5.93457818e-01 1.21199071e-01 -1.12908387e+00 -3.51100236e-01 -9.53965560e-02 -2.94743568e-01 -5.28390050e-01 7.79687941e-01 1.37137696e-01 1.25233459e+00 -2.11439013e+00 -2.23981723e-01 2.06229985e-01 3.18012208e-01 7.71900058e-01 -4.45838571e-01 1.15438186e-01 -8.64052624e-02 1.29761860e-01 -4.82761860e-01 -4.05270964e-01 1.23506173e-01 3.53760660e-01 -2.29129329e-01 3.11541706e-01 1.14883937e-01 7.65039325e-01 -1.11976743e+00 -4.29270208e-01 1.99458897e-01 2.39637613e-01 -6.95271194e-01 5.59066907e-02 -3.81642133e-01 4.21416402e-01 -4.02112395e-01 5.82041085e-01 6.98194981e-01 -5.83176136e-01 2.06994921e-01 2.22301427e-02 9.93415788e-02 3.91300470e-01 -1.37234855e+00 1.64405227e+00 -3.73196304e-01 5.26802421e-01 -2.74199009e-01 -1.12839949e+00 8.91511858e-01 2.24490613e-01 2.66860813e-01 -5.72034776e-01 2.37746071e-02 3.81840974e-01 -1.20376468e-01 -7.46010020e-02 5.80468588e-02 1.10757917e-01 -1.15206994e-01 5.57836056e-01 2.55610615e-01 3.33541811e-01 3.94105971e-01 2.52744943e-01 1.00688672e+00 7.50483125e-02 1.74552888e-01 -3.94552708e-01 6.21780694e-01 -3.15730018e-03 8.87162387e-01 9.60658729e-01 -2.45343089e-01 6.10638320e-01 4.31004226e-01 -3.56449604e-01 -6.68720245e-01 -1.24879348e+00 -3.33040506e-01 1.35418701e+00 8.42508599e-02 -3.55433494e-01 -7.36500800e-01 -1.33064473e+00 1.57964993e-02 6.79966688e-01 -6.13710165e-01 -1.57370657e-01 -4.91489679e-01 -8.17329764e-01 3.82934928e-01 5.46925426e-01 5.55744469e-01 -9.69630182e-01 -2.12944776e-01 1.35278985e-01 5.45080304e-02 -1.11826384e+00 -5.32325745e-01 5.44774890e-01 -8.99835706e-01 -1.19335639e+00 -4.98465329e-01 -1.00245047e+00 1.00219631e+00 2.44342715e-01 1.19126713e+00 5.32831550e-02 2.32377835e-02 6.40257969e-02 -4.34980154e-01 -1.99191168e-01 -3.95034611e-01 5.17663956e-01 -7.01811165e-02 -4.30814624e-02 4.96420175e-01 -3.76606315e-01 -5.60998440e-01 6.19882047e-01 -7.55420506e-01 1.53784677e-01 4.32569742e-01 9.43949938e-01 7.94535637e-01 -1.75734982e-01 7.71063805e-01 -1.56778848e+00 6.68640956e-02 -6.77188337e-01 -6.39899194e-01 3.51175845e-01 -9.90231693e-01 2.05504492e-01 7.02040136e-01 -7.26500511e-01 -8.03870678e-01 3.36795330e-01 -6.32504150e-02 -5.73587954e-01 -2.17089653e-01 4.99674261e-01 2.12836806e-02 9.66208503e-02 7.37255454e-01 -1.41495643e-02 -3.68821025e-01 -6.56901121e-01 2.72846818e-01 7.56855726e-01 5.22598565e-01 -7.51329184e-01 7.87358344e-01 2.30996579e-01 -2.75049508e-01 -2.77488649e-01 -1.58905196e+00 -7.22385108e-01 -6.06272221e-01 3.33464285e-03 3.12251002e-01 -1.05668330e+00 -6.68154120e-01 3.42529684e-01 -5.11581957e-01 -8.98421705e-01 -2.55693495e-01 4.80253547e-01 -3.33633542e-01 9.38810408e-02 -6.22733414e-01 -3.77727509e-01 -2.73254126e-01 -1.27041066e+00 7.06115723e-01 3.23423117e-01 -2.59830593e-03 -1.07157099e+00 -2.83710603e-02 3.21798980e-01 1.30027160e-01 -5.04854061e-02 7.46107519e-01 -1.06102276e+00 -5.31621993e-01 5.18570887e-03 -1.47304654e-01 5.64272225e-01 5.45781106e-03 -2.24230111e-01 -1.27034616e+00 -6.48822904e-01 -4.16507632e-01 -8.28054309e-01 1.17443454e+00 2.34329775e-01 1.41302788e+00 -1.73439845e-01 -3.22924644e-01 6.38794482e-01 1.48328435e+00 -5.16764994e-05 2.86039412e-01 2.47117445e-01 7.01345325e-01 4.61933613e-01 9.68795300e-01 2.14447111e-01 3.31364095e-01 4.80089784e-01 2.36395836e-01 -3.13987702e-01 -3.09222817e-01 -3.64340991e-01 4.28337008e-01 8.33692670e-01 4.90295917e-01 -1.66501492e-01 -1.05739653e+00 4.62835222e-01 -1.91469705e+00 -6.13950014e-01 -2.60290831e-01 2.44925618e+00 1.17992890e+00 4.29825306e-01 6.94248592e-03 2.89532989e-01 6.11640453e-01 -5.35867661e-02 -7.49979317e-01 -1.65485784e-01 3.70676517e-02 4.78280038e-01 7.22675145e-01 6.11408234e-01 -1.35814309e+00 1.17471933e+00 5.96358776e+00 1.10296738e+00 -1.12547481e+00 2.60935754e-01 9.05847013e-01 -1.70347825e-01 -2.44419605e-01 1.73009411e-02 -1.24793732e+00 4.76506948e-01 1.02874959e+00 3.09078470e-02 2.89500833e-01 9.06782269e-01 -3.27176303e-02 -7.30456337e-02 -1.21314549e+00 8.00127327e-01 -1.02626413e-01 -1.36467373e+00 -1.37099057e-01 -2.73796380e-01 9.84693229e-01 4.60846096e-01 3.01908460e-02 7.54979551e-01 6.51345134e-01 -8.57208550e-01 6.58205807e-01 -7.19125867e-02 8.05879354e-01 -7.99729705e-01 5.97124040e-01 5.25481939e-01 -1.10737336e+00 -1.72295734e-01 -3.80114168e-01 1.52832285e-01 -2.70370156e-01 6.75076067e-01 -9.02757764e-01 2.51583993e-01 5.90193629e-01 7.55414903e-01 -8.34244251e-01 1.13738143e+00 -5.44259310e-01 1.28714323e+00 -3.94484818e-01 1.36467442e-01 4.65730429e-01 9.76459309e-02 8.38861540e-02 1.39222193e+00 -4.49199416e-02 -3.38880159e-02 6.15419388e-01 5.79596758e-01 -2.34240189e-01 1.10579459e-02 -9.34083089e-02 2.71732152e-01 8.49867821e-01 1.21226227e+00 -7.76978850e-01 -5.79899848e-01 -4.53483760e-01 5.37147641e-01 5.53272784e-01 4.18042541e-01 -8.95806670e-01 -2.24571094e-01 5.29442847e-01 9.11318585e-02 3.05030912e-01 2.79672146e-02 -2.08031818e-01 -1.07329476e+00 -1.54264048e-01 -9.01235461e-01 7.87252009e-01 -3.47575992e-01 -1.27066481e+00 5.06179452e-01 5.57873957e-02 -1.36305034e+00 7.69696012e-02 -5.61519742e-01 -3.80465180e-01 5.82390487e-01 -1.89577723e+00 -9.52126026e-01 -2.06918344e-01 6.17186546e-01 7.93901563e-01 -9.10182521e-02 7.38671482e-01 3.57239872e-01 -7.32754230e-01 9.67579246e-01 2.38914356e-01 2.23461658e-01 1.04735291e+00 -1.39065838e+00 3.93709928e-01 7.74555027e-01 2.92931020e-01 4.40887809e-01 3.65033180e-01 -3.10547382e-01 -9.83757198e-01 -1.40420866e+00 9.11958992e-01 -3.30401480e-01 5.36341488e-01 -5.29529452e-01 -1.06363487e+00 8.77703786e-01 -6.97910041e-02 4.87356275e-01 8.18409145e-01 2.55827427e-01 -7.44373322e-01 -1.08961999e-01 -9.97307479e-01 4.71177131e-01 9.50742185e-01 -3.12838763e-01 -4.64142412e-01 4.22555983e-01 8.79412472e-01 -5.40671408e-01 -7.66087711e-01 5.72403491e-01 2.07469642e-01 -8.08191895e-01 8.00424874e-01 -5.28216839e-01 2.02558190e-01 -4.53112990e-01 -8.50502551e-02 -1.11975145e+00 -3.54123771e-01 -5.18975675e-01 -2.54244357e-01 1.33087063e+00 7.60740459e-01 -7.63164222e-01 1.00634623e+00 4.82049376e-01 -2.24129513e-01 -9.39573348e-01 -4.98174191e-01 -9.20651138e-01 2.96986610e-01 -5.23875177e-01 4.58978832e-01 1.16432714e+00 -1.94018066e-01 3.35122168e-01 -9.13682058e-02 1.96929321e-01 8.93695176e-01 3.93779337e-01 5.70039988e-01 -1.34970105e+00 -4.49429452e-01 -3.36587667e-01 1.72192812e-01 -1.17946589e+00 3.70307088e-01 -1.40355480e+00 -2.94733979e-02 -1.06349111e+00 3.70902240e-01 -1.20287311e+00 -7.49851584e-01 1.05181944e+00 -4.26535547e-01 3.81829202e-01 3.72722566e-01 2.39820987e-01 -8.18616986e-01 3.10794771e-01 1.22046852e+00 -1.55777812e-01 -2.58194625e-01 2.47812182e-01 -6.72836840e-01 6.51076436e-01 1.11496460e+00 -7.58173883e-01 -7.09112942e-01 -3.38079453e-01 -8.20853561e-02 -2.44376048e-01 1.04518376e-01 -1.00900638e+00 2.87832946e-01 -2.24171355e-01 2.33976066e-01 -5.17241180e-01 -8.08115602e-02 -5.60781598e-01 -1.47660539e-01 4.86911416e-01 -7.74091840e-01 -3.90569538e-01 2.14415133e-01 5.13133287e-01 -2.18878567e-01 -3.87085140e-01 1.12480783e+00 5.80210090e-02 -7.22147286e-01 5.58058739e-01 -1.60804138e-01 4.84329730e-01 9.66384172e-01 4.86680120e-02 -4.10370469e-01 1.49232388e-01 -8.17205727e-01 7.18066633e-01 4.13831949e-01 3.71250361e-01 2.78531790e-01 -1.19962120e+00 -6.53444588e-01 2.66484946e-01 2.52659500e-01 3.41866523e-01 7.99250752e-02 7.14064896e-01 -3.00146341e-01 4.80141789e-01 1.79098055e-01 -7.94640958e-01 -1.23005605e+00 5.76021910e-01 2.52026707e-01 -4.68574673e-01 -5.35858154e-01 8.90914738e-01 3.27455819e-01 -8.60426366e-01 7.05830097e-01 -2.13009253e-01 -4.01092656e-02 -6.13141358e-02 5.40386558e-01 2.03546986e-01 7.39487931e-02 -1.73721500e-02 -2.30489984e-01 4.76001054e-01 -3.81582946e-01 1.03179999e-01 1.16657174e+00 1.38588443e-01 4.36184019e-01 3.89148623e-01 1.29778099e+00 -1.98082924e-01 -1.68208981e+00 -6.90733731e-01 2.76585191e-01 -1.72050118e-01 1.90089360e-01 -1.05740392e+00 -1.33583808e+00 8.91992331e-01 5.47816277e-01 -3.23225349e-01 1.16377318e+00 4.78537269e-02 6.28117919e-01 3.82899106e-01 4.42205578e-01 -1.11281896e+00 6.55071512e-02 5.96941292e-01 3.32124829e-01 -1.44121253e+00 -4.98936996e-02 -5.48142016e-01 -6.83167517e-01 8.58499229e-01 8.00987959e-01 -3.94669827e-03 7.23437071e-01 2.71456182e-01 2.02952921e-01 7.46367052e-02 -1.02512908e+00 -2.14433596e-01 1.81374505e-01 2.77906597e-01 3.89141709e-01 1.71110079e-01 -2.62274623e-01 5.03816307e-01 6.93417788e-02 -2.73205508e-02 3.76847118e-01 9.19225931e-01 -5.22584975e-01 -1.51077163e+00 -1.92277193e-01 4.43707913e-01 -2.86907464e-01 -2.02605739e-01 -6.37525246e-02 6.82039201e-01 1.58979580e-01 8.62494707e-01 -6.50389027e-03 -2.42070705e-01 1.36248752e-01 2.08869725e-01 3.08003098e-01 -9.15813267e-01 -7.34862149e-01 7.23579898e-02 -1.13176756e-01 -4.54127789e-01 -4.79298204e-01 -6.28981054e-01 -1.64474487e+00 -7.31011331e-02 -2.64630705e-01 2.94613749e-01 3.63333970e-01 8.30819309e-01 2.34836116e-01 3.13209802e-01 8.02698851e-01 -3.33900124e-01 -6.28355086e-01 -7.85920680e-01 -3.98565620e-01 3.17917973e-01 2.38604426e-01 -6.69249117e-01 -4.29975748e-01 7.08858222e-02]	[9.493011474609375, 3.5035293102264404]
91c9d8fd-98cd-4bb4-a277-39bd342ee258	moral-machine-or-tyranny-of-the-majority	2305.17319	null	https://arxiv.org/abs/2305.17319v1	https://arxiv.org/pdf/2305.17319v1.pdf	Moral Machine or Tyranny of the Majority?	With Artificial Intelligence systems increasingly applied in consequential domains, researchers have begun to ask how these systems ought to act in ethically charged situations where even humans lack consensus. In the Moral Machine project, researchers crowdsourced answers to "Trolley Problems" concerning autonomous vehicles. Subsequently, Noothigattu et al. (2018) proposed inferring linear functions that approximate each individual's preferences and aggregating these linear models by averaging parameters across the population. In this paper, we examine this averaging mechanism, focusing on fairness concerns in the presence of strategic effects. We investigate a simple setting where the population consists of two groups, with the minority constituting an {\alpha} < 0.5 share of the population. To simplify the analysis, we consider the extreme case in which within-group preferences are homogeneous. Focusing on the fraction of contested cases where the minority group prevails, we make the following observations: (a) even when all parties report their preferences truthfully, the fraction of disputes where the minority prevails is less than proportionate in {\alpha}; (b) the degree of sub-proportionality grows more severe as the level of disagreement between the groups increases; (c) when parties report preferences strategically, pure strategy equilibria do not always exist; and (d) whenever a pure strategy equilibrium exists, the majority group prevails 100% of the time. These findings raise concerns about stability and fairness of preference vector averaging as a mechanism for aggregating diverging voices. Finally, we discuss alternatives, including randomized dictatorship and median-based mechanisms.	['Zachary C. Lipton', 'Hoda Heidari', 'Michael Feffer']	2023-05-27	null	null	null	null	['autonomous-vehicles']	['computer-vision']	[-8.22340101e-02 5.02365649e-01 -4.57080126e-01 -1.27060741e-01 -3.13316703e-01 -1.12276161e+00 7.04618394e-01 2.99336389e-02 -8.69373143e-01 1.09595823e+00 5.13059258e-01 -5.60596108e-01 -1.65830851e-01 -6.02627933e-01 -4.57599401e-01 -8.30932796e-01 3.34844798e-01 5.42999208e-01 -3.02228779e-01 -2.32678279e-01 2.57139653e-01 9.75698158e-02 -1.19364440e+00 -1.05248895e-02 1.28733087e+00 8.36626768e-01 -4.16886032e-01 3.87298495e-01 2.25645497e-01 1.11295950e+00 -7.37270892e-01 -7.61300087e-01 8.95167470e-01 -2.24805191e-01 -5.90158701e-01 -1.14959642e-01 1.15965255e-01 -3.76390845e-01 -9.51109529e-02 1.22443449e+00 3.97734016e-01 1.70586810e-01 8.67594123e-01 -1.65225708e+00 -7.49542892e-01 9.79798138e-01 -7.94679284e-01 -1.60755023e-01 1.47249803e-01 5.62078238e-01 1.27222896e+00 -1.20901734e-01 6.25768423e-01 1.18464470e+00 2.68191814e-01 7.36695766e-01 -1.30054164e+00 -6.93567216e-01 3.77333969e-01 -3.84817496e-02 -1.15603614e+00 -4.31301832e-01 6.65218353e-01 -6.17599070e-01 2.45614588e-01 6.38787985e-01 5.92921555e-01 1.02025187e+00 1.47227272e-01 5.07110894e-01 1.42155027e+00 -4.38160077e-02 6.14307702e-01 1.33037642e-01 1.82204768e-01 -1.31690830e-01 1.09617889e+00 -4.87944996e-03 -4.48693961e-01 -7.72425652e-01 2.24605784e-01 -3.34342539e-01 -1.57191038e-01 -2.10385233e-01 -9.00918365e-01 1.05698109e+00 1.89300805e-01 1.44327655e-01 -7.92653859e-01 -2.09105890e-02 2.33094975e-01 3.21994901e-01 3.98003459e-01 6.92422986e-01 -2.03552485e-01 -8.00291747e-02 -8.15398812e-01 8.22143734e-01 1.03385210e+00 4.18836176e-01 4.96560305e-01 -2.37015247e-01 -2.22535565e-01 4.21892554e-01 -7.50639662e-02 4.43404287e-01 1.26858473e-01 -1.75026166e+00 6.18488491e-01 5.16036093e-01 9.29638445e-01 -9.93903160e-01 -2.74045438e-01 -1.76777944e-01 -6.33195937e-01 2.96917975e-01 1.02394998e+00 -7.15247214e-01 -1.99287862e-01 2.25162530e+00 2.60592788e-01 -5.93277335e-01 1.45243749e-01 1.21410549e+00 7.99591467e-02 4.84667242e-01 2.65017778e-01 -5.52258968e-01 1.35486674e+00 -3.16703022e-01 -5.10971785e-01 -2.10542381e-01 4.47894216e-01 -4.78890061e-01 7.81699836e-01 1.98079303e-01 -1.09058928e+00 2.68257082e-01 -6.45650685e-01 2.23356709e-01 1.31834418e-01 -4.14938003e-01 4.81734246e-01 7.03626752e-01 -8.08953702e-01 3.00268829e-01 -2.31114939e-01 -1.67779177e-01 5.38358808e-01 2.55460739e-01 -2.65097558e-01 2.25110009e-01 -1.22462213e+00 9.99755979e-01 -2.85300136e-01 -4.49046493e-02 -4.02225763e-01 -7.07920134e-01 -3.58998418e-01 2.80055106e-01 6.41497254e-01 -7.48117507e-01 1.07590663e+00 -1.47467780e+00 -1.16459596e+00 8.45053017e-01 -8.56298506e-02 -4.85995859e-01 1.10728979e+00 1.09539032e-01 7.61787072e-02 -2.99192399e-01 6.08249187e-01 4.00853992e-01 8.61373618e-02 -1.38793552e+00 -6.83734715e-01 -6.52938426e-01 5.37712574e-01 5.24241626e-01 3.47827673e-02 1.43133953e-01 5.75712919e-01 -1.16924062e-01 -3.53843331e-01 -1.17734969e+00 -2.53892809e-01 -1.34357035e-01 -4.84491378e-01 -5.00747561e-01 3.42465222e-01 -1.75094590e-01 1.10657477e+00 -1.97559834e+00 -2.36810565e-01 3.42502803e-01 4.43904400e-01 -4.88412268e-02 7.15903863e-02 2.44591251e-01 2.64049143e-01 5.43652654e-01 -2.72445619e-01 -7.97985345e-02 5.59820235e-01 1.53734043e-01 -4.13111746e-01 7.16302633e-01 -2.02224046e-01 6.68154299e-01 -6.39526069e-01 -2.64941275e-01 -2.70100713e-01 -2.25119576e-01 -7.62655675e-01 -1.55659944e-01 3.24311182e-02 4.21836913e-01 -3.98603886e-01 3.26464206e-01 7.52948284e-01 -2.24210881e-02 5.51602006e-01 2.60559648e-01 -4.75898236e-01 2.76374072e-01 -1.16498518e+00 5.62758565e-01 6.06741272e-02 6.64595485e-01 7.17839837e-01 -8.70098770e-01 3.21737170e-01 1.44162267e-01 3.67359430e-01 -5.07597625e-01 5.32656550e-01 3.66690755e-01 4.49720204e-01 -2.96669841e-01 6.08398318e-01 -5.63386559e-01 -5.33008754e-01 8.88525665e-01 -6.45602405e-01 -1.13702290e-01 1.47495151e-01 2.64522851e-01 9.78590131e-01 -4.83849674e-01 5.64434350e-01 -6.02174640e-01 1.39914513e-01 1.97976351e-01 1.11701202e+00 9.50655341e-01 -7.47077703e-01 3.79609734e-01 1.22588134e+00 -2.99618423e-01 -1.02962351e+00 -9.28594589e-01 1.55363590e-01 1.11570728e+00 5.05455613e-01 1.72633156e-01 -9.14504647e-01 -4.58467335e-01 4.75986004e-01 1.13077319e+00 -8.68250787e-01 -3.66481431e-02 -2.62329489e-01 -7.45904207e-01 4.29964006e-01 1.31253511e-01 5.81939042e-01 -6.95028186e-01 -9.93320525e-01 -3.84561449e-01 -5.83033264e-01 -8.54390562e-01 -7.30812550e-01 -3.47079933e-01 -2.35399768e-01 -1.11332202e+00 -5.91593444e-01 1.65305451e-01 5.96200228e-01 2.52330035e-01 7.98919559e-01 -1.35680735e-01 3.10138077e-01 3.20397049e-01 -1.09121082e-02 -6.20481312e-01 -3.78171742e-01 -1.80821344e-01 3.55510414e-01 1.35748729e-01 5.91161251e-01 -3.11164320e-01 -8.09997737e-01 4.15338874e-01 -6.75198734e-01 -3.01946044e-01 2.15673055e-02 3.92153054e-01 -1.01619206e-01 -3.65940571e-01 7.46355116e-01 -9.01554286e-01 1.04575217e+00 -9.18925941e-01 -6.99073195e-01 1.55398160e-01 -4.43542093e-01 -1.48464873e-01 5.59623897e-01 -5.68189979e-01 -1.09612596e+00 -4.35496211e-01 6.17065549e-01 4.80270274e-02 1.03207096e-01 3.57475936e-01 -2.51804590e-01 2.96436191e-01 6.97804093e-01 -3.75201225e-01 3.30051988e-01 -3.87412794e-02 3.77127707e-01 9.60707545e-01 2.42620334e-01 -9.34014082e-01 5.91545105e-01 5.76969206e-01 -2.79684514e-01 -6.35575712e-01 -4.25946712e-01 1.67953521e-01 6.92234486e-02 -3.30670178e-01 8.52019668e-01 -8.13394785e-01 -1.53562939e+00 2.03804329e-01 -1.27222693e+00 -3.25185269e-01 -5.38088322e-01 6.10711575e-01 -5.98118544e-01 2.27538064e-01 -3.42323720e-01 -1.43279803e+00 5.43023981e-02 -1.22855759e+00 4.22203422e-01 5.09300232e-01 -7.27183282e-01 -5.37394583e-01 -1.78956985e-03 7.92773843e-01 5.55739045e-01 1.02771819e-01 6.68721974e-01 -8.99990559e-01 -2.85793424e-01 -9.11759436e-02 2.37005223e-02 -6.96799234e-02 8.18827823e-02 3.33813839e-02 -6.28714561e-01 7.61956498e-02 7.07279816e-02 -1.57382980e-01 3.37735504e-01 5.66284299e-01 5.62851787e-01 -9.74958301e-01 -1.80061743e-01 -1.75832525e-01 7.63962626e-01 3.50723803e-01 1.63605958e-01 2.56552875e-01 1.60204694e-01 1.30265701e+00 2.83590376e-01 7.03251541e-01 9.93745506e-01 5.97371936e-01 3.41456771e-01 3.92046183e-01 4.92385805e-01 -1.69326380e-01 4.41145897e-01 -1.18321052e-03 -3.96850109e-01 -4.46163952e-01 -6.12575293e-01 6.44374728e-01 -2.00656581e+00 -1.28636742e+00 1.28482997e-01 2.39390826e+00 8.16755712e-01 2.38967761e-02 6.81689858e-01 -1.44174755e-01 1.02673876e+00 1.23464711e-01 -6.63036883e-01 -7.22082257e-01 -4.13642466e-01 -4.98419642e-01 8.71827126e-01 7.96793878e-01 -6.11710310e-01 6.84722304e-01 6.07492208e+00 6.37563884e-01 -9.45591986e-01 1.41833410e-01 1.02320910e+00 -7.05575287e-01 -7.77148306e-01 3.19744349e-01 -3.08022410e-01 8.05386662e-01 6.03783369e-01 -8.92651737e-01 4.18862313e-01 4.86771584e-01 4.63949621e-01 -6.20163321e-01 -8.39246452e-01 4.76428360e-01 -2.21535489e-01 -1.07230949e+00 -3.28691512e-01 4.75161076e-01 8.89870167e-01 -1.14050120e-01 2.62838006e-01 -1.23485783e-02 1.24289870e+00 -1.14693141e+00 1.18188167e+00 2.64816672e-01 4.63946432e-01 -7.73650765e-01 6.28564119e-01 7.32437372e-01 -4.16356087e-01 -5.33650815e-01 -1.77354097e-01 -6.83395743e-01 2.59467781e-01 6.61208034e-01 -2.00884596e-01 2.45356575e-01 4.71322894e-01 -3.09195489e-01 1.25405431e-01 5.31994820e-01 -8.95267725e-02 5.84376156e-01 -5.19801795e-01 -8.76996666e-02 3.49436283e-01 -4.77730811e-01 6.89422250e-01 4.94959623e-01 1.17229901e-01 6.47947967e-01 -1.83227152e-01 1.17893970e+00 -2.51035750e-01 -8.97243917e-02 -7.23743677e-01 8.06406792e-03 9.02445674e-01 1.08443308e+00 -5.24179041e-01 -1.82464227e-01 -2.34144151e-01 3.27373475e-01 2.22119726e-02 5.44630289e-01 -6.44939661e-01 5.15672788e-02 1.03031468e+00 1.91555172e-01 -3.02041173e-01 1.38765108e-02 -8.64423454e-01 -1.14424860e+00 1.30905941e-01 -8.90409946e-01 3.59184533e-01 -5.01913548e-01 -1.34068716e+00 5.44573814e-02 5.88034913e-02 -8.19928586e-01 -2.67749637e-01 -2.87978828e-01 -8.36850584e-01 8.04198861e-01 -9.40102160e-01 -5.42897761e-01 2.70338029e-01 1.81985438e-01 8.25744793e-02 -1.74653716e-02 2.05612496e-01 -7.06134588e-02 -5.10031164e-01 5.43235779e-01 -2.50411164e-02 -4.33380492e-02 4.69682217e-01 -8.85187566e-01 -1.88461542e-01 6.72891319e-01 -3.01477879e-01 9.06584859e-01 1.09017980e+00 -5.77570736e-01 -1.14211833e+00 -6.80479288e-01 1.14874244e+00 -5.47342539e-01 5.45197964e-01 -3.00840706e-01 -4.90559042e-01 6.76759124e-01 3.04146916e-01 -1.17003538e-01 8.03886414e-01 3.80238332e-02 -3.90588075e-01 -7.92906880e-02 -1.60864067e+00 1.07795441e+00 8.51971209e-01 -3.89623046e-01 -4.76741612e-01 6.58414736e-02 3.35022241e-01 1.26734346e-01 -4.19426173e-01 2.15925246e-01 7.51038253e-01 -1.08057320e+00 3.03685069e-01 -6.16292953e-01 4.47313815e-01 -2.94333339e-01 -4.19389278e-01 -1.34300876e+00 -8.45988542e-02 -8.32928836e-01 5.59293568e-01 1.14853108e+00 4.66532141e-01 -1.09449756e+00 6.64253235e-01 1.38218033e+00 4.89664048e-01 -4.90028560e-01 -1.47895825e+00 -5.54754317e-01 8.27063560e-01 1.28247049e-02 5.40181160e-01 1.04502189e+00 4.54740047e-01 1.21163771e-01 -3.13141346e-01 -1.88036025e-01 6.82734311e-01 1.28037572e-01 6.82147563e-01 -1.33409870e+00 5.83194848e-03 -8.09099317e-01 4.93648574e-02 -6.95454001e-01 4.42071557e-01 -4.75898683e-01 7.80243287e-03 -1.24460721e+00 4.64288831e-01 -5.36415458e-01 1.00101054e-01 2.26559415e-01 -1.88027412e-01 -4.24694419e-02 6.76359296e-01 3.81246358e-01 -4.94867921e-01 2.84501553e-01 8.34884286e-01 -1.68440402e-01 -8.62095803e-02 2.54224055e-02 -1.71842182e+00 7.11010575e-01 7.05604017e-01 -2.22062826e-01 -2.92921156e-01 -1.10385761e-01 4.66126263e-01 1.80845484e-01 4.10051674e-01 -2.90813088e-01 4.33025599e-01 -8.53830695e-01 -8.12445432e-02 3.48637737e-02 -1.34708788e-02 -6.68076873e-01 4.63018775e-01 3.61875504e-01 -5.38705528e-01 -9.21729133e-02 -2.27173850e-01 3.55893761e-01 1.56911254e-01 -1.10640330e-03 7.69474566e-01 -9.36773643e-02 2.57668197e-01 -3.95676084e-02 -9.51036692e-01 2.54887044e-01 1.34125233e+00 -2.38336623e-01 -6.50653958e-01 -8.88898432e-01 -3.38113666e-01 6.37970150e-01 9.07963336e-01 2.04287812e-01 -1.42348513e-01 -1.18498886e+00 -9.88699079e-01 -3.71775150e-01 -1.01876900e-01 -2.38192976e-01 3.56183439e-01 1.04563808e+00 -8.18789527e-02 2.77016371e-01 -1.70032591e-01 4.01943885e-02 -1.01881337e+00 2.59647518e-01 6.06629968e-01 8.44437182e-02 1.32513925e-01 5.33889353e-01 7.06419468e-01 -4.57991600e-01 -2.41621405e-01 2.18560115e-01 -1.20003670e-01 5.06561756e-01 1.94504470e-01 7.29435146e-01 -5.10061204e-01 -1.04541206e+00 -5.96180618e-01 3.17410767e-01 1.68449134e-01 -5.74032485e-01 1.06769598e+00 -3.39605659e-01 -3.18794519e-01 3.87952000e-01 5.44367135e-01 4.79874790e-01 -1.20623648e+00 1.89310879e-01 -1.56862780e-01 -4.76412088e-01 -5.13194740e-01 -8.48479748e-01 -9.00109231e-01 3.53414446e-01 -2.29712471e-01 6.54728591e-01 6.45553410e-01 -1.73294991e-02 8.72727484e-02 6.84208125e-02 6.18957400e-01 -1.24367571e+00 -5.64022601e-01 3.33507031e-01 8.68359447e-01 -9.72102344e-01 -4.86604460e-02 -5.76352216e-02 -1.22945094e+00 3.91539544e-01 5.34003496e-01 -2.53793865e-01 2.69901365e-01 2.72847526e-02 1.78980589e-01 9.99797210e-02 -9.46193397e-01 -2.62323003e-02 -9.21827629e-02 4.37634200e-01 3.69028389e-01 7.49049962e-01 -1.30926633e+00 1.25001299e+00 -6.29047632e-01 -1.65298939e-01 1.08530354e+00 5.45662582e-01 -4.11805063e-01 -8.31523359e-01 -6.17331386e-01 6.45059586e-01 -6.34139299e-01 1.13521121e-01 -9.54022467e-01 6.89437032e-01 2.34471947e-01 1.38553214e+00 3.06863934e-01 1.02379387e-02 3.21761310e-01 -7.67015964e-02 1.15383230e-01 -2.17668146e-01 -6.15399063e-01 -2.69729555e-01 2.22039148e-01 -4.92303967e-01 -3.21662843e-01 -8.89018953e-01 -9.56461668e-01 -9.43742156e-01 -2.40855157e-01 4.86921638e-01 1.71308383e-01 9.35575962e-01 4.52457428e-01 -3.13135743e-01 6.46747410e-01 -5.85776746e-01 -9.06192005e-01 -7.58205175e-01 -7.50857413e-01 3.17266673e-01 3.80793631e-01 -6.49185538e-01 -9.44243073e-01 -4.95006084e-01]	[8.812207221984863, 5.42784309387207]
34ccb50a-a9ac-47c1-98b6-417f18c5b1d5	structural-neural-additive-models-enhanced	2302.09275	null	https://arxiv.org/abs/2302.09275v1	https://arxiv.org/pdf/2302.09275v1.pdf	Structural Neural Additive Models: Enhanced Interpretable Machine Learning	Deep neural networks (DNNs) have shown exceptional performances in a wide range of tasks and have become the go-to method for problems requiring high-level predictive power. There has been extensive research on how DNNs arrive at their decisions, however, the inherently uninterpretable networks remain up to this day mostly unobservable "black boxes". In recent years, the field has seen a push towards interpretable neural networks, such as the visually interpretable Neural Additive Models (NAMs). We propose a further step into the direction of intelligibility beyond the mere visualization of feature effects and propose Structural Neural Additive Models (SNAMs). A modeling framework that combines classical and clearly interpretable statistical methods with the predictive power of neural applications. Our experiments validate the predictive performances of SNAMs. The proposed framework performs comparable to state-of-the-art fully connected DNNs and we show that SNAMs can even outperform NAMs while remaining inherently more interpretable.	['Benjamin Säfken', 'Anton Thielmann', 'Mattias Luber']	2023-02-18	null	null	null	null	['additive-models', 'interpretable-machine-learning']	['methodology', 'methodology']	[ 3.87269080e-01 7.21567452e-01 1.80220693e-01 -5.65024197e-01 2.46246322e-03 -4.41583395e-01 9.62765157e-01 -1.78032875e-01 -1.77020401e-01 7.34732687e-01 4.48899001e-01 -7.45676041e-01 -4.58203614e-01 -6.15579069e-01 -8.54211032e-01 -7.18569934e-01 -6.46812394e-02 3.55288237e-01 -1.69492453e-01 -1.18927658e-01 9.15568024e-02 6.55205727e-01 -1.65774083e+00 4.83635366e-01 9.52130616e-01 1.11536682e+00 1.11664146e-01 5.86627483e-01 -1.30196214e-01 1.08283615e+00 -4.47728932e-01 -5.37493169e-01 -1.12715056e-02 -2.19589546e-01 -4.51433957e-01 -4.00514364e-01 6.75416887e-01 -1.36991292e-01 -4.49094504e-01 1.14890516e+00 1.46279246e-01 4.05316018e-02 7.06605494e-01 -1.46303320e+00 -1.54532075e+00 8.35188091e-01 -2.04533860e-01 -3.32843550e-02 -2.75656223e-01 3.63946736e-01 1.10318244e+00 -1.04904437e+00 3.54217440e-01 1.91556549e+00 5.62178969e-01 6.41833603e-01 -1.48434031e+00 -3.81132722e-01 4.29132462e-01 3.98774207e-01 -7.75397182e-01 -4.21840847e-01 6.53085649e-01 -4.79465634e-01 9.76546228e-01 6.96903884e-01 4.51325268e-01 1.27916813e+00 1.46602258e-01 9.10535634e-01 1.29758894e+00 -4.31014061e-01 5.59576988e-01 4.99712117e-02 4.21902835e-01 8.05897593e-01 6.49739206e-01 2.37376839e-01 -5.11380255e-01 2.89774865e-01 9.37697232e-01 2.18264967e-01 -1.46913096e-01 -2.01275468e-01 -1.12440646e+00 6.64848447e-01 7.60883152e-01 2.37412170e-01 -3.69572878e-01 5.54505646e-01 1.68568954e-01 5.27008846e-02 6.25592947e-01 5.75665236e-01 -3.14004689e-01 1.06409192e-01 -7.20419466e-01 2.14851096e-01 5.46295166e-01 6.98270440e-01 4.11834717e-01 6.52479470e-01 -2.63917416e-01 6.58934176e-01 3.50202054e-01 2.48014733e-01 1.92324176e-01 -9.80962157e-01 2.75827855e-01 8.57146621e-01 -1.55895665e-01 -9.91580844e-01 -4.66818571e-01 -5.74076474e-01 -1.30123055e+00 9.67599332e-01 4.09766883e-01 -2.91167246e-03 -1.32204533e+00 1.66655958e+00 -3.02311540e-01 -1.29655495e-01 2.48340815e-02 8.87843430e-01 9.37286854e-01 7.19634712e-01 3.91487420e-01 3.28174025e-01 1.16721237e+00 -9.40267205e-01 -9.92731929e-01 -3.02446187e-01 3.07455629e-01 -7.14299083e-02 1.20005214e+00 5.72790265e-01 -1.07569814e+00 -7.27478027e-01 -1.24415910e+00 -3.54377329e-01 -6.04646742e-01 2.60601580e-01 7.95194805e-01 5.64511120e-01 -1.49698901e+00 9.30873036e-01 -1.11412275e+00 -1.06039636e-01 7.94870675e-01 5.95100284e-01 -4.31374073e-01 2.46020809e-01 -9.29585278e-01 1.07769310e+00 7.42262423e-01 5.69766462e-01 -7.80818284e-01 -6.72960758e-01 -7.61133730e-01 3.46547484e-01 2.99624056e-01 -9.29657280e-01 1.06521225e+00 -1.34035909e+00 -1.36052847e+00 5.32557189e-01 -2.71928638e-01 -8.36789429e-01 6.75206423e-01 -5.29862285e-01 -3.85894507e-01 -1.80112466e-01 -5.17049313e-01 1.01869524e+00 5.57298601e-01 -1.59374022e+00 -2.52475113e-01 -4.39254612e-01 3.07774901e-01 -5.35185225e-02 -3.98738563e-01 -7.72608519e-02 1.22882478e-01 -8.89633417e-01 2.00561732e-01 -7.90037036e-01 -3.54115903e-01 7.61282682e-01 -9.38640356e-01 -1.97384134e-01 1.08576822e+00 -6.34517848e-01 1.01802707e+00 -2.02187610e+00 3.73293459e-01 -9.51168239e-02 9.80891645e-01 4.34503704e-01 -1.30240738e-01 -2.02539675e-02 -4.50004071e-01 5.92929065e-01 -2.24011526e-01 -6.69726074e-01 4.41794991e-01 5.74832678e-01 -3.02321970e-01 -7.12139392e-03 4.44966674e-01 1.12530708e+00 -8.01107824e-01 -1.80802152e-01 5.35712242e-01 7.41930127e-01 -2.15082675e-01 3.43091600e-02 -2.19150916e-01 1.97937801e-01 -1.14082105e-01 4.61642742e-01 6.12134576e-01 -3.53701144e-01 7.34826997e-02 -2.70125210e-01 -8.63848105e-02 6.34799227e-02 -9.74255741e-01 1.03611481e+00 -2.55152136e-01 1.36629677e+00 -2.24716261e-01 -9.57752287e-01 9.70022976e-01 8.40371624e-02 -2.50057787e-01 -6.37381911e-01 -9.36111435e-02 1.15859143e-01 4.10036981e-01 -3.52858692e-01 4.51528370e-01 -6.69199452e-02 5.17584741e-01 9.55718532e-02 1.58547491e-01 5.18635035e-01 -5.92320226e-02 3.88819724e-02 9.25664186e-01 3.03434700e-01 4.38859046e-01 -4.63620335e-01 1.73466906e-01 -2.28967071e-01 3.14347953e-01 8.57120037e-01 7.87926540e-02 5.98297477e-01 7.14936256e-01 -9.12106991e-01 -1.11316407e+00 -1.38160145e+00 1.49457613e-02 1.00197780e+00 -3.76807563e-02 -3.16826254e-02 -9.86467481e-01 -2.96404094e-01 -8.71479139e-02 1.16967010e+00 -9.84571278e-01 -1.11063071e-01 -3.68909806e-01 -6.00098372e-01 6.55713677e-01 1.04252255e+00 6.25475287e-01 -1.28546643e+00 -4.80764091e-01 2.01473422e-02 3.57123822e-01 -8.73560429e-01 3.74425173e-01 3.57371300e-01 -1.24334359e+00 -6.13034427e-01 -6.05238438e-01 -2.65811175e-01 8.54409635e-01 1.66487098e-02 1.28500557e+00 -9.08647552e-02 -2.12229341e-01 -1.71855941e-01 -2.39758845e-02 -8.18494439e-01 -6.98946893e-01 -3.44289571e-01 2.24319264e-01 4.78730462e-02 2.37073049e-01 -8.15407097e-01 -7.88309395e-01 1.03391297e-01 -1.10017681e+00 7.82588303e-01 9.04100537e-01 6.35116041e-01 4.63831633e-01 -2.29427487e-01 4.47943956e-01 -9.52477455e-01 7.11713910e-01 -2.84638852e-01 -3.13255847e-01 2.53193200e-01 -5.38941562e-01 5.77713430e-01 9.41386402e-01 -4.29216653e-01 -1.19007111e+00 -1.15780048e-01 6.16819523e-02 -4.40037221e-01 -4.32736129e-01 4.74682152e-01 -4.75588292e-01 -2.65041413e-03 7.59656012e-01 -5.05143143e-02 -6.61947355e-02 -7.32194781e-01 5.97644866e-01 3.61005574e-01 6.46113813e-01 -2.38088891e-01 7.38196850e-01 4.07741576e-01 2.80098051e-01 -7.54935920e-01 -6.87677503e-01 4.07014072e-01 -7.47201085e-01 -2.17625573e-01 8.47266495e-01 -5.53532302e-01 -6.81954265e-01 2.09592342e-01 -1.49376297e+00 -2.70594090e-01 -3.45996916e-01 5.02691232e-02 -3.10217857e-01 1.72446668e-01 -4.77221131e-01 -1.24244642e+00 -3.10340911e-01 -1.18047643e+00 8.16979527e-01 2.85637259e-01 -6.69658542e-01 -1.35074353e+00 -3.84594768e-01 1.36236727e-01 4.92053896e-01 6.78276420e-01 1.42128170e+00 -7.68910348e-01 -8.19982708e-01 1.54810753e-02 -6.74181342e-01 4.52697515e-01 -9.20440927e-02 2.09337637e-01 -1.52105522e+00 2.46444568e-01 -2.41307795e-01 1.15397125e-01 1.12098491e+00 6.71334982e-01 1.66665602e+00 -7.78848350e-01 -3.14591348e-01 4.64677602e-01 1.30825877e+00 3.85507762e-01 8.45017433e-01 2.21424997e-01 1.05811286e+00 5.89428306e-01 -7.72683322e-02 -1.42179048e-02 1.00960910e-01 3.90871942e-01 9.77491975e-01 -5.41460752e-01 -1.29191771e-01 -4.03758436e-01 2.23756239e-01 5.33545732e-01 -7.37431109e-01 -7.15462923e-01 -7.93236613e-01 2.41397023e-01 -2.03138113e+00 -1.00472105e+00 -4.05938387e-01 1.94968343e+00 3.00228566e-01 4.26094800e-01 -3.31503391e-01 3.45675856e-01 5.66317439e-01 2.12456912e-01 -8.26792061e-01 -7.87318885e-01 -4.62687314e-01 4.09625880e-02 2.29258075e-01 2.95791477e-01 -1.05745804e+00 5.00775695e-01 6.35283089e+00 7.63440907e-01 -8.93156230e-01 -1.40599385e-01 1.18670642e+00 1.71791837e-01 -5.91850162e-01 -2.34011680e-01 -3.92389894e-01 2.82667637e-01 1.04126513e+00 8.62767398e-02 2.60701954e-01 1.08539176e+00 6.25678539e-01 1.26326174e-01 -1.57753599e+00 1.02872801e+00 -2.39178181e-01 -1.73402643e+00 6.54313147e-01 1.77636340e-01 5.59463203e-01 -1.99855268e-01 5.55624783e-01 9.59522203e-02 4.32000637e-01 -1.74083245e+00 1.00912857e+00 8.41498852e-01 7.05249786e-01 -4.83871609e-01 6.51242316e-01 2.67727196e-01 -8.06648672e-01 -1.69145420e-01 -5.53772926e-01 -3.13419044e-01 -1.15022488e-01 4.91980970e-01 -6.47740185e-01 1.81246728e-01 6.36090219e-01 6.72149181e-01 -6.96395397e-01 1.00902188e+00 -2.19611421e-01 6.60139620e-01 -9.79878306e-02 -3.00083399e-01 3.11965913e-01 -7.79785514e-02 5.49618840e-01 1.16418684e+00 3.92583936e-01 -7.44277835e-02 -5.58023155e-01 1.49007726e+00 -1.17582716e-01 -3.59380901e-01 -7.89102793e-01 -3.09245110e-01 1.40413180e-01 1.13412905e+00 -7.70635903e-01 -4.59538043e-01 -2.00602204e-01 8.39167237e-01 4.13390338e-01 5.25923848e-01 -6.69808626e-01 -7.21876249e-02 7.64229536e-01 -3.71475029e-03 -1.02437750e-01 -2.21136704e-01 -1.00262690e+00 -8.58306050e-01 4.86613587e-02 -7.93049991e-01 -9.96546075e-03 -1.31547487e+00 -1.24974692e+00 9.10599768e-01 -8.38967934e-02 -1.20573211e+00 -9.32388529e-02 -1.41652215e+00 -7.05914676e-01 8.84511113e-01 -9.43775177e-01 -1.13030553e+00 -3.53854299e-01 -4.65734787e-02 7.11943805e-01 -4.59746458e-02 1.03528237e+00 -2.01579407e-01 -5.07039189e-01 4.41552401e-01 4.15654987e-01 -2.11018398e-02 2.15316154e-02 -1.63661957e+00 7.19754696e-01 7.54634798e-01 3.80784661e-01 8.19062054e-01 1.05894399e+00 -2.35884324e-01 -1.25917792e+00 -9.57152069e-01 5.27833879e-01 -5.26469529e-01 6.15580618e-01 -6.45838261e-01 -1.04492378e+00 7.58375227e-01 4.07471389e-01 -2.63756096e-01 5.81714272e-01 2.06340596e-01 -3.35875839e-01 -8.96395370e-02 -8.16191614e-01 1.10059738e+00 1.16091597e+00 -3.38165134e-01 -6.69201612e-01 6.44953847e-02 9.23987448e-01 -1.09351858e-01 -2.89115518e-01 4.46863890e-01 6.55415654e-01 -1.19729531e+00 1.01465964e+00 -8.59420300e-01 9.10872638e-01 -1.90757170e-01 5.41750081e-02 -1.51378274e+00 -4.22251135e-01 -5.92458367e-01 -4.01427031e-01 9.39637840e-01 6.83172762e-01 -6.14599228e-01 7.88502634e-01 1.22852993e+00 -3.78238976e-01 -8.67633641e-01 -7.56598771e-01 -6.40652716e-01 -7.29639009e-02 -6.35251403e-01 3.93149525e-01 6.02859616e-01 -1.75441563e-01 2.47267634e-01 -3.02416086e-01 3.05711683e-02 6.67572856e-01 -2.43242174e-01 2.62888312e-01 -1.65773129e+00 -3.70484680e-01 -8.43393385e-01 -7.50659823e-01 -1.02850235e+00 -1.17660075e-01 -6.49569571e-01 -1.46772623e-01 -1.78325450e+00 1.67920172e-01 -2.37568896e-02 -5.99010110e-01 6.60358429e-01 -1.66174948e-01 7.24520534e-02 4.15663749e-01 -2.36763377e-02 -5.47979891e-01 3.64914358e-01 1.33964670e+00 -2.34678358e-01 1.19965728e-02 -1.47613734e-01 -1.02376592e+00 1.08037877e+00 6.42244577e-01 -1.22348182e-01 -6.63325250e-01 -6.84434772e-01 2.61214852e-01 -1.77391052e-01 9.30002451e-01 -1.01177669e+00 -2.08631214e-02 -3.40420566e-02 7.25241005e-01 -4.49621856e-01 5.21234214e-01 -8.67151380e-01 4.55371678e-01 4.02641475e-01 -7.54037917e-01 1.42393619e-01 5.06694376e-01 6.53071523e-01 -1.80364519e-01 6.72716126e-02 5.14499784e-01 1.97846834e-02 -8.93638492e-01 1.64368182e-01 -3.52338523e-01 -3.26534212e-01 6.58955693e-01 -6.13905847e-01 -6.08586729e-01 -5.11250496e-01 -9.92220640e-01 -1.95896015e-01 6.83492124e-02 5.43363035e-01 8.88615012e-01 -1.07283819e+00 -4.15643573e-01 -4.16188203e-02 -1.80763900e-01 -1.31298034e-02 3.61037046e-01 3.74990851e-01 -6.53885126e-01 6.13479137e-01 -4.34546292e-01 -4.57360208e-01 -1.27099705e+00 5.12919068e-01 2.96587437e-01 1.40874786e-02 -5.88948190e-01 7.67400682e-01 8.59824538e-01 -1.91521391e-01 3.40628326e-01 -7.95538664e-01 -3.61715198e-01 -2.77144402e-01 7.51294911e-01 5.86250246e-01 -1.61484227e-01 -3.29172969e-01 -3.12096458e-02 -3.41447592e-02 1.15945660e-01 6.90913051e-02 1.66409576e+00 1.45817608e-01 -9.05122682e-02 7.19549477e-01 8.74124765e-01 -7.24474430e-01 -1.58121455e+00 5.57893924e-02 1.61908507e-01 -9.41973627e-02 4.91310097e-02 -1.19477499e+00 -9.13269341e-01 1.52749777e+00 7.31356084e-01 5.23546696e-01 1.04465747e+00 -1.59944907e-01 1.11144461e-01 5.88576257e-01 1.13943964e-02 -7.32499123e-01 -1.59594893e-01 3.02786350e-01 1.21797192e+00 -1.21901882e+00 -2.41714820e-01 -3.64789277e-01 -5.72294891e-01 1.45830667e+00 5.69138825e-01 -1.67262048e-01 3.00363541e-01 3.27437818e-01 -8.74909535e-02 -8.92952159e-02 -8.23492765e-01 1.87161669e-01 7.08933473e-01 7.13052034e-01 4.43141490e-01 3.41178983e-01 2.68654674e-01 7.67491698e-01 -2.47405037e-01 1.98390577e-02 2.15772256e-01 3.41913372e-01 -4.32141066e-01 -6.48654461e-01 -2.50067621e-01 5.23495376e-01 -3.32166404e-01 -4.49353188e-01 -5.94079912e-01 1.00147605e+00 6.93248883e-02 6.68948114e-01 1.25784427e-01 -4.67352480e-01 2.22095937e-01 7.53738061e-02 2.49177013e-02 -1.78000703e-01 -1.57924414e-01 -4.85233068e-01 2.25665674e-01 -4.41251218e-01 -3.46495658e-01 -2.58524567e-01 -1.10684919e+00 -5.68948865e-01 5.82864732e-02 -3.84211153e-01 6.50073230e-01 1.17140877e+00 5.55057526e-01 8.67250264e-01 -4.06170525e-02 -1.05057526e+00 -3.24781209e-01 -8.69690657e-01 -2.89525837e-01 2.16122076e-01 4.36101198e-01 -6.21590316e-01 -3.66436183e-01 2.68303156e-01]	[8.880178451538086, 5.560168743133545]
2bb08eaf-13a8-4832-9dfb-634e799ab7fe	jarvix-at-semeval-2022-task-2-it-takes-one-to	2202.02394	null	https://arxiv.org/abs/2202.02394v6	https://arxiv.org/pdf/2202.02394v6.pdf	JARVix at SemEval-2022 Task 2: It Takes One to Know One? Idiomaticity Detection using Zero and One-Shot Learning	Large Language Models have been successful in a wide variety of Natural Language Processing tasks by capturing the compositionality of the text representations. In spite of their great success, these vector representations fail to capture meaning of idiomatic multi-word expressions (MWEs). In this paper, we focus on the detection of idiomatic expressions by using binary classification. We use a dataset consisting of the literal and idiomatic usage of MWEs in English and Portuguese. Thereafter, we perform the classification in two different settings: zero shot and one shot, to determine if a given sentence contains an idiom or not. N shot classification for this task is defined by N number of common idioms between the training and testing sets. In this paper, we train multiple Large Language Models in both the settings and achieve an F1 score (macro) of 0.73 for the zero shot setting and an F1 score (macro) of 0.85 for the one shot setting. An implementation of our work can be found at https://github.com/ashwinpathak20/Idiomaticity_Detection_Using_Few_Shot_Learning.	['Yash Jakhotiya', 'Vaibhav Kumar', 'Raj Shah', 'Ashwin Pathak']	2022-02-04	null	https://aclanthology.org/2022.semeval-1.19	https://aclanthology.org/2022.semeval-1.19.pdf	semeval-naacl-2022-7	['one-shot-learning']	['methodology']	[ 6.70666248e-02 -2.23391697e-01 -5.88948429e-01 -2.75760055e-01 -6.24184549e-01 -5.29822946e-01 9.11463499e-01 1.24568924e-01 -4.33714598e-01 4.31255132e-01 4.55189586e-01 -2.38823071e-01 1.02167383e-01 -9.78371203e-01 -2.88231261e-02 -6.29155993e-01 3.32333952e-01 6.69831634e-01 1.45602068e-02 -5.37622273e-01 4.54406202e-01 1.59743637e-01 -1.43592417e+00 7.38686323e-01 4.97290105e-01 8.51135135e-01 -7.64509523e-03 5.90079546e-01 -5.91420352e-01 1.27274907e+00 -7.02269673e-01 -6.21609151e-01 1.48503287e-02 -7.60842443e-01 -8.24889481e-01 -1.97408244e-01 4.21071611e-02 1.09178856e-01 -8.33465233e-02 1.24943936e+00 5.53668976e-01 2.87187248e-01 7.39155054e-01 -1.15128911e+00 -5.93483210e-01 9.68019366e-01 -6.96816802e-01 6.29511356e-01 2.53819525e-01 -1.56100422e-01 1.24077559e+00 -1.07933593e+00 9.34023142e-01 1.32969999e+00 3.43401939e-01 4.63262469e-01 -1.14744699e+00 -8.03501904e-01 -3.54466408e-01 4.88184094e-01 -1.21718812e+00 -6.45089924e-01 1.04626667e+00 -6.39454484e-01 1.15464389e+00 -4.09531174e-04 2.98545182e-01 1.43770361e+00 3.19148988e-01 6.98606431e-01 1.11238205e+00 -7.32566714e-01 2.67200917e-01 3.31773460e-01 5.62520087e-01 7.05628335e-01 -2.42456004e-01 -1.67739749e-01 -4.51448649e-01 -2.10119247e-01 2.42848709e-01 -4.08382006e-02 1.40724450e-01 1.73497573e-01 -8.77794862e-01 1.56468594e+00 1.49284899e-01 7.90826559e-01 -3.98894727e-01 -1.47287741e-01 9.18290854e-01 6.15624785e-01 4.26885277e-01 4.35175806e-01 -2.62005478e-01 -4.19290930e-01 -6.34965062e-01 9.88146961e-02 6.59117222e-01 4.96451914e-01 3.34980339e-01 1.49776489e-01 -8.01452473e-02 1.47154760e+00 -3.37417513e-01 2.35157073e-01 1.21042788e+00 -6.19505644e-01 3.13903898e-01 6.06832862e-01 -3.46649766e-01 -1.08605683e+00 -3.24480832e-01 1.43803898e-02 -7.64667094e-01 5.30097587e-03 2.52379775e-01 -3.09268236e-01 -5.61637044e-01 1.80709410e+00 1.86549455e-01 -2.93844491e-01 2.02302769e-01 6.46856904e-01 9.79219615e-01 8.68319571e-01 5.71456999e-02 -3.82211804e-01 1.74631739e+00 -8.83772671e-01 -8.95634115e-01 -2.57249981e-01 7.91737795e-01 -9.63617682e-01 1.22962236e+00 2.97540396e-01 -8.85106921e-01 -3.72443438e-01 -1.04814613e+00 -1.15638040e-01 -5.53655148e-01 -1.58150792e-01 6.07059896e-01 5.70425034e-01 -3.36466461e-01 3.32813144e-01 -5.86616635e-01 -3.38904470e-01 4.54595476e-01 8.47672857e-03 -2.84066767e-01 8.16060454e-02 -1.33843839e+00 1.18641746e+00 3.67772728e-01 -6.24009192e-01 -4.43990976e-01 -3.71933192e-01 -9.55645442e-01 -2.46718246e-03 4.80705082e-01 -1.37845799e-01 1.19046962e+00 -1.17900753e+00 -1.32684255e+00 1.35070157e+00 -8.69401917e-02 -4.17212665e-01 2.90852278e-01 -2.65057012e-02 -4.36353028e-01 -9.34885591e-02 2.86784202e-01 3.61609429e-01 6.35301828e-01 -6.95756853e-01 -3.12041789e-01 -3.89630675e-01 1.44420162e-01 -9.94006991e-02 -2.81384110e-01 8.80064428e-01 3.86800505e-02 -8.05073142e-01 -1.15612917e-01 -7.47631788e-01 1.00500681e-01 -2.59793252e-01 -1.93610057e-01 -4.40804064e-01 6.45960689e-01 -4.98612940e-01 1.19253111e+00 -2.26173186e+00 1.19455747e-01 -2.31235832e-01 -3.42205688e-02 1.66703120e-01 -6.89140558e-02 5.36781549e-01 -1.81548893e-01 4.49907556e-02 -3.13126653e-01 -1.04492456e-01 -1.36536092e-01 4.10822898e-01 -3.73848081e-01 4.56379384e-01 1.96076736e-01 8.19012463e-01 -9.52056348e-01 -5.97172201e-01 1.45097464e-01 4.21219856e-01 -4.29313719e-01 4.94353436e-02 -8.44442546e-02 -1.76305808e-02 -2.03582793e-02 6.24793291e-01 2.80557007e-01 -1.65100545e-01 5.88694453e-01 -1.51649872e-02 5.43801785e-02 3.84942472e-01 -9.83698249e-01 1.38172424e+00 -8.35712194e-01 7.09748566e-01 -2.42277339e-01 -1.14199018e+00 8.78293514e-01 4.35617685e-01 2.82883555e-01 -8.42910707e-01 6.78114533e-01 8.35620910e-02 5.22059143e-01 -4.37640101e-01 3.58686566e-01 -8.55256975e-01 -4.77902442e-01 8.10283124e-01 3.37268263e-01 9.46673825e-02 4.00363892e-01 -4.59960010e-03 9.47358727e-01 -4.22870338e-01 1.20063090e+00 -2.80436516e-01 5.22094250e-01 -7.11683184e-02 6.74329400e-01 4.55739528e-01 -2.34910861e-01 3.13136429e-01 8.06515753e-01 -5.99673867e-01 -8.81698549e-01 -9.45302904e-01 -3.32116902e-01 1.47708237e+00 -1.63275555e-01 -5.29826641e-01 -3.58056754e-01 -5.83617747e-01 -2.35405892e-01 1.12495708e+00 -6.94265962e-01 -3.24535072e-01 -6.90158725e-01 -9.74743009e-01 5.90291500e-01 5.06621122e-01 3.78750682e-01 -1.46002650e+00 -1.10543537e+00 9.83017758e-02 -3.05284619e-01 -9.63532090e-01 -1.38203561e-01 3.53381187e-01 -2.82629907e-01 -1.10309100e+00 -2.07099855e-01 -8.04122448e-01 1.45879060e-01 7.19353231e-03 1.04321742e+00 -2.10989751e-02 -4.85308081e-01 -3.50222498e-01 -5.97440660e-01 -3.38940173e-01 -6.68780267e-01 -1.10936925e-01 9.82134417e-02 -6.96617831e-03 7.58813143e-01 -4.19447064e-01 5.08439951e-02 3.33731286e-02 -7.99393237e-01 3.13957669e-02 2.20285147e-01 1.13634598e+00 2.13628933e-01 -5.65743297e-02 5.04987478e-01 -1.17188132e+00 8.20430994e-01 -7.94371843e-01 -3.48464638e-01 -1.73197739e-04 -3.10812443e-01 -6.93752617e-02 7.46870220e-01 -7.14786291e-01 -7.75296986e-01 -3.56667876e-01 -3.67116392e-01 -2.87172884e-01 -5.98690584e-02 6.05868816e-01 -3.70200345e-04 4.31277961e-01 7.32205689e-01 -1.00353956e-01 -1.46618649e-01 -3.42017144e-01 3.86739224e-01 9.00821805e-01 2.52278239e-01 -4.94756013e-01 2.69788951e-01 2.52217025e-01 -3.23388070e-01 -1.03216326e+00 -1.15068519e+00 -4.26217794e-01 -5.54540217e-01 1.02039337e-01 9.20010030e-01 -6.09183669e-01 -2.41001725e-01 2.42176577e-01 -1.13742304e+00 -1.04612716e-01 -2.53467858e-01 1.71100751e-01 -5.47870696e-01 1.96273372e-01 -8.29851270e-01 -8.30551922e-01 -6.30623281e-01 -9.97956634e-01 6.47108436e-01 1.76792890e-02 -8.69895458e-01 -1.01516366e+00 3.16837728e-01 2.68503994e-01 3.83416861e-01 3.74566585e-01 1.33303881e+00 -1.12390208e+00 4.39401031e-01 -1.84797242e-01 -1.19154222e-01 8.21776539e-02 3.38130951e-01 -2.28158325e-01 -1.08073390e+00 -3.61722969e-02 4.40623254e-01 -6.17945075e-01 8.37954938e-01 5.59915565e-02 7.78547466e-01 -5.42075217e-01 1.99716300e-01 2.85706013e-01 1.46006513e+00 3.63447398e-01 4.19108450e-01 3.24596465e-01 3.14218104e-01 4.73244458e-01 5.01104832e-01 5.16859174e-01 -9.99213289e-03 9.36709583e-01 -8.82739201e-03 4.10918593e-01 -3.40920426e-02 3.01565509e-02 5.52669406e-01 8.09682369e-01 1.77212194e-01 -9.76320580e-02 -1.23808432e+00 6.54673755e-01 -1.67652082e+00 -1.39236438e+00 2.21510038e-01 1.91293788e+00 1.01665354e+00 2.82713979e-01 1.41048551e-01 3.32150370e-01 5.82641482e-01 5.05800307e-01 -1.57967269e-01 -9.56580639e-01 -2.19726384e-01 5.60771525e-01 -7.78610483e-02 7.08290160e-01 -1.20270765e+00 1.15832639e+00 4.97736025e+00 9.12370920e-01 -1.26334929e+00 6.37692928e-01 3.35697532e-01 -2.36701310e-01 -5.58878630e-02 -1.70316041e-01 -6.70936167e-01 5.89238644e-01 1.32308674e+00 -4.27188426e-01 3.14775527e-01 8.95798862e-01 -1.39165059e-01 2.69308556e-02 -6.63753450e-01 1.04428232e+00 4.67081130e-01 -1.33359742e+00 1.20595835e-01 -1.48610964e-01 7.05753326e-01 1.66642457e-01 -1.96708322e-01 5.91679633e-01 3.42510581e-01 -8.98966968e-01 5.83680928e-01 -1.01446167e-01 8.27616096e-01 -8.21046889e-01 9.59723592e-01 4.87950146e-01 -8.36371899e-01 -3.89880359e-01 -3.87959242e-01 -3.23259443e-01 3.17553759e-01 3.69111627e-01 -5.98594844e-01 1.15731537e-01 2.52879113e-01 5.60679257e-01 -3.25666338e-01 2.42648810e-01 -3.66261661e-01 3.84952664e-01 -4.26610792e-03 -1.69257134e-01 2.70854414e-01 -3.43980491e-02 4.49234396e-01 1.44020689e+00 -8.66207331e-02 1.12761594e-01 4.61366147e-01 7.91213989e-01 -5.31282425e-02 4.90783364e-01 -8.02044332e-01 -3.13794374e-01 3.95421714e-01 1.26251912e+00 -8.25654805e-01 -5.78628182e-01 -3.90574753e-01 1.02123833e+00 4.87940043e-01 -1.08665287e-01 -8.46332431e-01 -5.14043748e-01 7.59027004e-01 -1.43783674e-01 1.56467304e-01 -2.91880295e-02 -3.30592245e-01 -1.35877872e+00 -3.41682941e-01 -9.24256265e-01 6.87037647e-01 -4.20146048e-01 -1.45203900e+00 6.40665829e-01 3.81272100e-02 -8.10529113e-01 -5.73191106e-01 -9.20915127e-01 -9.09856617e-01 5.71544647e-01 -9.92553174e-01 -1.02397573e+00 -1.62820369e-02 3.32478344e-01 1.20643508e+00 -3.53214025e-01 1.23628783e+00 2.09265232e-01 -6.09467447e-01 5.79961896e-01 -2.00019479e-01 5.57908416e-01 5.13237178e-01 -1.03408062e+00 1.38005257e-01 6.78021908e-01 4.16958809e-01 6.82418227e-01 7.73503363e-01 -4.40065920e-01 -9.68022823e-01 -7.20606446e-01 1.26319301e+00 -3.10199678e-01 1.01126826e+00 -3.98611426e-01 -8.42053890e-01 6.81739688e-01 2.73058534e-01 -2.85827041e-01 1.08702826e+00 3.55775505e-01 -7.31593907e-01 3.92670393e-01 -1.22803462e+00 6.33419275e-01 7.71851182e-01 -6.84442103e-01 -9.60402429e-01 5.80704451e-01 3.58694851e-01 4.18459531e-03 -7.76324630e-01 3.10753316e-01 5.85162461e-01 -7.62544036e-01 7.54039824e-01 -1.13754165e+00 1.07691574e+00 2.52870649e-01 -7.66556561e-01 -1.06283891e+00 -1.81092992e-01 -5.57482801e-02 -1.44545048e-01 1.13418162e+00 3.32616210e-01 -4.83901352e-01 2.81482071e-01 -5.80382720e-02 2.28609353e-01 -7.07322598e-01 -1.10888898e+00 -8.07726085e-01 3.38293672e-01 -4.12973404e-01 1.44472256e-01 1.46179998e+00 4.61466879e-01 1.05263805e+00 -3.79514486e-01 -3.80160958e-01 3.35780203e-01 5.43171346e-01 4.90996033e-01 -1.11175537e+00 -4.01633888e-01 -8.44637811e-01 -5.90934873e-01 -2.51837969e-01 5.56675375e-01 -1.27876782e+00 -2.42812186e-01 -1.02076590e+00 5.97672641e-01 1.37626216e-01 -3.13827306e-01 6.72675431e-01 1.04387663e-01 4.42776948e-01 1.91730663e-01 1.39706939e-01 -2.73614973e-01 4.45525944e-01 4.65917528e-01 -1.88267738e-01 -6.15170002e-02 -3.88236672e-01 -5.55267453e-01 9.77028012e-01 1.23106694e+00 -4.67451364e-01 -2.21613824e-01 -1.04592644e-01 1.09286241e-01 -1.06764875e-01 3.22252661e-02 -6.22633457e-01 -2.04707056e-01 -3.30799729e-01 1.06271952e-01 -2.46848688e-01 5.58784664e-01 -3.24740529e-01 -1.20639950e-01 7.16284215e-01 -3.75797004e-01 4.29579556e-01 1.18126772e-01 -2.72933953e-02 -1.52708977e-01 -6.20777905e-01 1.24194443e+00 -5.49991250e-01 -1.00309992e+00 -2.30773702e-01 -4.89494979e-01 3.27940345e-01 9.18544590e-01 1.83843151e-01 -4.06099677e-01 -1.49700552e-01 -5.55270970e-01 -2.71266669e-01 4.14774388e-01 8.01201344e-01 6.71301901e-01 -1.32859159e+00 -6.37039423e-01 3.57452303e-01 3.77861351e-01 -8.68373990e-01 -2.07759254e-02 1.04466414e+00 -3.96560669e-01 1.25242516e-01 -5.29978096e-01 -2.27902219e-01 -1.51314259e+00 6.21690094e-01 2.31507286e-01 -3.05028647e-01 -8.28188539e-01 6.76942587e-01 -6.41026646e-02 -5.19275367e-01 -1.94145337e-01 3.02268207e-01 -3.38568449e-01 6.02971375e-01 9.11325276e-01 2.11584449e-01 -1.56719625e-01 -1.00855136e+00 -3.76354009e-01 4.10149217e-01 -3.41393799e-01 -1.81622699e-01 1.30234289e+00 2.36895591e-01 -2.30814785e-01 1.13086808e+00 1.29526365e+00 1.23767853e-01 -1.78937420e-01 -1.30964726e-01 2.89028913e-01 -4.23657000e-01 -6.32331520e-02 -6.91076875e-01 -7.38242090e-01 9.11593676e-01 4.46558714e-01 4.59074825e-02 6.89853430e-01 2.73958445e-01 6.67151809e-01 2.40475982e-01 2.10674822e-01 -9.75973129e-01 4.68559936e-03 9.64216232e-01 8.42430890e-01 -1.34724820e+00 -2.18285426e-01 -1.95618987e-01 -9.15200055e-01 1.03077924e+00 3.28187913e-01 -3.16657662e-01 4.65203196e-01 3.07910144e-01 3.53089720e-01 -6.25984669e-01 -1.01671112e+00 -1.80748239e-01 -8.75899009e-03 1.71802133e-01 8.94557178e-01 1.50621027e-01 -7.42097676e-01 6.78287029e-01 -6.13388836e-01 -2.47560829e-01 5.60279667e-01 8.21888626e-01 -5.86885929e-01 -9.21537876e-01 1.86718553e-02 4.54857439e-01 -6.68547153e-01 -2.02238739e-01 -6.18859708e-01 6.19513571e-01 4.74805795e-02 8.84650409e-01 2.00165734e-01 -3.71994883e-01 1.23386495e-01 5.79546452e-01 3.67854983e-01 -7.92822361e-01 -6.10720336e-01 -2.11408406e-01 3.54242265e-01 -4.11153287e-01 -1.73644722e-01 -5.98732591e-01 -1.29204929e+00 -4.21609879e-01 -9.23923403e-02 1.39384763e-02 3.34400624e-01 9.40148592e-01 -1.15816578e-01 2.24710405e-01 4.73509163e-01 -5.04073381e-01 -7.57710755e-01 -1.18851471e+00 -4.55536574e-01 7.30982423e-01 1.29273925e-02 -9.99524713e-01 -3.53408396e-01 -2.59121954e-01]	[10.586067199707031, 9.5941801071167]
858b2710-9a9c-4284-929f-7012ff8aa0f5	attentive-recurrent-tensor-model-for	1801.06792	null	http://arxiv.org/abs/1801.06792v1	http://arxiv.org/pdf/1801.06792v1.pdf	Attentive Recurrent Tensor Model for Community Question Answering	A major challenge to the problem of community question answering is the lexical and semantic gap between the sentence representations. Some solutions to minimize this gap includes the introduction of extra parameters to deep models or augmenting the external handcrafted features. In this paper, we propose a novel attentive recurrent tensor network for solving the lexical and semantic gap in community question answering. We introduce token-level and phrase-level attention strategy that maps input sequences to the output using trainable parameters. Further, we use the tensor parameters to introduce a 3-way interaction between question, answer and external features in vector space. We introduce simplified tensor matrices with L2 regularization that results in smooth optimization during training. The proposed model achieves state-of-the-art performance on the task of answer sentence selection (TrecQA and WikiQA datasets) while outperforming the current state-of-the-art on the tasks of best answer selection (Yahoo! L4) and answer triggering task (WikiQA).	['Balasubramanian Raman', 'Shivam Sharma', 'Gaurav Bhatt']	2018-01-21	null	null	null	null	['l2-regularization']	['methodology']	[ 2.93753166e-02 -2.25957006e-01 1.56857058e-01 -6.10235393e-01 -1.29063606e+00 -7.31964111e-01 3.14532220e-01 3.47516328e-01 -6.48850620e-01 2.58575112e-01 4.55920488e-01 -3.79747152e-01 -2.16562569e-01 -5.41801989e-01 -3.62545997e-01 -2.32060730e-01 7.52016678e-02 5.94680429e-01 2.58606523e-01 -6.97077751e-01 5.81362784e-01 -2.85110295e-01 -1.37873566e+00 9.77308989e-01 9.10101652e-01 1.19891369e+00 1.00384712e-01 1.16355348e+00 -6.45369589e-01 1.25277841e+00 -5.14049709e-01 -7.24549532e-01 -3.80982868e-02 -3.78458291e-01 -1.74049497e+00 -1.75547332e-01 6.11832500e-01 -9.91470516e-02 -1.73501804e-01 5.77069581e-01 6.39394581e-01 3.66513610e-01 3.27289969e-01 -8.50841820e-01 -9.67643261e-01 5.63111126e-01 5.00893220e-02 6.38574839e-01 6.31538868e-01 1.61315426e-01 1.72530901e+00 -1.04678214e+00 4.16819811e-01 1.29712522e+00 5.30056834e-01 4.99408484e-01 -1.10462153e+00 -5.51281869e-02 2.76437312e-01 7.86909521e-01 -1.10660005e+00 -3.31562638e-01 6.21871352e-01 -6.22565389e-01 1.49991059e+00 6.27735674e-01 2.31898963e-01 7.63667405e-01 -1.77496895e-01 1.14415765e+00 7.12893605e-01 -3.61018717e-01 1.51385948e-01 5.23817502e-02 7.25322485e-01 8.67969453e-01 -6.90316379e-01 -7.25238264e-01 -7.54127443e-01 -5.32581151e-01 1.50634199e-01 -2.52563030e-01 -1.41544908e-01 -1.82051230e-02 -1.16267991e+00 1.14851713e+00 4.76722181e-01 2.03188092e-01 -3.95129472e-01 3.62549245e-01 7.78368950e-01 8.12274337e-01 4.11646456e-01 8.93200755e-01 -6.89550042e-01 -2.45044261e-01 -4.69595850e-01 5.03250301e-01 7.81902671e-01 6.61382496e-01 6.37701273e-01 -2.22850874e-01 -9.40944910e-01 1.13905680e+00 3.04756254e-01 2.55376726e-01 5.77482760e-01 -1.08338332e+00 1.09959662e+00 8.67359102e-01 7.49058053e-02 -9.78275359e-01 -2.40526393e-01 -3.47098261e-01 -4.54325914e-01 -7.33442962e-01 5.61758459e-01 2.11060438e-02 -6.81363702e-01 1.44895625e+00 3.86963338e-01 -2.49117717e-01 1.10458881e-02 1.03965330e+00 1.11990547e+00 7.01001346e-01 -2.01355703e-02 2.36390471e-01 1.80825615e+00 -1.45501363e+00 -8.67976189e-01 -2.21132994e-01 1.11133039e+00 -8.35874498e-01 1.43694830e+00 -1.62236750e-01 -1.15150964e+00 -4.90234196e-01 -5.94136119e-01 -6.58173084e-01 -2.66910464e-01 1.63096845e-01 8.27802598e-01 4.11182880e-01 -9.66842890e-01 3.08424830e-01 -4.52084154e-01 -3.16078931e-01 1.08853623e-01 4.58081365e-01 -2.42327407e-01 -1.40518084e-01 -1.46783745e+00 8.82345140e-01 -2.01403908e-02 3.31911415e-01 -7.14316130e-01 -7.28757799e-01 -6.95481420e-01 4.83875647e-02 2.57476062e-01 -1.10531402e+00 1.50171018e+00 -6.29587293e-01 -1.77647793e+00 8.81289184e-01 -2.85868883e-01 -5.06518185e-01 -7.06863357e-03 -5.81628561e-01 1.74084574e-01 4.58654940e-01 3.51718873e-01 6.50305569e-01 7.73835599e-01 -4.59220290e-01 -2.88167536e-01 -2.96549678e-01 5.08875310e-01 2.99041331e-01 -4.65167522e-01 4.89363194e-01 -3.76448244e-01 -4.45218265e-01 8.59757587e-02 -7.30002463e-01 -4.17459309e-01 -3.91286969e-01 -3.32326859e-01 -1.01168227e+00 4.34346765e-01 -8.95610273e-01 1.41291058e+00 -1.68377984e+00 5.89859366e-01 -1.76893368e-01 1.54845953e-01 2.75944620e-01 -6.44870520e-01 7.33368814e-01 8.72854516e-02 2.16793358e-01 -1.01176806e-01 -5.60662985e-01 -5.95314354e-02 4.71093692e-02 -3.89348954e-01 1.57531023e-01 5.69901347e-01 1.11801434e+00 -1.01362741e+00 -5.38505077e-01 -3.39523703e-01 1.58023313e-01 -9.31590199e-01 5.89983761e-01 -4.85001594e-01 4.27998036e-01 -7.57831216e-01 4.04231280e-01 2.57263184e-01 -5.29071510e-01 -8.78996551e-02 -5.25581613e-02 1.02882020e-01 1.17097080e+00 -7.42212117e-01 1.90103447e+00 -5.67836404e-01 3.82252634e-01 1.24461040e-01 -9.47658002e-01 9.17452574e-01 5.42151332e-01 1.45223677e-01 -5.85319340e-01 1.44493625e-01 2.92121619e-01 7.77494311e-02 -1.15011501e+00 7.02740550e-01 2.20049843e-01 -2.45680794e-01 8.38565588e-01 3.26145321e-01 -3.15823883e-01 3.08304697e-01 4.94114906e-01 1.32869947e+00 -1.27493620e-01 -2.42211208e-01 -2.42861137e-01 9.76830482e-01 -9.24760252e-02 1.45672202e-01 8.70498598e-01 -6.79971874e-02 6.99231565e-01 7.01172948e-01 -5.43629050e-01 -8.57362092e-01 -5.70869207e-01 2.97759593e-01 1.68995547e+00 -4.92633820e-01 -6.60441697e-01 -8.08978796e-01 -9.67139006e-01 -7.89174810e-02 4.97375011e-01 -7.05616176e-01 1.83147974e-02 -9.63477612e-01 -5.79000235e-01 5.93369484e-01 3.63763690e-01 4.18196887e-01 -9.15823996e-01 -2.59286195e-01 3.10392290e-01 -8.09785366e-01 -1.12903869e+00 -8.58908415e-01 7.79495686e-02 -9.20455396e-01 -8.98254871e-01 -5.31122446e-01 -8.29532087e-01 3.71478409e-01 3.28181833e-01 1.41451228e+00 3.19892645e-01 7.70161450e-02 5.54690659e-01 -6.82727873e-01 2.15280369e-01 2.32311532e-01 6.65715992e-01 -4.39939111e-01 3.43741745e-01 2.34016299e-01 -2.46638089e-01 -7.99715161e-01 1.91684797e-01 -7.21539736e-01 -2.80295521e-01 -1.58874486e-02 1.18048441e+00 1.91575557e-01 -8.22365642e-01 8.98274720e-01 -7.12134838e-01 1.23573518e+00 -4.92498577e-01 -8.84026811e-02 5.38710296e-01 -1.73617397e-02 5.62296271e-01 5.15258431e-01 -3.30338150e-01 -8.98290277e-01 -1.75134614e-01 -4.50665057e-01 -6.24258481e-02 4.40178514e-01 6.91292465e-01 2.46550754e-01 -6.25804358e-04 6.68375373e-01 -1.28093557e-02 -1.88522786e-01 -7.42589593e-01 6.03271127e-01 7.20016539e-01 1.02081252e-02 -7.42137849e-01 4.68670130e-01 2.25311428e-01 -5.54375947e-01 -5.91295302e-01 -1.42233384e+00 -8.37083995e-01 -5.95179856e-01 -2.51290891e-02 1.00799704e+00 -5.92032254e-01 -9.95096087e-01 2.64410436e-01 -1.62960279e+00 -3.69546562e-02 -2.42263377e-01 2.36195773e-01 -3.01550746e-01 4.27849591e-01 -8.97084117e-01 -7.20061660e-01 -8.24756265e-01 -1.28914392e+00 1.05154014e+00 -7.29272440e-02 -2.92494446e-01 -9.52947855e-01 1.80169120e-01 1.25045121e+00 7.84122825e-01 -5.05572140e-01 1.10060418e+00 -8.45092297e-01 -7.97112107e-01 -2.06041142e-01 -2.16720626e-01 4.79903907e-01 -3.18686843e-01 -4.60693061e-01 -9.03937995e-01 -1.09142125e-01 3.48256886e-01 -7.64163315e-01 1.19569933e+00 -6.75141886e-02 1.04044235e+00 -3.52163881e-01 2.25125268e-01 1.82182506e-01 9.30319667e-01 -3.53240877e-01 4.23220158e-01 3.71648669e-01 7.87184775e-01 1.01968455e+00 3.22682500e-01 1.97601840e-01 1.01538479e+00 8.32844257e-01 3.07213545e-01 3.79700601e-01 5.79794608e-02 -5.56152575e-02 2.20731422e-01 1.54365778e+00 2.08461687e-01 -2.15049699e-01 -8.56521904e-01 8.88479292e-01 -2.11833692e+00 -8.05359364e-01 -4.98328388e-01 2.05380321e+00 1.00716639e+00 -1.95866153e-01 -6.60861358e-02 -7.20202252e-02 3.86107713e-01 2.59126097e-01 -2.81758904e-01 -7.79060662e-01 -2.51167100e-02 4.01991934e-01 3.04751545e-02 8.35736275e-01 -1.21784782e+00 1.01005280e+00 6.17233706e+00 8.12699556e-01 -8.44276607e-01 5.02593935e-01 4.83620524e-01 -2.60167606e-02 -5.86470306e-01 1.20116100e-01 -7.60630310e-01 1.03271641e-01 1.02758133e+00 1.92178577e-01 4.18963164e-01 5.18780053e-01 -3.17133591e-02 7.19428733e-02 -1.20327187e+00 8.55316103e-01 3.17416161e-01 -1.53014839e+00 -4.72692475e-02 -6.01842284e-01 5.09344876e-01 1.07848085e-01 6.89606890e-02 7.49626100e-01 8.70148540e-02 -9.36143219e-01 4.35779035e-01 5.38374186e-01 1.87848508e-01 -3.07777137e-01 8.96176338e-01 1.21519111e-01 -1.00531614e+00 -2.11341739e-01 -2.79882282e-01 -3.40496361e-01 2.68043071e-01 1.91360474e-01 -7.71805584e-01 3.56268793e-01 7.13941813e-01 3.11895490e-01 -9.41400409e-01 8.46087337e-01 -3.33755791e-01 8.51891339e-01 -2.85842061e-01 -3.69575351e-01 4.94909704e-01 4.89176996e-03 4.03216004e-01 1.08098292e+00 -1.59323707e-01 1.43532492e-02 1.75836489e-01 6.49016142e-01 -4.00810152e-01 5.27382910e-01 -2.96139300e-01 -3.42429765e-02 1.68981031e-01 1.08736610e+00 -1.03405342e-01 -4.22803104e-01 -3.27496558e-01 1.11502826e+00 8.72556567e-01 4.38983381e-01 -3.71713042e-01 -5.61165094e-01 5.38297832e-01 -1.85209006e-01 5.58751404e-01 -1.50273845e-01 -2.58341908e-01 -1.50141490e+00 3.72970045e-01 -1.06835568e+00 6.26952827e-01 -4.03218657e-01 -1.60611260e+00 7.64453828e-01 -3.37641299e-01 -7.49248862e-01 -3.97794843e-01 -5.00205398e-01 -6.78334117e-01 9.97072339e-01 -1.50096059e+00 -1.08384490e+00 2.55292326e-01 5.12176752e-01 7.56902933e-01 -1.36881903e-01 1.00402117e+00 5.54411948e-01 -6.11901999e-01 7.56133854e-01 -4.00707545e-03 2.57218063e-01 6.20005846e-01 -1.14759791e+00 5.70601165e-01 3.97508442e-01 -2.42684875e-02 8.85577023e-01 4.28766936e-01 -1.83176607e-01 -1.64352906e+00 -8.26875210e-01 1.69727921e+00 -8.23271453e-01 9.80271161e-01 -4.98888940e-01 -9.42513764e-01 5.41246057e-01 4.87799257e-01 -2.43043061e-02 9.05779123e-01 5.83830297e-01 -7.37364590e-01 -1.04700178e-02 -8.52773547e-01 4.70178425e-01 5.51393270e-01 -1.21364832e+00 -7.43352234e-01 7.31176317e-01 1.14704037e+00 -1.43805385e-01 -9.69096661e-01 9.31191891e-02 2.66662747e-01 -6.19270146e-01 1.02979779e+00 -1.23529422e+00 3.92925948e-01 -2.43867829e-01 -3.31072569e-01 -9.58009899e-01 -2.73400277e-01 -6.71380639e-01 -1.75969198e-01 1.13203120e+00 7.44725347e-01 -3.24792624e-01 8.16701055e-01 6.42673850e-01 -2.40205690e-01 -1.11151361e+00 -1.13380444e+00 -2.72052199e-01 1.55072808e-01 -3.49834830e-01 4.01416510e-01 7.39664853e-01 1.58421099e-01 1.16297257e+00 -3.70826542e-01 -2.48173907e-01 1.75480753e-01 1.04036219e-01 5.20921886e-01 -7.43369281e-01 -4.42659706e-01 -2.44800523e-01 -1.22396275e-01 -1.53718853e+00 1.61823794e-01 -1.21153009e+00 -1.50427148e-01 -1.65152681e+00 2.80593842e-01 -2.64455050e-01 -1.69897601e-01 1.49324179e-01 -5.92414498e-01 -1.44515643e-02 3.68755043e-01 1.30941957e-01 -1.29402280e+00 8.74419570e-01 1.29913390e+00 -3.74255836e-01 -9.18833762e-02 -6.82811663e-02 -6.18499994e-01 3.02870244e-01 7.05390275e-01 -6.05954468e-01 -2.09762067e-01 -1.20191705e+00 7.78581381e-01 1.74913898e-01 1.91210076e-01 -3.65982443e-01 4.32715446e-01 1.91605181e-01 -3.64007831e-01 -7.02815473e-01 5.28486431e-01 -1.61834419e-01 -7.85203099e-01 2.74359226e-01 -1.02387071e+00 3.79088640e-01 -2.02711329e-01 3.78693819e-01 -4.60939825e-01 -7.35101163e-01 2.92018980e-01 -3.09237689e-01 -1.69781387e-01 -3.29258442e-02 -5.68870485e-01 6.70757294e-01 3.06739897e-01 3.18622470e-01 -5.91888487e-01 -4.78745431e-01 -5.46748281e-01 8.65587473e-01 -3.96415591e-01 5.97163439e-01 9.03562129e-01 -1.32950127e+00 -1.05296350e+00 -7.33036399e-02 2.16004372e-01 -2.61098117e-01 4.97906566e-01 9.85269606e-01 -4.67990041e-01 8.28846514e-01 3.90065283e-01 -5.21986187e-01 -1.19523609e+00 2.78439581e-01 5.00005841e-01 -8.65600288e-01 -5.68348840e-02 1.57508838e+00 -9.83134285e-02 -1.13794947e+00 2.94890791e-01 -3.20463210e-01 -5.54328382e-01 1.87040851e-01 5.16889036e-01 2.91649222e-01 3.16162944e-01 -4.12392199e-01 -3.88874620e-01 5.31989217e-01 -5.18375397e-01 -1.93364024e-02 1.13498247e+00 -1.18382283e-01 -5.35547316e-01 4.12533998e-01 1.54303432e+00 -3.31649482e-01 -4.28356498e-01 -5.16412556e-01 4.12982196e-01 -2.84292698e-01 -1.22316971e-01 -6.10108554e-01 -6.46300614e-01 1.23856771e+00 4.87425417e-01 3.21334749e-01 5.98299861e-01 1.79000258e-01 1.13839543e+00 1.00571668e+00 -1.86979815e-01 -1.10455787e+00 7.45107293e-01 1.18193841e+00 1.15909159e+00 -1.30284894e+00 -3.03129524e-01 -2.36597523e-01 -6.51780844e-01 9.38444197e-01 6.34570539e-01 -2.04489276e-01 4.15971577e-01 -4.90739703e-01 2.25955248e-01 -5.05913317e-01 -1.43342435e+00 -3.37296814e-01 4.18088347e-01 1.65767759e-01 8.96184564e-01 -1.90001577e-01 -5.08249640e-01 4.27292585e-01 -1.41635835e-01 -3.85731131e-01 1.82241783e-01 8.26304734e-01 -3.83002043e-01 -1.38841546e+00 -3.29332352e-01 4.97568816e-01 -8.60038221e-01 -5.02246618e-01 -3.50727826e-01 4.75952588e-02 -2.35336989e-01 1.44482839e+00 -1.04293101e-01 -4.79919910e-01 3.96514595e-01 2.92800575e-01 2.97853947e-01 -7.65966952e-01 -1.48400211e+00 -4.05586004e-01 6.49460733e-01 -4.80618060e-01 -2.70796061e-01 -6.09242082e-01 -7.99547970e-01 7.44539872e-02 -5.79686642e-01 5.82011402e-01 5.94164670e-01 1.04044878e+00 6.21850193e-01 5.06295800e-01 6.01758361e-01 -3.29101264e-01 -9.61737990e-01 -1.33353961e+00 1.91467211e-01 5.42899430e-01 4.28875118e-01 -2.34384194e-01 -2.57273644e-01 -1.61442891e-01]	[11.32333755493164, 8.053642272949219]
a3f44305-cf82-46d9-bd70-c9454d7dd720	unichart-a-universal-vision-language	2305.14761	null	https://arxiv.org/abs/2305.14761v1	https://arxiv.org/pdf/2305.14761v1.pdf	UniChart: A Universal Vision-language Pretrained Model for Chart Comprehension and Reasoning	Charts are very popular for analyzing data, visualizing key insights and answering complex reasoning questions about data. To facilitate chart-based data analysis using natural language, several downstream tasks have been introduced recently such as chart question answering and chart summarization. However, most of the methods that solve these tasks use pretraining on language or vision-language tasks that do not attempt to explicitly model the structure of the charts (e.g., how data is visually encoded and how chart elements are related to each other). To address this, we first build a large corpus of charts covering a wide variety of topics and visual styles. We then present UniChart, a pretrained model for chart comprehension and reasoning. UniChart encodes the relevant text, data, and visual elements of charts and then uses a chart-grounded text decoder to generate the expected output in natural language. We propose several chart-specific pretraining tasks that include: (i) low-level tasks to extract the visual elements (e.g., bars, lines) and data from charts, and (ii) high-level tasks to acquire chart understanding and reasoning skills. We find that pretraining the model on a large corpus with chart-specific low- and high-level tasks followed by finetuning on three down-streaming tasks results in state-of-the-art performance on three downstream tasks.	['Shafiq Joty', 'Enamul Hoque', 'Xuan Long Do', 'Parsa Kavehzadeh', 'Ahmed Masry']	2023-05-24	null	null	null	null	['chart-question-answering', 'chart-question-answering']	['computer-code', 'computer-vision']	[ 3.00423384e-01 1.88755855e-01 9.64426771e-02 -4.27183777e-01 -7.97596753e-01 -8.53734374e-01 6.05049253e-01 7.79169202e-01 2.16850683e-01 1.11868240e-01 7.86553442e-01 -8.49216282e-01 3.68195683e-01 -7.07477748e-01 -9.65259910e-01 7.99923390e-02 -2.55417526e-01 4.17656302e-01 2.08138198e-01 -1.42046943e-01 5.40491700e-01 3.65488350e-01 -1.51885569e+00 1.22275376e+00 1.02535462e+00 8.93316507e-01 8.78262967e-02 1.21453094e+00 -8.67079198e-01 1.36907160e+00 -9.51695204e-01 -4.63001817e-01 -2.16685101e-01 -6.93866432e-01 -8.42115283e-01 2.11005911e-01 6.10652566e-01 -4.64962721e-01 1.10270292e-01 8.04324687e-01 1.94898307e-01 -4.77357469e-02 9.19606030e-01 -1.16106391e+00 -1.07839727e+00 6.46217704e-01 -8.36938143e-01 1.37157470e-01 8.64153206e-01 4.03044283e-01 1.15316403e+00 -8.51771474e-01 7.54187584e-01 1.47981441e+00 3.15172672e-01 3.17919016e-01 -1.20125079e+00 -2.46377781e-01 5.47010541e-01 6.11433648e-02 -6.74720228e-01 -2.74607688e-01 7.66965032e-01 -6.94205105e-01 1.25605059e+00 3.89922142e-01 6.92656219e-01 7.99684525e-01 7.07795992e-02 1.16119254e+00 5.93093932e-01 -5.93331933e-01 3.21513206e-01 -8.56489018e-02 2.32211262e-01 9.23463821e-01 2.58980840e-01 -6.33484006e-01 -5.93932331e-01 2.13990033e-01 9.04688299e-01 -2.10328892e-01 -2.08693203e-02 -4.59709764e-01 -1.36747456e+00 6.91575646e-01 6.48506522e-01 -4.22823541e-02 -2.63764262e-01 5.28763272e-02 7.51555979e-01 2.71978676e-01 -6.53939024e-02 7.31959164e-01 -1.20596111e-01 2.49681957e-02 -6.51159525e-01 2.39118218e-01 8.32464516e-01 1.44932842e+00 6.93298340e-01 2.30414346e-02 -6.98360920e-01 4.60841388e-01 3.41165394e-01 4.13388908e-01 1.50563315e-01 -6.61301434e-01 1.40975618e+00 1.12165666e+00 -4.54227701e-02 -8.12829912e-01 -5.90692639e-01 -1.58566572e-02 -7.93619037e-01 3.04430932e-01 4.57343429e-01 -2.52717078e-01 -1.36080909e+00 1.17492235e+00 7.99320489e-02 -5.75652242e-01 1.41171426e-01 5.99362493e-01 1.18340266e+00 1.10764217e+00 1.40924841e-01 7.78991356e-02 1.81947851e+00 -1.16592371e+00 -7.67523408e-01 -3.62556338e-01 8.88238668e-01 -6.21992886e-01 1.89319158e+00 3.42427790e-01 -1.14905131e+00 -6.67383552e-01 -1.13538241e+00 -1.00064945e+00 -6.33167267e-01 2.48073190e-01 4.92004663e-01 3.10945064e-01 -9.41438079e-01 -1.02999412e-01 -6.89410865e-01 -3.83802235e-01 7.86001801e-01 -3.29103440e-01 -6.48932680e-02 5.41785806e-02 -6.49100244e-01 5.98444760e-01 3.37122560e-01 -1.89880729e-01 -6.04834914e-01 -8.38538527e-01 -1.27393115e+00 5.41709542e-01 5.77516556e-01 -7.45775521e-01 1.53948915e+00 -5.59698403e-01 -1.15936518e+00 7.04359055e-01 -4.62049127e-01 -3.11289489e-01 5.05863786e-01 -4.13249999e-01 -2.19285220e-01 3.30917448e-01 1.44863680e-01 8.41622829e-01 6.31395519e-01 -1.48495615e+00 -5.70648551e-01 -3.17745566e-01 2.32474998e-01 1.94902390e-01 -9.15737599e-02 -4.15792204e-02 -8.66231203e-01 -6.54389322e-01 -1.47251055e-01 -3.57002199e-01 -1.47245470e-02 3.19685727e-01 -8.27308416e-01 -1.39576614e-01 7.88452029e-01 -9.44893122e-01 1.49203694e+00 -2.00986290e+00 -7.49028753e-03 2.25857124e-01 2.27492005e-01 1.22070588e-01 -2.24145442e-01 8.50519717e-01 -1.38728186e-01 2.28691489e-01 -3.53839368e-01 -1.15255445e-01 1.82053879e-01 -1.58981413e-01 -6.48558319e-01 -4.13885534e-01 5.45859933e-01 1.25135720e+00 -6.56526804e-01 -5.97217202e-01 3.03091675e-01 1.36821762e-01 -7.98187494e-01 5.27327716e-01 -7.95420170e-01 1.88001156e-01 -3.35497290e-01 6.31415486e-01 2.90953755e-01 -6.57863557e-01 -6.86531886e-02 -1.98926970e-01 -1.39528990e-01 3.51623714e-01 -9.62564468e-01 1.80861199e+00 -4.38523412e-01 9.83911932e-01 -3.74064624e-01 -6.14149988e-01 9.39376593e-01 4.07907180e-02 -1.69390529e-01 -1.03194833e+00 -1.96948126e-01 -3.45194668e-01 -1.65005594e-01 -9.32219028e-01 6.88409746e-01 3.30430239e-01 -2.80381620e-01 4.93701756e-01 -1.72639325e-01 -4.10744160e-01 7.54504800e-01 6.13430321e-01 9.82634187e-01 1.47833750e-01 6.62973404e-01 2.18787655e-01 3.96975577e-01 5.98047853e-01 -3.65486145e-01 6.53351247e-01 2.18824744e-01 7.41680443e-01 1.35870135e+00 -5.50241530e-01 -1.18591309e+00 -9.61483181e-01 4.91958052e-01 1.40331829e+00 -1.70190364e-01 -8.36434186e-01 -7.07114339e-01 -4.51131284e-01 -1.26689440e-02 1.24435532e+00 -8.37338567e-01 1.88184872e-01 -5.49141049e-01 -1.07610688e-01 2.74499208e-01 1.11239326e+00 4.35979933e-01 -1.37550151e+00 -1.05421484e+00 -1.43384799e-01 8.82366374e-02 -1.00830054e+00 -4.63072360e-01 1.90770924e-01 -8.93717110e-01 -1.02708662e+00 -5.46526849e-01 -8.67170691e-01 9.78448749e-01 2.17061132e-01 1.37344599e+00 4.35549766e-02 -1.86537758e-01 2.52687544e-01 -3.19075942e-01 -9.56579804e-01 -3.94204050e-01 6.12827241e-02 -1.08127940e+00 -4.40949261e-01 1.81056976e-01 -9.88277793e-02 -2.64982790e-01 -1.96972311e-01 -1.18023455e+00 6.71574593e-01 7.69616783e-01 2.69646645e-01 4.77872491e-01 -3.82658184e-01 2.31497511e-01 -1.23607540e+00 1.12226844e+00 -3.84892188e-02 -5.01145720e-01 7.09524155e-01 -1.07401311e-01 5.56116164e-01 9.66331124e-01 -6.39381185e-02 -1.07750607e+00 5.43229766e-02 -1.98184792e-02 -2.37625152e-01 -3.32464129e-01 8.18378329e-01 -2.22472474e-01 9.59320962e-01 7.72816062e-01 4.36899215e-02 -3.45223039e-01 -2.91042715e-01 9.53400075e-01 2.57029444e-01 7.98824370e-01 -5.84274650e-01 7.52532363e-01 4.15897846e-01 -1.68749973e-01 -7.59339452e-01 -7.35833228e-01 -3.00431609e-01 -7.73272693e-01 -2.10247055e-01 1.33753371e+00 -7.36360848e-01 -8.00593197e-01 -1.35110527e-01 -1.43932331e+00 -5.31294823e-01 -4.84351277e-01 -5.49980029e-02 -5.45076489e-01 3.52891907e-02 -3.47009063e-01 -7.77360678e-01 -2.94908762e-01 -1.17772388e+00 1.28146398e+00 2.35149801e-01 -4.61130738e-01 -8.10602665e-01 -9.73340049e-02 8.53110012e-03 1.16876103e-01 6.06432378e-01 1.66125846e+00 -3.15555453e-01 -7.21728683e-01 3.41819748e-02 -7.64934003e-01 -2.17162505e-01 7.74184009e-03 4.32731032e-01 -7.56103456e-01 8.48801211e-02 -6.77639723e-01 -6.19801164e-01 9.28381443e-01 1.64614245e-01 1.47539401e+00 -3.92525613e-01 -2.24982828e-01 6.22124732e-01 1.15825367e+00 5.84824026e-01 6.08468592e-01 1.80656433e-01 1.15173876e+00 8.00354779e-01 8.93612355e-02 3.64371687e-01 8.32364261e-01 4.94271368e-02 3.80204529e-01 -3.85321081e-01 -1.18574016e-01 -8.57746840e-01 2.67358959e-01 6.10410273e-01 8.74122679e-02 -3.37947577e-01 -1.20342791e+00 4.32999462e-01 -1.89959073e+00 -8.83295715e-01 -2.88395047e-01 1.69080985e+00 4.73682910e-01 4.06511307e-01 3.96505594e-01 -6.54780418e-02 3.98741335e-01 4.95230407e-01 -8.20068181e-01 -9.08451021e-01 2.29850665e-01 -1.66473165e-01 9.95226428e-02 2.25191027e-01 -9.29907441e-01 8.48963499e-01 6.26278591e+00 8.85549858e-02 -9.43447828e-01 -6.63286269e-01 6.71670735e-01 1.82576403e-01 -6.24104619e-01 4.36030328e-03 -3.79324883e-01 -1.98779926e-01 7.06219673e-01 -3.12102467e-01 2.42582798e-01 8.67738545e-01 1.61478728e-01 -1.47360072e-01 -1.58625770e+00 1.01989043e+00 3.77166182e-01 -1.72574139e+00 8.55246365e-01 -3.95704597e-01 5.43988764e-01 -5.76685548e-01 -1.03196599e-01 5.37096441e-01 3.92697990e-01 -1.24709129e+00 1.06784189e+00 5.20896614e-01 1.00644779e+00 -5.60029209e-01 1.69129327e-01 6.34792745e-02 -1.40200627e+00 -1.40624503e-02 -1.57593653e-01 -2.46184304e-01 1.64621413e-01 1.46436408e-01 -8.34493756e-01 5.90631366e-01 5.41036606e-01 7.79136479e-01 -1.20682073e+00 8.69633198e-01 -5.73183060e-01 4.40750182e-01 2.04080299e-01 -4.05200839e-01 3.36279541e-01 4.34225872e-02 8.37998986e-02 1.51033437e+00 1.81115255e-01 -3.54433879e-02 -4.57645282e-02 9.61889625e-01 -2.34690294e-01 1.73724890e-01 -7.20181346e-01 -3.67536277e-01 4.99902032e-02 8.42051327e-01 -9.81602550e-01 -7.78875411e-01 -8.77797306e-01 6.70960248e-01 5.34234405e-01 6.69130981e-01 -5.82181513e-01 -7.48945653e-01 1.89392924e-01 1.23086296e-01 5.13944030e-01 -3.55345726e-01 -8.63366485e-01 -1.18005943e+00 2.90711403e-01 -1.01640856e+00 8.32751095e-01 -1.47609913e+00 -8.00725639e-01 5.33826768e-01 1.39465600e-01 -1.27198410e+00 -3.90471756e-01 -7.99633265e-01 -8.88742149e-01 6.26002967e-01 -1.27859938e+00 -8.42376053e-01 -6.22189164e-01 4.68285888e-01 9.02315915e-01 7.75802583e-02 6.00693107e-01 -3.02841485e-01 -4.71620470e-01 8.70398134e-02 -2.69539922e-01 6.73755646e-01 5.73134661e-01 -1.70959008e+00 1.17120898e+00 1.00498092e+00 2.82991588e-01 6.21923268e-01 6.73840880e-01 -5.17268956e-01 -1.59086633e+00 -1.05156636e+00 5.26723981e-01 -5.50461709e-01 6.84099197e-01 -9.90679562e-01 -1.14362097e+00 9.84700322e-01 5.93658805e-01 -3.04665744e-01 5.10063946e-01 8.62286165e-02 -7.47412741e-01 -4.75274585e-02 -5.47306836e-01 9.46385682e-01 1.08295190e+00 -6.46026969e-01 -1.04566276e+00 2.10094512e-01 9.44917858e-01 -7.28137255e-01 -3.15903634e-01 -1.50729671e-01 5.48981011e-01 -1.02563298e+00 7.56967127e-01 -1.00095797e+00 1.19843280e+00 -3.74700993e-01 1.82843193e-01 -1.33983827e+00 -1.14645451e-01 -5.26064217e-01 -2.64669895e-01 1.00019085e+00 7.85454452e-01 -4.37941216e-02 5.74955463e-01 6.35417581e-01 -3.39560360e-01 -5.04176021e-01 -1.39433339e-01 -2.70908147e-01 -1.88803017e-01 -5.29886127e-01 6.42950416e-01 6.83709741e-01 4.40544933e-01 8.13310623e-01 -2.02283710e-02 1.43091813e-01 4.34782095e-02 4.95072842e-01 1.29058456e+00 -1.08432579e+00 2.41016448e-01 -6.48361325e-01 1.34329041e-02 -1.29447103e+00 -3.21899861e-01 -8.93382668e-01 -1.06752940e-01 -2.63984084e+00 -9.65819322e-03 4.47898597e-01 3.61875087e-01 4.66151237e-01 -3.24080050e-01 -5.01181424e-01 5.72321117e-01 1.68014318e-01 -8.98512244e-01 2.14573175e-01 1.64115119e+00 -2.90822715e-01 -6.68913782e-01 -3.87933850e-01 -1.01326036e+00 5.35653353e-01 4.65353161e-01 1.04484268e-01 -8.54477823e-01 -8.73999417e-01 5.51894009e-01 4.01842028e-01 1.31896749e-01 -1.00511563e+00 4.09452051e-01 -2.01498583e-01 9.83647466e-01 -1.23862994e+00 -1.15741909e-01 -5.88581443e-01 -5.55835724e-01 3.75389308e-01 -9.80812192e-01 8.07068229e-01 5.41334569e-01 4.37168449e-01 -3.68691593e-01 1.49793595e-01 2.82190502e-01 -3.47105414e-01 -7.87575185e-01 -1.69050738e-01 -3.96039784e-01 3.35271984e-01 8.49597454e-01 -2.47930720e-01 -8.86387289e-01 -6.60198390e-01 -4.89916772e-01 7.49738216e-01 3.00778925e-01 5.88947177e-01 7.16885567e-01 -1.11190605e+00 -6.12355888e-01 3.10922146e-01 6.97041154e-01 5.68466961e-01 5.56269661e-03 2.34091133e-01 -1.11422694e+00 6.47882342e-01 -4.49951351e-01 -5.97552717e-01 -9.41653490e-01 9.65419173e-01 -2.16996580e-01 -3.43594164e-01 -8.99643838e-01 5.28987646e-01 5.67616284e-01 -1.23758897e-01 4.53206748e-01 -1.36849821e+00 -4.48093712e-01 2.98062980e-01 9.07120943e-01 1.35891691e-01 -4.81364205e-02 -6.41391277e-02 -2.22711787e-01 4.49826270e-01 -1.42585859e-01 -1.31661370e-01 1.21957588e+00 6.61538765e-02 1.02386475e-01 8.16811740e-01 1.03525293e+00 -1.75261907e-02 -1.43728864e+00 8.84180740e-02 4.48875308e-01 -1.93638697e-01 -5.74156702e-01 -9.27848935e-01 -4.82780337e-01 1.35136175e+00 -2.20995888e-01 5.72413027e-01 1.17501509e+00 6.84162378e-02 3.04321676e-01 7.73949921e-01 -2.83776194e-01 -8.54397893e-01 3.68766516e-01 4.75302994e-01 1.47929406e+00 -1.12815845e+00 1.47151887e-01 -4.40251738e-01 -1.20779192e+00 1.53525174e+00 9.02786255e-01 3.70999761e-02 1.31812856e-01 3.80728900e-01 3.50262970e-01 -5.21387577e-01 -1.16788864e+00 -2.67972022e-01 6.18275821e-01 6.16061032e-01 7.10173488e-01 -5.10216951e-02 2.06734255e-01 9.67807919e-02 -5.49620569e-01 -1.67205676e-01 6.79592490e-01 9.65222478e-01 -4.84057218e-01 -3.23905885e-01 -6.20361865e-01 8.50325644e-01 2.21695825e-01 -5.38146868e-02 -8.06298673e-01 9.99846220e-01 -5.43884933e-01 7.19455659e-01 4.98559654e-01 6.79175630e-02 8.95601630e-01 2.65714556e-01 2.19323620e-01 -6.75649524e-01 -4.29917037e-01 4.11186554e-02 1.14948846e-01 -4.77958292e-01 -2.35561594e-01 -5.10241210e-01 -1.62534046e+00 8.75700042e-02 5.49617231e-01 -4.32951637e-02 4.22549367e-01 7.20586300e-01 3.68449628e-01 1.10291970e+00 -1.51221737e-01 -5.07228374e-01 3.03540360e-02 -7.86434233e-01 -3.63961756e-02 5.78905761e-01 6.67306125e-01 -1.26498351e-02 2.44508609e-01 5.26575148e-01]	[11.226428985595703, 2.065953254699707]
2c132b4d-8714-4d06-b5db-3086af40ee81	dialogue-act-classification-in-group-chats	1908.01821	null	https://arxiv.org/abs/1908.01821v1	https://arxiv.org/pdf/1908.01821v1.pdf	Dialogue Act Classification in Group Chats with DAG-LSTMs	Dialogue act (DA) classification has been studied for the past two decades and has several key applications such as workflow automation and conversation analytics. Researchers have used, to address this problem, various traditional machine learning models, and more recently deep neural network models such as hierarchical convolutional neural networks (CNNs) and long short-term memory (LSTM) networks. In this paper, we introduce a new model architecture, directed-acyclic-graph LSTM (DAG-LSTM) for DA classification. A DAG-LSTM exploits the turn-taking structure naturally present in a multi-party conversation, and encodes this relation in its model structure. Using the STAC corpus, we show that the proposed method performs roughly 0.8% better in accuracy and 1.2% better in macro-F1 score when compared to existing methods. The proposed method is generic and not limited to conversation applications.	['Brendan Fahy', 'Mu-Hsin Wei', 'Ozan İrsoy', 'Haimin Zhang', 'Rakesh Gosangi', 'Peter Lund', 'Neophytos Nephytou', 'Duccio Pappadopulo', 'Camilo Ortiz']	2019-08-02	null	null	null	null	['dialogue-act-classification']	['natural-language-processing']	[ 1.79332793e-01 4.73376453e-01 2.18191966e-02 -5.20554602e-01 -1.22335128e-01 -3.22089612e-01 9.71240401e-01 3.80830973e-01 -2.86783338e-01 8.42808247e-01 6.25572562e-01 -4.87546802e-01 1.38019502e-01 -7.59371936e-01 -1.16290256e-01 -4.40510184e-01 -2.50404596e-01 5.23708761e-01 1.48573339e-01 -4.29781318e-01 1.72438160e-01 4.30628240e-01 -1.14844012e+00 6.72997534e-01 7.32047081e-01 9.81889904e-01 -1.63025819e-02 9.72985029e-01 -5.71048737e-01 1.62854373e+00 -8.67113411e-01 -4.26664412e-01 -3.37732464e-01 -5.31587303e-01 -1.52478504e+00 -3.73916030e-02 -1.22793190e-01 -3.27622950e-01 -5.51136076e-01 5.11174917e-01 2.63216227e-01 4.12856519e-01 3.29883397e-01 -1.04555213e+00 -1.51295021e-01 7.66088605e-01 -2.07188353e-01 1.45970866e-01 5.52651823e-01 1.38437331e-01 1.04873204e+00 -4.51210409e-01 3.99141431e-01 1.49161530e+00 5.38008392e-01 6.84968054e-01 -8.40375364e-01 -3.82420838e-01 1.44304842e-01 3.14541191e-01 -6.33735359e-01 -2.07898200e-01 8.66050065e-01 -3.59271646e-01 1.41437674e+00 5.72938025e-02 7.35066593e-01 1.39270592e+00 5.04526138e-01 1.19798017e+00 1.17677939e+00 -6.59179747e-01 -9.15531069e-02 -1.98203832e-01 7.80179679e-01 1.03039300e+00 -4.22248423e-01 -2.86671340e-01 -5.72635472e-01 -3.29105496e-01 6.02579355e-01 -1.12477168e-01 4.10794429e-02 3.81950259e-01 -1.08354652e+00 9.64892924e-01 4.24655378e-01 8.00385535e-01 -4.47053522e-01 8.67765173e-02 8.37612867e-01 4.55727428e-01 5.25611162e-01 5.36149621e-01 -3.84565055e-01 -7.13818014e-01 -3.77111226e-01 2.71646500e-01 1.26864946e+00 7.14249432e-01 4.22559112e-01 1.71216428e-01 -4.03899223e-01 9.79787767e-01 8.82961452e-02 -3.34470123e-01 7.58752167e-01 -9.59829569e-01 5.87558985e-01 8.77360404e-01 -1.17794983e-01 -1.03578389e+00 -9.36256468e-01 -2.15243623e-01 -1.20287848e+00 -2.93496042e-01 6.15828514e-01 -3.96295130e-01 -4.22501504e-01 1.46058059e+00 1.50348216e-01 -1.62901476e-01 1.00626990e-01 4.87356603e-01 1.02783537e+00 8.03962171e-01 4.12185863e-02 -3.04215282e-01 1.29262471e+00 -1.25159252e+00 -1.20651579e+00 -1.20089769e-01 8.47434759e-01 -5.17862201e-01 1.00183380e+00 3.07517380e-01 -1.10670424e+00 -4.45330083e-01 -9.00437832e-01 -2.83707470e-01 -4.30399597e-01 -8.13536048e-02 7.49225318e-01 5.83389342e-01 -9.77137268e-01 5.44754386e-01 -5.88700235e-01 -5.37285268e-01 1.23342268e-01 4.37227368e-01 -1.16889879e-01 2.79554009e-01 -1.39183807e+00 9.71900225e-01 3.06521267e-01 3.68148059e-01 -7.56729782e-01 -1.64533351e-02 -8.85537863e-01 2.87858605e-01 3.48041356e-01 -4.41301793e-01 1.75954425e+00 -1.06899691e+00 -2.29336166e+00 6.18074477e-01 -3.45595181e-01 -8.38395417e-01 2.19972476e-01 -2.65232529e-02 -2.07085893e-01 -1.83708698e-01 -4.93307412e-01 3.28401715e-01 3.67072254e-01 -9.53047335e-01 -6.93366230e-01 -4.89094943e-01 6.37681007e-01 9.43602249e-02 -3.25244009e-01 2.30437770e-01 2.27319196e-01 -3.57453316e-01 2.03937106e-02 -9.50991452e-01 -2.04095930e-01 -7.43260860e-01 -5.88693082e-01 -8.13233912e-01 9.19440508e-01 -8.55118573e-01 1.39607191e+00 -1.51594734e+00 1.14770159e-01 -7.43125454e-02 4.48251545e-01 5.11407733e-01 7.91729018e-02 8.50773752e-01 2.50405252e-01 2.33338755e-02 -1.68671161e-01 -4.35510457e-01 1.78993610e-03 4.75812465e-01 6.48921281e-02 1.83217674e-01 -5.66789135e-02 9.05552149e-01 -8.01893294e-01 -2.19373330e-01 2.02604845e-01 2.02070370e-01 -2.70147063e-02 5.47554553e-01 -3.80509019e-01 5.98389030e-01 -2.57174373e-01 3.32268208e-01 1.84346408e-01 -2.66631365e-01 8.21179569e-01 2.30978698e-01 -2.07037896e-01 7.52160847e-01 -3.84223431e-01 1.61537850e+00 -7.32972026e-01 1.08883512e+00 4.53991741e-02 -1.07923603e+00 9.99086499e-01 7.33229637e-01 2.61747181e-01 -5.65081120e-01 3.20028424e-01 3.12281232e-02 4.83585089e-01 -5.94032884e-01 5.90532780e-01 6.37391657e-02 -1.54301552e-02 6.10101461e-01 2.60402292e-01 3.99519175e-01 1.58736363e-01 7.11729154e-02 1.11648345e+00 -3.35210949e-01 6.34024382e-01 -1.67379960e-01 1.01988149e+00 -2.19241068e-01 4.35218215e-01 5.26592016e-01 -2.85241365e-01 -3.34893651e-02 9.92468715e-01 -9.12752271e-01 -6.38984144e-01 -2.12086648e-01 3.34882051e-01 1.39590549e+00 -5.16525507e-01 -5.57965100e-01 -8.99255097e-01 -8.10436130e-01 -3.38165671e-01 7.53366530e-01 -5.54687023e-01 -2.29010926e-04 -1.08959770e+00 -3.09745669e-01 1.02464271e+00 5.85542858e-01 9.89205897e-01 -1.46348643e+00 -3.64230752e-01 4.74120677e-01 -2.50946641e-01 -1.30498731e+00 -5.08950576e-02 1.71230938e-02 -8.36111486e-01 -1.00318730e+00 -4.79616016e-01 -8.51836741e-01 8.25925693e-02 -6.01985268e-02 1.23576140e+00 1.13473415e-01 1.66787848e-01 1.82351425e-01 -3.35568875e-01 -2.50632077e-01 -7.60883152e-01 6.13359332e-01 -1.33879349e-01 3.23971212e-01 4.08583075e-01 -5.06450415e-01 -1.56377658e-01 3.92183736e-02 -4.12769586e-01 2.47643396e-01 3.33168775e-01 1.09079349e+00 -3.26073796e-01 -4.29508597e-01 7.92842805e-01 -1.26976216e+00 1.27672291e+00 -2.60691017e-01 -3.58436584e-01 2.00667456e-01 -5.65268338e-01 3.46907303e-02 8.15740287e-01 6.73192670e-04 -1.42569089e+00 -1.86393633e-01 -4.07049596e-01 7.88857862e-02 -4.69696283e-01 8.24270248e-01 -1.28661349e-01 7.52700865e-02 3.79697651e-01 2.51049370e-01 8.02817419e-02 -4.85491037e-01 2.03393567e-02 9.92077470e-01 6.63204640e-02 -4.82972294e-01 -1.05117977e-01 1.03867523e-01 1.67294722e-02 -9.58514035e-01 -7.75289714e-01 -2.72664607e-01 -8.44051659e-01 -5.15496612e-01 1.20858514e+00 -3.93450648e-01 -1.35780978e+00 7.74472594e-01 -1.63354909e+00 -6.08600318e-01 3.17983329e-01 1.90355375e-01 -5.10027409e-01 3.77655774e-01 -1.23566234e+00 -1.10859501e+00 -7.01292515e-01 -1.07231128e+00 3.83621961e-01 2.41477624e-01 -3.51272970e-01 -1.43541586e+00 -2.23147646e-02 5.66068470e-01 4.37276691e-01 3.58836979e-01 1.20817006e+00 -1.02519953e+00 -1.37565345e-01 -6.28389195e-02 -5.28270565e-02 4.33040589e-01 2.72897661e-01 -9.41559896e-02 -1.20208621e+00 -1.45343035e-01 -3.99125926e-02 -4.58320200e-01 6.66892588e-01 3.79011244e-01 1.12872481e+00 -5.99511981e-01 -6.28287718e-02 -1.18187264e-01 6.40208542e-01 5.54990053e-01 5.58928430e-01 1.83617383e-01 8.27943563e-01 9.77345347e-01 2.15017334e-01 3.96397889e-01 7.80088663e-01 5.66692770e-01 2.37587690e-01 -3.37263308e-02 1.57484323e-01 -1.47475721e-02 3.07424903e-01 1.23351169e+00 -3.20837408e-01 -3.84835601e-01 -1.22254741e+00 3.25170994e-01 -2.21899843e+00 -8.79310727e-01 -4.64942724e-01 1.71020305e+00 6.82993770e-01 2.70581394e-01 4.47868764e-01 3.34088922e-01 6.76507831e-01 5.28722107e-01 -1.95134580e-01 -1.10877895e+00 -2.70116758e-02 2.21693918e-01 -2.90937219e-02 6.10544682e-01 -1.02538788e+00 1.00282323e+00 5.90106583e+00 4.96391773e-01 -9.21674848e-01 1.43847138e-01 8.12738419e-01 3.82390171e-01 2.66428471e-01 -3.35180759e-01 -4.78037417e-01 1.73701420e-01 1.42564118e+00 -2.20392972e-01 3.15463990e-01 6.28949463e-01 4.22438979e-01 -2.32696533e-02 -1.10215604e+00 7.58141458e-01 -3.19276452e-02 -1.50301909e+00 7.29296803e-02 1.57913208e-01 5.18229723e-01 -3.53951991e-01 -3.10713828e-01 6.90351188e-01 3.94379765e-01 -1.23504913e+00 7.14136958e-02 5.03023744e-01 2.23044112e-01 -9.65261877e-01 1.27132154e+00 6.21928334e-01 -1.12550092e+00 -2.52699763e-01 -1.10849358e-01 -6.64328218e-01 -3.39249484e-02 2.24078730e-01 -1.21263385e+00 5.68284452e-01 3.76677126e-01 7.10202873e-01 -2.45403096e-01 4.66675550e-01 -1.78176373e-01 9.78924394e-01 2.80998051e-01 -5.09686470e-01 7.97514021e-01 -2.95500964e-01 4.32160765e-01 1.41893756e+00 -3.19499701e-01 1.57312557e-01 2.69329906e-01 2.92851537e-01 -3.55851740e-01 1.28329126e-02 -6.83683872e-01 -2.29366496e-01 1.70678616e-01 1.25976574e+00 -4.28511471e-01 -5.84876120e-01 -4.78760779e-01 7.77347982e-01 5.11308849e-01 1.48984477e-01 -5.27806163e-01 -3.73163968e-01 5.25790632e-01 -2.64654607e-01 -2.17601031e-01 -4.24044698e-01 -3.39161247e-01 -8.57796967e-01 -1.27875835e-01 -1.06138027e+00 4.46134627e-01 -2.91819900e-01 -1.17244554e+00 9.45272624e-01 -2.83861220e-01 -8.31317902e-01 -8.31069529e-01 -5.92700243e-01 -8.16182613e-01 7.23221004e-01 -1.14229083e+00 -1.34425402e+00 -4.82638210e-01 6.01778924e-01 1.02710605e+00 -4.19442862e-01 1.27640474e+00 8.66716951e-02 -6.94539785e-01 3.66108060e-01 -7.49977008e-02 6.28347099e-01 3.25206578e-01 -1.47936690e+00 4.48399335e-01 4.08783168e-01 -1.50637880e-01 6.42443419e-01 3.93117458e-01 -2.11778298e-01 -1.03834426e+00 -9.67937350e-01 1.33960319e+00 -3.21119167e-02 6.41760290e-01 -5.81904352e-01 -9.72165585e-01 9.30417717e-01 8.52776289e-01 -4.16465849e-01 7.42847741e-01 3.94045502e-01 -9.93621349e-02 -6.22479506e-02 -1.06053007e+00 4.41126525e-01 8.31284583e-01 -6.89130425e-01 -4.93468493e-01 3.93542290e-01 7.70759523e-01 -5.51603198e-01 -9.94328856e-01 2.11119652e-03 6.55899525e-01 -1.13799691e+00 4.12797660e-01 -1.09390879e+00 3.95661116e-01 3.89439493e-01 1.47436768e-01 -1.37158895e+00 -3.10668588e-01 -9.26676035e-01 -4.50963616e-01 1.03275478e+00 3.78451109e-01 -6.66644931e-01 6.23875499e-01 5.91309547e-01 -5.05516708e-01 -9.30743158e-01 -9.17647302e-01 -4.95584726e-01 8.14069137e-02 -1.91491142e-01 5.62113285e-01 1.08428657e+00 2.67308563e-01 9.53974068e-01 -6.11099899e-01 -3.55237514e-01 1.06277848e-02 1.10408887e-01 8.31806421e-01 -1.39722228e+00 -4.66908924e-02 -5.51415384e-01 -2.12624520e-01 -1.35156333e+00 5.01307428e-01 -5.26659548e-01 -1.14793397e-01 -1.82274282e+00 -2.14177758e-01 2.62109227e-02 -1.01298340e-01 4.96735245e-01 1.50840312e-01 -5.18474340e-01 1.08976588e-01 -1.01214834e-01 -5.43943226e-01 6.71257913e-01 1.12083995e+00 -3.22902560e-01 -4.25075382e-01 4.01050419e-01 -5.28665558e-02 7.60478318e-01 1.15583980e+00 -8.58353898e-02 -5.59849143e-02 -2.43950695e-01 -2.12046877e-01 5.22968709e-01 2.33506020e-02 -9.01419461e-01 3.74236315e-01 3.78222838e-02 4.03425703e-03 -5.49478114e-01 4.82161552e-01 -3.38639319e-01 -4.84858721e-01 6.53728724e-01 -8.12314510e-01 2.30569661e-01 1.07147209e-02 4.61424232e-01 -4.61008400e-01 -2.00855747e-01 4.40945625e-01 -3.78904492e-01 -3.58341992e-01 4.48974706e-02 -1.22746110e+00 -3.32321137e-01 6.45910919e-01 2.15079099e-01 -4.16584164e-01 -9.34006929e-01 -7.36396074e-01 2.42515668e-01 -3.40998501e-01 3.77762645e-01 3.88789266e-01 -1.04267609e+00 -5.26112139e-01 -1.77452996e-01 -1.03509910e-01 -1.54241621e-01 2.36811206e-01 9.34637904e-01 -6.08617008e-01 9.33322370e-01 -3.05679917e-01 -4.74411488e-01 -1.55199623e+00 8.61596018e-02 4.90561336e-01 -8.28381121e-01 -5.42321920e-01 6.39122248e-01 -1.20716199e-01 -7.07329750e-01 3.69251907e-01 -6.14296317e-01 -6.41422510e-01 2.20645014e-02 3.74501646e-01 5.66600204e-01 1.33265823e-01 -5.80257416e-01 -1.25932306e-01 -2.03177482e-01 -1.58144176e-01 -1.75992716e-02 1.18725121e+00 6.02977611e-02 -4.58858430e-01 1.01787019e+00 1.03280926e+00 -3.61620575e-01 -6.34224892e-01 -4.17433858e-01 5.82309425e-01 1.32037504e-02 -7.47991949e-02 -6.42845511e-01 -7.81563103e-01 1.21789420e+00 2.19643354e-01 1.00404143e+00 6.81249738e-01 -4.11136508e-01 1.13341200e+00 9.89056468e-01 2.75756985e-01 -1.10759676e+00 2.54319221e-01 1.34567952e+00 9.32303309e-01 -1.10623324e+00 -4.53996360e-01 -8.33443403e-02 -7.50167787e-01 1.61349332e+00 8.34154725e-01 3.65632735e-02 4.27159697e-01 -2.54260842e-02 -1.72277838e-02 -3.36015254e-01 -1.09270787e+00 -1.67271674e-01 -7.70135671e-02 3.49384189e-01 9.70954239e-01 2.18885407e-01 -4.41864699e-01 6.66255236e-01 -1.34745866e-01 -7.63540715e-02 6.76891208e-01 8.34965169e-01 -2.93605179e-01 -1.12703502e+00 4.02030833e-02 5.97596705e-01 -4.09516484e-01 1.27943292e-01 -9.52710211e-01 7.45761514e-01 -2.32489869e-01 1.35758662e+00 2.46842131e-01 -7.57435858e-01 2.13517502e-01 5.13024628e-01 2.18992099e-01 -5.51896214e-01 -1.31091034e+00 -2.83999681e-01 8.56135428e-01 -3.82695407e-01 -7.04013050e-01 -3.47602218e-01 -1.31463730e+00 -6.29696906e-01 -2.19044715e-01 2.19774485e-01 4.38657701e-01 1.28416884e+00 2.59563833e-01 8.49305212e-01 7.44120300e-01 -6.16785407e-01 -2.07180381e-01 -1.66632199e+00 -3.51015896e-01 1.19484460e-03 2.21180275e-01 -4.46944356e-01 6.36819750e-02 -2.84871280e-01]	[12.825575828552246, 7.767239570617676]
17a5a936-1afe-4ba3-ab81-e7274bdcf143	can-knowledge-graph-embeddings-tell-us-what	null	null	https://aclanthology.org/2020.insights-1.11	https://aclanthology.org/2020.insights-1.11.pdf	Can Knowledge Graph Embeddings Tell Us What Fact-checked Claims Are About?	The web offers a wealth of discourse data that help researchers from various fields analyze debates about current societal issues and gauge the effects on society of important phenomena such as misinformation spread. Such analyses often revolve around claims made by people about a given topic of interest. Fact-checking portals offer partially structured information that can assist such analysis. However, exploiting the network structure of such online discourse data is as of yet under-explored. We study the effectiveness of using neural-graph embedding features for claim topic prediction and their complementarity with text embeddings. We show that graph embeddings are modestly complementary with text embeddings, but the low performance of graph embedding features alone indicate that the model fails to capture topological features pertinent of the topic prediction task.	['Andon Tchechmedjiev', 'Konstantin Todorov', 'Luke Lo Seen', 'Katarina Boland', 'Sébastien Harispe', 'Valentina Beretta']	null	null	null	null	emnlp-insights-2020-11	['knowledge-graph-embeddings', 'knowledge-graph-embeddings']	['graphs', 'methodology']	[-1.10076979e-01 7.26893127e-01 -6.47665858e-01 9.74041373e-02 -4.35066134e-01 -7.48602450e-01 1.22434664e+00 1.36564827e+00 -1.67421088e-01 3.72538179e-01 1.26478386e+00 -9.53023672e-01 -2.26553738e-01 -1.15436745e+00 -4.96811748e-01 -2.15820774e-01 -3.29238653e-01 4.78570879e-01 4.37022954e-01 -5.28120518e-01 6.56261742e-01 1.41146228e-01 -9.52293992e-01 1.39212564e-01 8.81920159e-01 6.66431785e-01 -2.55570501e-01 3.83838594e-01 -5.75076342e-01 8.69963467e-01 -4.17818040e-01 -7.19189525e-01 -3.11125100e-01 -1.96867347e-01 -9.41626251e-01 -1.04912169e-01 4.03631836e-01 5.65646738e-02 -8.41546774e-01 1.11356783e+00 1.20649800e-01 -1.37626395e-01 8.22763622e-01 -9.21908796e-01 -8.93384278e-01 6.37114525e-01 -4.50380325e-01 7.60886192e-01 3.76116663e-01 -1.89731166e-01 1.56079900e+00 -4.12171453e-01 1.27691388e+00 1.33870292e+00 9.61030722e-01 -3.57694060e-01 -1.32064092e+00 1.45354301e-01 2.24728882e-02 2.14878187e-01 -7.49408185e-01 -2.57987142e-01 1.20729256e+00 -8.01647007e-01 5.98226547e-01 3.42032522e-01 8.43634009e-01 1.06414866e+00 3.91791642e-01 3.07299823e-01 9.86630440e-01 -1.96383566e-01 -6.87443558e-03 9.26892161e-02 4.81518775e-01 8.89966011e-01 6.93767309e-01 -4.32227850e-01 -4.98592466e-01 -9.01408613e-01 2.15281248e-01 -1.58568829e-01 -1.85318723e-01 -2.83310294e-01 -1.29065335e+00 1.48270488e+00 7.11968422e-01 6.71169162e-01 -4.57753360e-01 1.47927210e-01 6.92784727e-01 4.04393792e-01 1.19620287e+00 5.84033251e-01 -2.16060147e-01 -1.79942846e-01 -5.97718954e-01 4.21846092e-01 1.00960720e+00 2.33291671e-01 6.48772418e-01 -3.94007802e-01 5.86619191e-02 6.60126209e-01 2.95858592e-01 1.13825180e-01 1.16995066e-01 -6.72353387e-01 6.94073677e-01 1.29379928e+00 1.00985244e-01 -2.09286666e+00 -5.62327981e-01 2.33711582e-03 -3.32375556e-01 -1.48624763e-01 7.46944308e-01 -1.60485879e-01 -1.45855471e-01 1.22382557e+00 5.88145494e-01 -2.95058101e-01 -4.16550934e-01 8.50332737e-01 8.26545537e-01 6.17698431e-01 -1.11279488e-01 5.73748583e-03 1.62319195e+00 -3.79966468e-01 -8.29979539e-01 -6.60258085e-02 8.64492238e-01 -7.33392656e-01 7.82259643e-01 -2.32850313e-01 -7.89450049e-01 2.03390539e-01 -8.38401079e-01 -4.51839745e-01 -8.51683617e-01 -6.19508684e-01 8.64981472e-01 6.26266062e-01 -8.17031264e-01 5.80344558e-01 -5.36697924e-01 -5.91162801e-01 8.64131749e-01 -3.69574636e-01 -1.83519155e-01 -3.42986211e-02 -1.29588795e+00 1.11756635e+00 4.05923501e-02 -4.11854863e-01 3.78281884e-02 -5.91415286e-01 -8.17178130e-01 -8.01783130e-02 5.99806786e-01 -2.69571811e-01 4.81456965e-01 -4.57131565e-01 -7.34274924e-01 8.72082710e-01 1.30194183e-02 -6.18226647e-01 4.42718089e-01 1.52735904e-01 -4.73447531e-01 4.46367621e-01 2.58947521e-01 1.13958359e-01 6.01348102e-01 -8.39723229e-01 -1.37867406e-01 -5.49844980e-01 3.81208122e-01 7.82923475e-02 -6.90819383e-01 1.47175834e-01 1.95185050e-01 -7.67162144e-01 2.54563868e-01 -7.90113270e-01 -6.63307980e-02 3.51069689e-01 -6.17868483e-01 -4.67735797e-01 1.15913522e+00 -1.16464126e+00 1.23153281e+00 -1.80834198e+00 3.01455171e-03 1.79242283e-01 8.04471612e-01 -9.41788852e-02 1.25666797e-01 1.24604714e+00 4.44322228e-01 8.37975264e-01 4.14549112e-02 4.40238029e-01 8.05619285e-02 -1.49236366e-01 -3.51876557e-01 8.62923801e-01 1.02155544e-01 1.19191349e+00 -9.49306071e-01 -5.85692644e-01 -8.46295506e-02 5.22793174e-01 -5.21428108e-01 -5.15685618e-01 -5.23762345e-01 6.07807823e-02 -9.45714474e-01 4.11969095e-01 1.21124446e-01 -7.42746174e-01 7.29377568e-01 -3.44403803e-01 -1.32377788e-01 9.50200856e-01 -5.12992263e-01 1.16298997e+00 5.42864427e-02 1.45788622e+00 1.38257757e-01 -1.01673388e+00 6.47054791e-01 5.73622100e-02 4.74794686e-01 -3.88507158e-01 2.75973678e-01 -3.13101530e-01 3.41033936e-01 -5.61267078e-01 8.18010271e-01 8.42191428e-02 1.48060415e-02 1.06358910e+00 -3.28285873e-01 2.58476406e-01 -1.03111334e-01 5.83337128e-01 1.34907818e+00 -4.36262608e-01 3.46501887e-01 -6.77755296e-01 -9.54685956e-02 6.02974415e-01 1.22559868e-01 3.85892153e-01 -1.92780450e-01 1.18916124e-01 1.18232477e+00 -5.47357500e-01 -1.36669099e+00 -7.17861295e-01 -3.62967730e-01 1.06022608e+00 2.82186568e-01 -7.31278121e-01 -2.88705558e-01 -8.20674300e-01 4.64400411e-01 6.00136817e-01 -8.64466667e-01 2.24016055e-01 -3.74089897e-01 -8.14371884e-01 3.17597389e-01 9.74338129e-02 9.86454487e-02 -6.28668189e-01 -5.08966625e-01 2.04981774e-01 -3.20301324e-01 -9.28046167e-01 -3.14382702e-01 -3.75045061e-01 -9.70439613e-01 -1.70747638e+00 -3.17029148e-01 -5.36131620e-01 4.72378284e-01 3.92225891e-01 1.01837230e+00 4.57322001e-01 -2.31625408e-01 3.90054196e-01 -4.16794389e-01 -1.57808945e-01 -6.47369504e-01 1.71763107e-01 -1.06995173e-01 -1.12867229e-01 2.84026027e-01 -6.49732471e-01 -3.82891297e-01 -3.43009531e-02 -7.70791531e-01 -1.26167342e-01 4.02497388e-02 7.08781064e-01 -1.68011263e-01 -9.43882391e-02 5.66316664e-01 -1.17252791e+00 1.31874037e+00 -1.01244986e+00 -2.41485551e-01 1.56648323e-01 -6.51574910e-01 5.28218783e-02 -8.08071867e-02 -4.56620902e-01 -5.47682762e-01 -1.11331284e+00 2.91998416e-01 2.08840996e-01 2.17070654e-01 1.01309168e+00 4.32147890e-01 3.67855914e-02 7.34766543e-01 -4.45522457e-01 3.75332057e-01 -4.16285962e-01 5.87720871e-01 5.21620691e-01 -2.68163055e-01 -3.77579391e-01 7.31395245e-01 7.04621613e-01 -1.77777335e-01 -1.22179532e+00 -6.93700850e-01 -2.52934784e-01 -2.92487830e-01 -1.78958431e-01 9.09304857e-01 -6.32367015e-01 -7.85220861e-01 -3.15544903e-01 -1.24721789e+00 1.06933892e-01 -1.06960118e-01 2.06219807e-01 -1.72596380e-01 3.87792528e-01 -8.63601387e-01 -6.63809180e-01 6.73833638e-02 -6.32250786e-01 5.01895249e-01 -2.86736906e-01 -5.02392530e-01 -1.59807301e+00 1.71819299e-01 6.59539282e-01 6.77285910e-01 7.80208945e-01 1.24284923e+00 -1.07871413e+00 -4.90381598e-01 -2.37122104e-01 -5.20380497e-01 -5.79629242e-01 2.21060321e-01 -1.84898272e-01 -7.46994555e-01 -1.21954195e-02 -2.87373096e-01 -2.13210672e-01 8.68398488e-01 2.21964598e-01 4.99838144e-01 -1.01113343e+00 -5.62286198e-01 -1.72621772e-01 1.26258683e+00 -5.44616997e-01 3.26384783e-01 5.18554330e-01 7.66186357e-01 9.89126265e-01 -1.43215358e-01 2.59262562e-01 7.41057754e-01 6.34318352e-01 3.68091077e-01 2.38663957e-01 -5.48544452e-02 -5.26849866e-01 -4.28675162e-03 7.72039473e-01 -3.85275632e-02 -2.15409860e-01 -1.34338772e+00 8.19656730e-01 -1.71660495e+00 -1.15888095e+00 -4.66996104e-01 1.51174760e+00 6.32590532e-01 4.84431922e-01 3.29297870e-01 9.36522782e-02 9.15641487e-01 1.05998027e+00 -1.69259518e-01 -4.51676846e-01 -2.20028803e-01 -4.64138657e-01 7.26557970e-01 7.03875661e-01 -8.48118424e-01 6.52086377e-01 6.24080801e+00 4.97469038e-01 -7.60266066e-01 4.05521095e-01 6.06554806e-01 4.01755631e-01 -8.87124360e-01 2.72428334e-01 -1.74901262e-01 5.41898549e-01 9.46548879e-01 -5.40709198e-01 2.42623955e-01 4.37782079e-01 1.70359716e-01 -1.88941911e-01 -6.24347806e-01 2.49833211e-01 1.16250440e-01 -2.25139999e+00 -1.07115008e-01 6.62365139e-01 7.59318590e-01 2.69622535e-01 -8.80336575e-03 -2.80182362e-01 5.13154387e-01 -7.85946906e-01 4.42873299e-01 4.15296108e-01 4.40902799e-01 -4.27709788e-01 4.80816096e-01 2.32237607e-01 -6.63973451e-01 -1.00304127e-01 -2.04590231e-01 -4.40502435e-01 3.88160050e-01 9.41713929e-01 -1.04548991e+00 2.12567374e-01 2.66276062e-01 6.51558816e-01 -7.01369107e-01 6.01835966e-01 -1.96975350e-01 9.37607229e-01 -1.94217622e-01 -3.44459981e-01 3.81816298e-01 -3.28655899e-01 1.01078224e+00 9.65761423e-01 4.92095295e-03 -7.37528726e-02 8.45346302e-02 9.40905392e-01 -3.78019631e-01 2.15428218e-01 -1.16603160e+00 -8.91238213e-01 5.01581550e-01 1.01759052e+00 -1.04116380e+00 -6.69203233e-03 -5.71229100e-01 2.70912647e-01 6.12370133e-01 1.10136136e-01 -2.70405680e-01 1.19379244e-03 6.94775939e-01 8.74021709e-01 2.26066560e-01 -6.28348589e-01 -3.81121397e-01 -1.03120804e+00 -1.64764911e-01 -5.64818025e-01 2.28693575e-01 -4.71546680e-01 -1.48793828e+00 1.70448273e-01 -2.20507860e-01 -6.99148357e-01 -2.15901551e-03 -3.91162306e-01 -1.04418552e+00 4.74746913e-01 -1.28871095e+00 -1.30384398e+00 2.27549419e-01 -2.33530886e-02 1.57918125e-01 6.32522851e-02 5.19723952e-01 -2.53290683e-01 -1.39651254e-01 -1.25238221e-04 3.53342056e-01 3.48297417e-01 3.22392046e-01 -1.07961082e+00 5.68602741e-01 3.03305030e-01 5.86556613e-01 4.99704570e-01 9.69675124e-01 -1.13059378e+00 -1.45921409e+00 -7.30860949e-01 1.37463677e+00 -8.01825762e-01 1.53227639e+00 -3.63719523e-01 -1.03202415e+00 6.11265063e-01 4.43869919e-01 -7.79483616e-02 7.60580420e-01 8.12302232e-01 -4.67313498e-01 5.60680807e-01 -8.54384780e-01 7.24152327e-01 1.04430401e+00 -9.81166720e-01 -9.59115505e-01 6.40571058e-01 8.76735389e-01 3.72808874e-02 -1.12834311e+00 -2.36428291e-01 2.75722265e-01 -6.40348852e-01 8.80825341e-01 -9.85165954e-01 8.61952484e-01 3.73844773e-01 -1.90456718e-01 -1.11581314e+00 -3.32184911e-01 -5.08115172e-01 -2.22648829e-01 1.18490720e+00 5.50639331e-01 -8.39196384e-01 7.01058388e-01 4.24667299e-01 2.91439176e-01 -6.24913275e-01 -8.48820269e-01 -3.76646608e-01 2.93641388e-01 -1.56961486e-01 3.56899977e-01 1.73370779e+00 4.88621622e-01 5.10022223e-01 1.10219128e-01 -1.54983308e-02 5.54817915e-01 3.07534218e-01 5.82259297e-01 -1.81073737e+00 3.02493781e-01 -7.16125309e-01 -6.76375628e-01 -4.27451700e-01 1.59495279e-01 -1.05456281e+00 -8.64836693e-01 -1.90326428e+00 1.13002673e-01 -5.50027013e-01 1.74163043e-01 3.13791409e-02 1.57060809e-02 8.64383876e-02 1.92234948e-01 4.18713421e-01 -4.41503614e-01 3.91584486e-01 1.02705967e+00 -3.36928040e-01 -4.31420468e-02 -5.85600257e-01 -1.01448286e+00 8.10673594e-01 7.03372836e-01 -4.04062927e-01 -1.33677304e-01 -2.58944392e-01 8.92408013e-01 1.79884776e-01 5.43028653e-01 -4.68018234e-01 2.03734607e-01 -1.83432236e-01 1.03770062e-01 -3.51063639e-01 1.53274566e-01 -5.21826744e-01 -2.19179645e-01 2.72718370e-01 -6.49490654e-01 2.33597890e-01 6.02767356e-02 1.16659904e+00 -1.17178991e-01 6.04847632e-02 2.90882047e-02 1.55324847e-01 -2.07976624e-01 6.12644628e-02 -3.87596488e-01 5.31045198e-01 6.92870617e-01 -7.75749758e-02 -9.57356751e-01 -6.56453490e-01 -6.00395441e-01 2.25804243e-02 7.51969755e-01 5.78372061e-01 2.99295217e-01 -1.35894632e+00 -8.26855421e-01 -4.74366367e-01 -4.41939570e-02 -6.45623744e-01 -1.90667957e-01 1.04019201e+00 -3.20412606e-01 3.22160840e-01 9.00012534e-03 -1.59031555e-01 -1.00457597e+00 6.01395249e-01 -2.78719276e-01 -3.09551001e-01 -1.13204265e+00 2.60480106e-01 -1.45234481e-01 -3.56193036e-01 -4.30888772e-01 -7.11387768e-02 -4.08764809e-01 7.05756843e-01 3.60726833e-01 5.84692240e-01 -3.54766011e-01 -8.71752143e-01 -2.34806389e-01 2.27485895e-02 1.47465080e-01 -1.38127282e-01 1.67937350e+00 -2.59501696e-01 -3.22951525e-01 7.56068587e-01 1.21672225e+00 3.39127332e-01 -8.07892442e-01 -2.89681554e-01 5.45310199e-01 -5.66569269e-01 2.19258353e-01 -4.72211003e-01 -6.39795661e-01 7.28789508e-01 -2.02363551e-01 1.24022138e+00 7.36688590e-03 4.83111501e-01 7.28320122e-01 3.15377712e-01 7.55379423e-02 -1.10959363e+00 5.34339882e-02 2.65949398e-01 9.40098345e-01 -1.21303082e+00 4.55580622e-01 -4.77625161e-01 -5.19900024e-01 1.11785161e+00 -6.08683005e-02 -3.00498903e-01 1.03657699e+00 -1.87897637e-01 -3.82705182e-01 -9.12308574e-01 -6.32340193e-01 -7.44289830e-02 4.78269935e-01 4.02361006e-01 4.27812934e-01 9.24819633e-02 -7.98419774e-01 2.50439733e-01 -1.14581287e-01 -5.94415069e-01 6.42159939e-01 7.36258745e-01 -7.60685384e-01 -6.57628834e-01 -3.98835868e-01 9.93966401e-01 -7.29832709e-01 -1.11811511e-01 -7.20757723e-01 8.76483202e-01 -3.05719733e-01 9.02184665e-01 2.46530533e-01 -2.15624779e-01 -3.43973309e-01 1.45214781e-01 3.72619741e-02 -3.95064741e-01 -4.83074486e-01 -3.11428308e-01 9.52756107e-01 -4.21037525e-01 -4.60072547e-01 -9.18813050e-01 -6.89343393e-01 -8.32599640e-01 -4.29892778e-01 2.32742205e-01 7.24319279e-01 9.60589409e-01 4.97243673e-01 2.73217231e-01 2.77422011e-01 -5.13245165e-01 -2.44974375e-01 -9.55128014e-01 -4.39156353e-01 4.83406276e-01 3.91480207e-01 -7.89517760e-01 -3.23019415e-01 -1.64681107e-01]	[8.706696510314941, 9.934659957885742]
25f66e4f-f98d-45e8-b914-37ccd9d528c6	leverage-financial-news-to-predict-stock	1506.07220	null	http://arxiv.org/abs/1506.07220v1	http://arxiv.org/pdf/1506.07220v1.pdf	Leverage Financial News to Predict Stock Price Movements Using Word Embeddings and Deep Neural Networks	Financial news contains useful information on public companies and the market. In this paper we apply the popular word embedding methods and deep neural networks to leverage financial news to predict stock price movements in the market. Experimental results have shown that our proposed methods are simple but very effective, which can significantly improve the stock prediction accuracy on a standard financial database over the baseline system using only the historical price information.	['Hui Jiang', 'Yangtuo Peng']	2015-06-24	leverage-financial-news-to-predict-stock-1	https://aclanthology.org/N16-1041	https://aclanthology.org/N16-1041.pdf	naacl-2016-6	['stock-prediction']	['time-series']	[-1.16539991e+00 -4.48227376e-01 -6.93705618e-01 -6.48424104e-02 -4.10363019e-01 -4.81809020e-01 8.20441127e-01 -2.16184542e-01 -4.53446507e-01 9.25735831e-01 8.69548738e-01 -3.96124691e-01 3.81362975e-01 -1.50836253e+00 -3.03978145e-01 -2.54685938e-01 6.07064692e-03 7.82482326e-02 4.53871757e-01 -6.33173764e-01 6.83779776e-01 1.99599236e-01 -1.06992185e+00 1.02558814e-01 6.82088407e-03 1.55924046e+00 -3.15334290e-01 3.00765604e-01 -9.09016669e-01 1.57890546e+00 -6.81055069e-01 -1.02157724e+00 7.30477452e-01 -1.02010351e-02 -2.47297674e-01 -2.98985839e-01 -1.46688744e-01 -7.16780066e-01 -6.40192151e-01 1.24884677e+00 3.70262265e-01 -1.38010755e-01 3.13190162e-01 -6.73388839e-01 -1.44221723e+00 1.28476512e+00 -7.10830808e-01 1.12913048e+00 -4.94887196e-02 -1.59540400e-01 1.70587313e+00 -1.23317134e+00 5.95525146e-01 7.84026444e-01 8.44143748e-01 1.10820187e-02 -3.89061838e-01 -1.03236425e+00 1.47032633e-01 2.43036687e-01 -5.22401690e-01 -9.40332841e-03 1.08825719e+00 -5.63353539e-01 1.12664413e+00 -2.73048729e-02 1.24245465e+00 8.70804965e-01 7.77903795e-01 7.31530309e-01 8.53337228e-01 -3.56988497e-02 2.37175822e-02 2.01103523e-01 6.83760703e-01 1.27161592e-01 8.32535744e-01 3.15214664e-01 -6.57606661e-01 -4.51759815e-01 6.73736572e-01 5.74577987e-01 -6.37195557e-02 4.70176220e-01 -1.07521331e+00 1.68251443e+00 3.63032222e-01 6.32418454e-01 -8.62111330e-01 2.20316872e-01 6.41856790e-01 7.60946810e-01 1.34642863e+00 7.00972438e-01 -6.49706423e-01 -4.72366810e-01 -8.13418031e-01 6.26898885e-01 1.06273854e+00 3.66832465e-01 -4.39992454e-03 8.50196362e-01 2.55921096e-01 3.95899713e-01 3.88304323e-01 4.55914170e-01 1.36270082e+00 -1.27176598e-01 5.77879250e-01 5.18435061e-01 3.83683324e-01 -1.48592949e+00 -2.45922282e-01 -6.95968151e-01 -3.65550935e-01 2.58681387e-01 -2.05321878e-01 -5.32421649e-01 -3.91402692e-01 6.72043622e-01 -1.53936356e-01 6.44048572e-01 4.56630468e-01 4.96808499e-01 8.36016715e-01 1.22817802e+00 -2.89905697e-01 -2.08067641e-01 1.16464365e+00 -1.20363951e+00 -1.32479167e+00 -2.46527761e-01 1.92388371e-01 -8.92416835e-01 2.94542432e-01 1.01063885e-01 -1.12161732e+00 -3.41326714e-01 -1.37954938e+00 3.86565238e-01 -8.48765314e-01 -5.69551229e-01 7.62403607e-01 4.13015008e-01 -7.29016662e-01 8.32429111e-01 -6.68037713e-01 6.94346488e-01 5.09366453e-01 -1.86197132e-01 4.12987433e-02 5.05106509e-01 -1.80921721e+00 1.00710630e+00 7.53334999e-01 -1.50789350e-01 5.32859713e-02 -8.36091638e-01 -8.89070272e-01 8.60915631e-02 5.83222061e-02 -2.45112896e-01 1.56593192e+00 -3.87878865e-01 -1.56064677e+00 2.59868473e-01 5.30557871e-01 -1.23313057e+00 4.00531083e-01 -5.47507226e-01 -9.50778425e-01 -1.47060826e-01 -1.86325312e-01 -2.63138890e-01 4.43736494e-01 -3.87015007e-02 -9.74183023e-01 1.46063432e-01 -2.14507744e-01 -3.38354379e-01 -5.59198618e-01 4.25503314e-01 7.43885860e-02 -1.46054900e+00 -1.49280697e-01 -3.73420596e-01 -9.74291340e-02 -4.52144921e-01 4.55023460e-02 -4.07203525e-01 6.85566127e-01 -9.86702263e-01 1.34339345e+00 -1.80071557e+00 -5.34267664e-01 5.27242720e-02 7.97711834e-02 1.71641260e-01 1.56968310e-01 6.94534779e-01 -2.35374570e-01 2.67002642e-01 3.99080575e-01 3.03092778e-01 2.03669265e-01 -7.64011443e-02 -1.21275055e+00 4.46403027e-01 1.08146980e-01 1.50403738e+00 -5.31950653e-01 2.19705999e-01 -1.46347024e-02 2.59694099e-01 -1.62625760e-01 4.90980707e-02 -1.90666348e-01 -6.32981002e-01 -5.71084976e-01 6.58301532e-01 4.88545299e-01 -1.49778202e-01 -3.91352028e-01 9.86370146e-02 -3.15441221e-01 7.72668481e-01 -9.51308608e-01 7.15552986e-01 -6.20935299e-02 6.97544515e-01 -8.32315266e-01 -7.21348345e-01 1.34225368e+00 3.11303139e-01 3.04503828e-01 -8.34720254e-01 2.64298886e-01 4.92216080e-01 -1.53649703e-01 -1.08805269e-01 6.14129007e-01 -6.92545056e-01 -7.80984536e-02 8.86027873e-01 -2.40415215e-01 1.28323466e-01 5.97527288e-02 -2.46026531e-01 7.95852363e-01 -3.57931584e-01 6.14680409e-01 -1.63332105e-01 6.55667856e-02 -1.65237457e-01 6.46097958e-01 3.00164491e-01 -1.31736532e-01 2.06909567e-01 5.63771784e-01 -1.16962457e+00 -1.10948563e+00 -5.83872437e-01 -2.15323672e-01 7.63141572e-01 -3.49940300e-01 -1.17384098e-01 -3.42912303e-04 -5.70050895e-01 6.18531823e-01 9.68500137e-01 -8.44656825e-01 1.11329198e-01 -4.32918370e-01 -1.20481849e+00 1.59821287e-01 9.71106112e-01 6.40366852e-01 -1.13604045e+00 -5.31585991e-01 7.53966868e-01 5.49309850e-01 -7.42107749e-01 -3.76360506e-01 -1.20232441e-01 -7.55558193e-01 -9.74225104e-01 -1.21410108e+00 -3.86030406e-01 -1.18620351e-01 9.81531013e-03 1.04982424e+00 -4.53643888e-01 1.27845719e-01 -2.41617620e-01 -4.49116468e-01 -9.09981728e-01 -8.19969624e-02 -2.79316127e-01 -3.91515717e-02 4.57336269e-02 6.81451201e-01 -3.25408161e-01 -4.41994011e-01 -1.89063564e-01 -8.27527761e-01 -4.10503328e-01 2.57957995e-01 8.43707979e-01 2.07822844e-01 3.27849001e-01 1.02334249e+00 -9.85396087e-01 1.15979838e+00 -9.28076446e-01 -1.08830547e+00 -4.95486446e-02 -1.17445958e+00 1.30292669e-01 1.12895921e-01 -2.56363630e-01 -8.62790465e-01 -7.31105447e-01 -2.59665608e-01 -1.14545703e-01 8.65512192e-01 1.39637423e+00 7.32168138e-01 2.67350346e-01 3.42177637e-02 4.79839236e-01 -1.22131892e-01 -7.56320715e-01 1.49660587e-01 3.71439517e-01 2.52562672e-01 2.74596423e-01 9.71837699e-01 2.49175787e-01 -5.11082828e-01 -1.37481973e-01 -1.04753268e+00 -3.57574105e-01 -1.97458431e-01 1.21495932e-01 5.54535568e-01 -1.08588696e+00 -2.06868589e-01 7.14252114e-01 -1.00604272e+00 5.12902558e-01 -4.03211743e-01 8.72411609e-01 6.17542267e-02 4.13215533e-02 -1.33917165e+00 -8.48773718e-01 -8.03043246e-01 -4.68185127e-01 2.39336178e-01 1.20389506e-01 -6.70605898e-02 -1.49603057e+00 6.80925488e-01 -1.06441475e-01 9.82086003e-01 4.63110954e-01 4.59649771e-01 -1.50885773e+00 -2.86136538e-01 -1.00892150e+00 -2.50436515e-01 5.66691160e-01 3.16076428e-01 -8.37515965e-02 -5.90129614e-01 8.43234733e-02 4.61915761e-01 1.25742499e-02 1.38531208e+00 1.79430872e-01 3.55454385e-01 -6.37927115e-01 2.19728559e-01 4.73519057e-01 1.42511213e+00 5.94511569e-01 4.28231627e-01 1.13704264e+00 3.55321318e-01 1.10898927e-01 3.04499805e-01 7.82878101e-01 2.65487969e-01 4.46237065e-02 7.77519047e-02 3.08568150e-01 4.27510798e-01 -1.39829829e-01 7.88769424e-01 1.41676807e+00 -1.51191771e-01 1.28497347e-01 -8.62477720e-01 4.96707767e-01 -1.87351704e+00 -1.48906779e+00 2.10890234e-01 1.39071941e+00 9.50991213e-01 7.44329154e-01 2.28637651e-01 1.00409143e-01 5.08562684e-01 9.40349162e-01 -4.18178380e-01 -2.43639648e-01 -4.30758893e-01 3.94845217e-01 9.51281130e-01 2.32483968e-01 -1.26025295e+00 5.53331614e-01 7.60389423e+00 3.90320510e-01 -1.28940165e+00 1.62794530e-01 5.31290829e-01 -2.11818114e-01 -6.19063854e-01 -7.25147486e-01 -1.19565856e+00 7.87913740e-01 1.34591413e+00 -1.15703905e+00 -4.35529113e-01 1.37390220e+00 -2.07695618e-01 6.43296838e-01 -5.33504307e-01 7.72538900e-01 3.35242823e-02 -2.30929708e+00 2.57182837e-01 1.19181745e-01 7.26611078e-01 4.67388988e-01 5.08810103e-01 3.90399486e-01 5.73262215e-01 -5.11629939e-01 8.04857731e-01 8.41814041e-01 -1.48090437e-01 -1.24553537e+00 1.60391152e+00 1.06231291e-02 -9.91943538e-01 -1.43298060e-01 -7.93768167e-01 -4.93492901e-01 2.75143564e-01 8.29126120e-01 -5.62085569e-01 3.49977821e-01 5.39584994e-01 1.39100325e+00 -2.74263561e-01 7.98402607e-01 -1.62401304e-01 6.93588793e-01 2.54967250e-02 -5.75227976e-01 7.69417286e-01 -1.93668649e-01 1.91849872e-01 9.64523315e-01 6.87216461e-01 -3.03491671e-02 -1.80257604e-01 7.93350577e-01 -2.44212314e-01 3.22637290e-01 -8.35922420e-01 -5.99581003e-01 -1.21807344e-02 6.28643632e-01 -5.31127512e-01 -5.83179951e-01 -1.07661605e+00 4.90161359e-01 -2.15666573e-02 -3.26792635e-02 -7.68868625e-01 -6.09825790e-01 7.36753523e-01 1.07982814e-01 9.85699832e-01 -2.18970571e-02 -1.51975006e-01 -1.54305112e+00 9.81244221e-02 -3.49321157e-01 5.51213205e-01 -5.78828454e-01 -1.87375569e+00 6.03672683e-01 -3.63680571e-01 -1.36512506e+00 -4.71356660e-01 -9.45679963e-01 -1.00465345e+00 9.01181579e-01 -2.09959364e+00 -4.19899553e-01 8.50621104e-01 2.76862651e-01 6.22650743e-01 -1.21151829e+00 6.79064691e-01 -2.02947017e-02 -6.67078793e-01 1.69443175e-01 7.41160095e-01 7.88603425e-01 1.08204506e-01 -1.29684567e+00 1.33316910e+00 7.93656349e-01 5.90912402e-01 4.24811363e-01 4.01412219e-01 -9.48325753e-01 -1.09939075e+00 -9.90167916e-01 1.14501858e+00 -2.41323113e-01 1.89856589e+00 1.54500842e-01 -8.47351372e-01 8.01010907e-01 7.51281440e-01 1.16483472e-01 1.09287453e+00 -1.56987146e-01 -7.36107767e-01 -1.51097462e-01 -9.57877159e-01 2.54004419e-01 -1.30490750e-01 -3.40999365e-01 -1.65311670e+00 1.87885627e-01 1.29616714e+00 -4.31606807e-02 -9.43317652e-01 -1.60054695e-02 5.46808839e-01 -7.00238883e-01 9.84970450e-01 -8.60579848e-01 8.15449953e-01 2.40184158e-01 -7.40564540e-02 -1.35256064e+00 -5.22056401e-01 -3.90592694e-01 -5.83356321e-01 1.00716794e+00 7.61327982e-01 -1.32815099e+00 5.35187066e-01 7.94628978e-01 3.26361090e-01 -7.40177512e-01 -1.02839255e+00 -8.83541048e-01 2.87342697e-01 -4.12027031e-01 1.09807575e+00 1.28127825e+00 1.76653922e-01 -1.80971026e-02 -5.13474286e-01 -3.00607413e-01 1.46032035e-01 5.18557310e-01 1.74086362e-01 -1.29989064e+00 -3.53002101e-01 -7.76145041e-01 -7.17643917e-01 -5.06192207e-01 3.49936634e-01 -6.38958931e-01 -8.44922006e-01 -1.27481198e+00 -1.35739431e-01 4.15679693e-01 -1.34583461e+00 1.36145458e-01 1.09396400e-02 1.81972966e-01 1.63339004e-01 3.55404973e-01 -1.25654489e-01 5.68467140e-01 9.65600967e-01 -5.14110684e-01 2.37457082e-01 2.58283585e-01 -9.05407608e-01 8.50495815e-01 9.42482471e-01 -4.98603046e-01 9.81689990e-02 -5.85687980e-02 8.76251817e-01 -1.04436755e-01 -6.38938248e-02 -5.48417449e-01 8.87131467e-02 -1.53730392e-01 5.67341089e-01 -8.87022793e-01 5.56805283e-02 -6.70163572e-01 -1.62967131e-01 9.07292068e-01 -4.03080732e-01 7.65856802e-01 3.21338356e-01 8.04232657e-01 -9.26237881e-01 -3.80852222e-01 2.65558660e-01 -1.94435641e-01 -9.42562997e-01 3.26988399e-01 -2.70005196e-01 -1.61273032e-01 1.04897702e+00 5.48905842e-02 -3.94928664e-01 -4.64677632e-01 -4.76529568e-01 -6.67636469e-02 -3.17821145e-01 7.30728805e-01 8.35408866e-01 -1.85512817e+00 -1.07730830e+00 9.10283625e-02 -5.13787866e-02 -8.02168667e-01 -4.42232639e-01 2.88228720e-01 -9.50628638e-01 7.27859914e-01 -2.12291598e-01 5.93217432e-01 -5.35356820e-01 6.97595000e-01 2.04493105e-01 -3.69685054e-01 -9.35395062e-01 1.04631484e+00 -4.21797752e-01 2.41957247e-01 1.11914650e-01 -8.18873525e-01 -7.81088948e-01 9.37043607e-01 1.22858870e+00 2.63245642e-01 -2.55742341e-01 -4.09758508e-01 -1.83069799e-02 4.63748306e-01 -4.48609084e-01 -1.63655415e-01 2.34754944e+00 3.40432465e-01 -1.92013681e-01 1.14750743e+00 1.33680832e+00 2.28454620e-01 -6.99397385e-01 -5.77651381e-01 6.60537779e-01 -6.10088170e-01 5.39624274e-01 -4.77842599e-01 -1.65954101e+00 4.71760511e-01 1.73032090e-01 8.11086416e-01 4.79673058e-01 -2.59282082e-01 1.50112665e+00 6.69131398e-01 1.13653921e-01 -1.26809216e+00 -3.11893970e-01 5.20033062e-01 1.10659289e+00 -1.24186218e+00 2.30532676e-01 2.77916461e-01 -6.72623575e-01 1.62345147e+00 -1.52658433e-01 -7.75571227e-01 1.54422176e+00 4.33861554e-01 5.12118459e-01 -3.18284750e-01 -1.08951795e+00 2.29155555e-01 2.56757170e-01 -2.13076845e-01 4.51359898e-01 -1.76056027e-01 -4.35663432e-01 1.23515296e+00 -5.50124407e-01 7.92791173e-02 7.83504367e-01 9.53657925e-01 -7.32488096e-01 -9.20175314e-01 -1.91164657e-01 7.16640353e-01 -1.48338354e+00 -3.72464508e-01 -4.04146254e-01 9.25214350e-01 -6.29475296e-01 3.08353186e-01 4.60739195e-01 -3.95804077e-01 3.92930150e-01 6.09456480e-01 -5.06644309e-01 -6.39495730e-01 -6.92050874e-01 2.56413996e-01 1.57174200e-01 -2.92136669e-01 -4.18567181e-01 -6.93444014e-01 -9.73392725e-01 -5.44619560e-01 -2.81944662e-01 1.51464626e-01 3.64294797e-01 7.08292544e-01 1.58051074e-01 5.70853889e-01 9.90855932e-01 -3.20643604e-01 -1.21589434e+00 -1.00003970e+00 -1.23523164e+00 1.24414809e-01 5.92361331e-01 -5.28874815e-01 -6.42023802e-01 -1.41692132e-01]	[4.434328079223633, 4.252329349517822]
669befbb-efc7-4c24-915b-722d41fab3d1	exploiting-negative-learning-for-implicit	2106.12123	null	https://arxiv.org/abs/2106.12123v1	https://arxiv.org/pdf/2106.12123v1.pdf	Exploiting Negative Learning for Implicit Pseudo Label Rectification in Source-Free Domain Adaptive Semantic Segmentation	It is desirable to transfer the knowledge stored in a well-trained source model onto non-annotated target domain in the absence of source data. However, state-of-the-art methods for source free domain adaptation (SFDA) are subject to strict limits: 1) access to internal specifications of source models is a must; and 2) pseudo labels should be clean during self-training, making critical tasks relying on semantic segmentation unreliable. Aiming at these pitfalls, this study develops a domain adaptive solution to semantic segmentation with pseudo label rectification (namely \textit{PR-SFDA}), which operates in two phases: 1) \textit{Confidence-regularized unsupervised learning}: Maximum squares loss applies to regularize the target model to ensure the confidence in prediction; and 2) \textit{Noise-aware pseudo label learning}: Negative learning enables tolerance to noisy pseudo labels in training, meanwhile positive learning achieves fast convergence. Extensive experiments have been performed on domain adaptive semantic segmentation benchmark, \textit{GTA5 $\to$ Cityscapes}. Overall, \textit{PR-SFDA} achieves a performance of 49.0 mIoU, which is very close to that of the state-of-the-art counterparts. Note that the latter demand accesses to the source model's internal specifications, whereas the \textit{PR-SFDA} solution needs none as a sharp contrast.	['Xiaogang Jia', 'Yulin He', 'Chen Li', 'Yusong Tan', 'Wei Chen', 'Xin Luo']	2021-06-23	null	null	null	null	['source-free-domain-adaptation']	['computer-vision']	[ 8.04808736e-01 4.26853478e-01 -4.70994502e-01 -5.56260765e-01 -1.04485810e+00 -5.60958028e-01 3.98396552e-01 -8.19238648e-02 -5.37771106e-01 8.72426867e-01 -3.34476948e-01 -2.32608616e-01 -1.06227577e-01 -4.89141345e-01 -7.75873005e-01 -7.76672184e-01 4.48871076e-01 7.59147108e-01 5.42179942e-01 -1.17499463e-01 8.83649066e-02 9.84678790e-03 -1.30751276e+00 1.41457180e-02 1.34788644e+00 1.45361662e+00 3.55875283e-01 3.82443696e-01 -4.84783977e-01 6.70546234e-01 -5.95834136e-01 -3.56489748e-01 3.96583825e-01 -5.78518450e-01 -1.08074617e+00 2.88085222e-01 1.86636418e-01 -8.51732641e-02 2.19249651e-01 1.42113256e+00 3.62476230e-01 1.51567563e-01 9.09837246e-01 -1.33972657e+00 -8.87512922e-01 4.63172197e-01 -5.81367731e-01 9.68266726e-02 -1.59065068e-01 5.89237846e-02 8.28837395e-01 -6.77897453e-01 5.86327374e-01 8.98088515e-01 6.75507545e-01 7.66386509e-01 -1.23420393e+00 -7.40258455e-01 4.86821443e-01 -1.57104626e-01 -1.36189783e+00 -4.36335623e-01 7.14804590e-01 -4.16599900e-01 6.61754131e-01 8.40102211e-02 9.40151215e-02 1.18601656e+00 -3.37611437e-01 9.69115496e-01 1.19034791e+00 -5.63435376e-01 5.96305490e-01 5.60993314e-01 3.30728233e-01 3.75521898e-01 1.28846362e-01 -1.19659670e-01 -3.30323011e-01 -2.94827353e-02 7.59265661e-01 -4.11089718e-01 -2.47938242e-02 -5.06461620e-01 -9.09599960e-01 7.58245051e-01 2.68108130e-01 1.89713880e-01 -1.25924513e-01 -4.15158421e-01 5.49814045e-01 3.21506888e-01 8.11560988e-01 1.88953042e-01 -9.57366347e-01 -1.34074120e-02 -9.30135608e-01 1.47916609e-02 6.54411495e-01 1.53987026e+00 7.74505973e-01 2.62482762e-01 1.78111792e-01 1.26753604e+00 3.78959507e-01 6.70973480e-01 6.70867205e-01 -1.08175123e+00 6.14014268e-01 6.72042131e-01 1.42521635e-01 -4.06788230e-01 -2.74112374e-01 -5.49410582e-01 -8.20308745e-01 1.39235422e-01 7.35029757e-01 -2.39971623e-01 -1.39928496e+00 1.66044605e+00 4.11282718e-01 2.46880531e-01 3.99445653e-01 8.44084084e-01 8.43348265e-01 4.51380789e-01 5.85834444e-01 -3.16136986e-01 1.10583138e+00 -1.02695072e+00 -5.69694459e-01 -7.39651620e-01 6.85390353e-01 -5.65921664e-01 1.40618610e+00 3.74349594e-01 -8.48421276e-01 -7.59954154e-01 -1.14066494e+00 4.51777466e-02 -4.84508693e-01 1.81081757e-01 2.26519421e-01 8.53462636e-01 -7.60964751e-01 3.64461005e-01 -7.38701284e-01 -4.16644663e-01 5.00568926e-01 4.48989719e-01 -1.52158856e-01 9.27604176e-03 -1.22569537e+00 6.62539959e-01 8.62644553e-01 -2.11683184e-01 -7.03708053e-01 -7.36015618e-01 -8.72510016e-01 -2.33196333e-01 7.48303771e-01 -4.41623569e-01 1.44224906e+00 -1.63072920e+00 -1.66388798e+00 1.26484740e+00 -1.08844638e-01 -3.81433994e-01 6.47758543e-01 -2.07166091e-01 -6.52162194e-01 1.27136126e-01 4.90967840e-01 8.61226618e-01 9.36132729e-01 -1.59523106e+00 -8.99994195e-01 -3.83221775e-01 -4.40403640e-01 3.29596996e-01 -3.57390046e-01 -1.38266534e-01 -6.60345137e-01 -8.94685507e-01 3.39137018e-01 -8.87827158e-01 -2.41605401e-01 -2.22391710e-01 -3.40987086e-01 -2.34236687e-01 8.77238810e-01 -6.69629753e-01 1.16722155e+00 -2.16530108e+00 -1.52173355e-01 2.48298690e-01 -1.99271396e-01 5.70240557e-01 -1.79223642e-02 -3.09448242e-01 -1.48816511e-01 8.10060427e-02 -8.65294397e-01 -4.29947406e-01 1.54430285e-01 3.20232987e-01 -2.02003181e-01 2.42410719e-01 2.30503008e-01 6.76338971e-01 -8.43410373e-01 -6.87977374e-01 2.15958163e-01 7.20314533e-02 -3.01497906e-01 1.08056866e-01 -5.59787571e-01 5.64479411e-01 -7.31916070e-01 7.16527998e-01 8.15747857e-01 -3.40265274e-01 5.90915792e-02 1.41067326e-01 1.82112187e-01 4.32432257e-02 -1.44615996e+00 1.87212431e+00 -1.56846151e-01 -6.73545003e-02 1.05465963e-01 -1.17522871e+00 1.08526635e+00 2.82909304e-01 5.92750967e-01 -8.26539338e-01 1.96080416e-01 5.53093553e-01 -5.61713457e-01 -2.81079680e-01 3.48296463e-01 -9.32497978e-02 -1.53664783e-01 2.43789986e-01 3.75474513e-01 -2.54685074e-01 5.77156618e-02 -6.51040077e-02 6.93999946e-01 6.46192670e-01 1.69623062e-01 -3.93493772e-01 5.85959733e-01 4.56360549e-01 9.72304165e-01 6.82876885e-01 -4.70553756e-01 7.76576638e-01 2.60590702e-01 -3.43232341e-02 -9.71713424e-01 -1.26708615e+00 -2.08115622e-01 1.37588608e+00 3.07631940e-01 1.29717678e-01 -1.08958554e+00 -1.13846588e+00 -3.70158821e-01 1.00715435e+00 -3.23445648e-01 -1.18844971e-01 -3.78689975e-01 -9.26042438e-01 4.95873898e-01 6.31959736e-01 9.74529028e-01 -9.57727015e-01 -3.01250786e-01 2.65608788e-01 -2.94756204e-01 -1.15152407e+00 -4.97258395e-01 5.41754007e-01 -8.76254737e-01 -8.34427297e-01 -8.98300052e-01 -1.06212378e+00 7.87195086e-01 -9.64672416e-02 9.90411639e-01 -4.75931287e-01 2.07631901e-01 2.16405541e-01 -5.31460464e-01 -5.15114069e-01 -5.89679241e-01 3.40187907e-01 -1.75182000e-01 -3.20360400e-02 7.21513808e-01 -4.73497599e-01 -4.33540136e-01 5.34550130e-01 -9.88600850e-01 -9.19496566e-02 4.74110663e-01 7.07434595e-01 1.15520477e+00 1.47157297e-01 9.50554550e-01 -1.36718059e+00 2.41712511e-01 -5.48695982e-01 -7.25040913e-01 3.32212836e-01 -9.57927763e-01 -1.60209522e-01 6.48379385e-01 -5.76973617e-01 -1.60285401e+00 4.60891932e-01 -2.81942368e-01 -3.06607991e-01 -5.50721049e-01 1.76742867e-01 -5.68338931e-01 2.28308693e-01 9.43105996e-01 2.81079590e-01 -1.03593186e-01 -5.76857805e-01 2.81309932e-01 9.35702801e-01 8.11340153e-01 -7.29244649e-01 6.02224767e-01 3.69706273e-01 -3.64665687e-01 -7.09930301e-01 -1.23673940e+00 -7.44189322e-01 -9.74454284e-01 1.73950449e-01 9.88462806e-01 -1.10615587e+00 -2.27216794e-03 6.65404022e-01 -6.44960284e-01 -6.92184627e-01 -5.04571795e-01 2.06367388e-01 -8.69158626e-01 3.46746266e-01 -3.56081843e-01 -6.85617983e-01 -3.07683051e-01 -9.70576048e-01 9.29311872e-01 4.55450654e-01 -1.67031750e-01 -1.22750771e+00 -3.80506128e-01 6.62686169e-01 1.83679521e-01 1.20976068e-01 7.50639796e-01 -1.18799114e+00 -1.17779866e-01 8.07125866e-02 -3.25067073e-01 9.67901707e-01 2.49868616e-01 -4.39125061e-01 -1.24126995e+00 -1.20899774e-01 1.59183547e-01 -3.68592441e-01 5.84887743e-01 5.22064805e-01 1.06273913e+00 -9.08955261e-02 -2.58114100e-01 4.23626304e-01 1.34040546e+00 3.79271328e-01 4.69630688e-01 5.39097190e-01 5.84969938e-01 4.48559105e-01 1.07429957e+00 3.10681939e-01 5.14120519e-01 3.83023679e-01 1.90161705e-01 -3.26502472e-01 -2.36294419e-01 -3.17896575e-01 2.15471148e-01 5.71496546e-01 2.77900964e-01 -2.28996843e-01 -9.64691043e-01 6.65484905e-01 -1.90824771e+00 -3.92964542e-01 -3.26290876e-01 2.22783709e+00 1.02381444e+00 4.66310352e-01 1.26458570e-01 1.74533457e-01 7.05830872e-01 -1.16754957e-01 -9.56436276e-01 -1.34624228e-01 -2.18583658e-01 1.45467326e-01 7.69591570e-01 2.46215045e-01 -1.57127774e+00 1.29465151e+00 5.22196054e+00 1.24950588e+00 -9.92870092e-01 5.19639850e-01 8.22876811e-01 4.06162202e-01 2.38846987e-02 -6.57635853e-02 -9.24575806e-01 5.02989173e-01 9.61189866e-01 1.62675202e-01 7.95570388e-02 1.21581352e+00 1.78867914e-02 -3.17412496e-01 -9.21608567e-01 7.52600133e-01 -2.16807257e-02 -8.39777529e-01 -1.85814887e-01 -1.67515859e-01 9.50407922e-01 8.79148990e-02 6.58674985e-02 5.44054985e-01 6.49335325e-01 -5.70833683e-01 9.07619357e-01 1.12044260e-01 1.03678679e+00 -5.92289925e-01 6.43046379e-01 6.35568440e-01 -9.74348128e-01 -2.63454895e-02 -2.77161419e-01 4.15280044e-01 2.62086868e-01 5.37756205e-01 -7.00156569e-01 7.62618423e-01 8.72257769e-01 6.34505510e-01 -4.21143979e-01 6.81435406e-01 -3.18893731e-01 9.93349850e-01 -3.55039060e-01 4.23578888e-01 3.19083422e-01 -2.48482451e-01 3.64690155e-01 1.22130740e+00 1.24158435e-01 1.57539025e-01 3.52497697e-01 8.18472266e-01 2.41550468e-02 2.16277868e-01 -2.25926772e-01 2.16894507e-01 3.61718714e-01 8.04556251e-01 -1.05947375e+00 -4.47668284e-01 -3.92464429e-01 1.22416925e+00 9.47938636e-02 5.92324138e-01 -8.41073990e-01 -2.33272731e-01 4.01028126e-01 1.83864743e-01 2.92002350e-01 5.34499027e-02 -8.80505204e-01 -9.80899751e-01 -4.68233647e-03 -7.29594409e-01 7.90208042e-01 -7.23444402e-01 -1.42729580e+00 5.09888470e-01 7.95075819e-02 -1.26882625e+00 -5.05245961e-02 -4.08891618e-01 1.97980199e-02 9.23344314e-01 -1.77908218e+00 -1.30792046e+00 -1.14881918e-01 7.94481099e-01 8.99923086e-01 -1.57619551e-01 7.59097576e-01 4.86252457e-01 -5.12318492e-01 8.02868009e-01 2.19405815e-01 1.18279986e-01 9.24110174e-01 -1.27737546e+00 2.56422848e-01 8.77246678e-01 -1.50311038e-01 1.13636293e-01 6.28343940e-01 -8.75862062e-01 -5.93629539e-01 -1.46958494e+00 7.25272536e-01 -5.02753139e-01 4.80057269e-01 -2.14585155e-01 -1.12392879e+00 8.00288379e-01 -2.18724832e-01 2.35815402e-02 6.12204790e-01 -1.95119709e-01 -1.85311496e-01 -1.04671255e-01 -1.51149511e+00 3.10202241e-01 1.08557570e+00 -2.45062634e-01 -6.65148079e-01 3.01891655e-01 9.05709088e-01 -4.76852119e-01 -8.55640233e-01 5.12928247e-01 -4.14901525e-02 -6.54673040e-01 9.31468725e-01 -2.87295014e-01 1.07425101e-01 -3.84735465e-01 -1.75351039e-01 -1.15124655e+00 -6.26134649e-02 -3.38987112e-01 2.54063457e-01 1.71879649e+00 6.54071331e-01 -6.35473311e-01 8.70306909e-01 8.91772091e-01 -4.61384922e-01 -1.28007814e-01 -1.12606335e+00 -9.95714545e-01 3.33177745e-01 -5.33775151e-01 3.96820694e-01 1.24874139e+00 -4.84924018e-01 2.22726211e-01 -1.22839101e-01 3.34629595e-01 6.86124563e-01 -3.60599101e-01 4.87513959e-01 -1.33364737e+00 -1.01584986e-01 -3.50893021e-01 2.33941317e-01 -1.27847683e+00 3.35873663e-01 -8.15492392e-01 2.17335314e-01 -1.47638083e+00 -2.40449924e-02 -1.09572172e+00 -4.46990609e-01 6.75127923e-01 -2.07852125e-01 1.93350658e-01 -8.44271109e-02 2.56302744e-01 -7.31060445e-01 4.76973653e-01 1.01713324e+00 -1.62413001e-01 -4.34557527e-01 3.30621064e-01 -8.52904081e-01 9.51269865e-01 7.22285151e-01 -6.36711657e-01 -6.95714295e-01 -4.38468993e-01 -2.16189548e-01 -1.26336962e-01 1.58418402e-01 -8.48426104e-01 1.13948345e-01 -3.87162656e-01 1.54427096e-01 -4.01532799e-01 1.65918708e-01 -9.65855062e-01 -4.26441059e-03 -2.42565256e-02 -2.46948212e-01 -6.05709732e-01 1.71725705e-01 5.85721910e-01 -2.59511143e-01 -4.75848705e-01 1.17148483e+00 -2.51999021e-01 -1.15381682e+00 1.33649454e-01 -1.55637473e-01 5.49043894e-01 1.01224697e+00 -5.70275486e-01 -7.80300349e-02 1.26738846e-01 -8.54499638e-01 2.98461407e-01 4.01348919e-01 4.11348432e-01 1.52911589e-01 -9.99108672e-01 -4.42190558e-01 1.25507459e-01 2.23630115e-01 6.52656615e-01 3.21537197e-01 5.39513528e-01 -1.81716487e-01 1.84184656e-01 6.03947230e-02 -7.36769080e-01 -9.28389788e-01 5.56916535e-01 3.40096503e-01 -2.68103033e-01 -5.52702010e-01 9.59487617e-01 2.75376469e-01 -8.28042567e-01 3.48676801e-01 -1.25069395e-01 -1.03774726e-01 -7.41526065e-03 2.09989995e-01 2.45846808e-01 2.92429090e-01 -6.20615661e-01 -1.97489992e-01 4.79202747e-01 -1.29719764e-01 -1.33077269e-02 1.24022353e+00 -4.87368941e-01 3.06894243e-01 4.00681138e-01 9.00524914e-01 -4.40404832e-01 -1.77502477e+00 -6.26572490e-01 3.87307763e-01 -1.58762023e-01 4.34210291e-03 -1.35161150e+00 -9.16820526e-01 6.07625544e-01 7.24807739e-01 -1.03287041e-01 1.42857504e+00 2.72973217e-02 8.51463675e-01 9.21177492e-02 3.78366768e-01 -1.79439247e+00 1.31131262e-02 4.38996762e-01 4.76443529e-01 -1.61633909e+00 -2.54084677e-01 -6.69953525e-01 -1.19040978e+00 6.39993191e-01 7.15775847e-01 7.93246478e-02 7.40899205e-01 8.70567188e-02 2.72690445e-01 1.72580332e-01 -2.12036327e-01 -3.09614956e-01 2.25161672e-01 8.39904070e-01 -5.65014519e-02 -3.50060835e-02 -1.17822401e-01 1.12970948e+00 3.74905318e-02 1.60478368e-01 1.04904048e-01 9.70465302e-01 -4.87516016e-01 -1.08256304e+00 -3.75221342e-01 2.93671012e-01 -4.51798588e-01 -3.78231928e-02 -3.37958515e-01 7.69055486e-01 4.45094049e-01 1.15840566e+00 -1.97848901e-01 1.45442545e-01 5.84781826e-01 3.89614224e-01 -5.14279455e-02 -7.40747452e-01 -4.39835936e-01 4.04595971e-01 9.08383280e-02 -2.02831730e-01 -5.25462270e-01 -7.72537053e-01 -1.61382639e+00 2.36642748e-01 -3.16127181e-01 5.72659560e-02 5.79734385e-01 1.01551294e+00 1.90343842e-01 4.50811684e-01 2.91045040e-01 -3.66785258e-01 -5.80770969e-01 -9.59558487e-01 -6.83042347e-01 5.79046428e-01 9.60655510e-02 -6.37839973e-01 -3.42376202e-01 6.43162966e-01]	[9.620694160461426, 1.3619030714035034]
d2a59b7f-7ca8-4f8d-989a-4f0edc84aacf	doremi-grounding-language-model-by-detecting	2307.00329	null	https://arxiv.org/abs/2307.00329v1	https://arxiv.org/pdf/2307.00329v1.pdf	DoReMi: Grounding Language Model by Detecting and Recovering from Plan-Execution Misalignment	Large language models encode a vast amount of semantic knowledge and possess remarkable understanding and reasoning capabilities. Previous research has explored how to ground language models in robotic tasks to ensure that the sequences generated by the language model are both logically correct and practically executable. However, low-level execution may deviate from the high-level plan due to environmental perturbations or imperfect controller design. In this paper, we propose DoReMi, a novel language model grounding framework that enables immediate Detection and Recovery from Misalignments between plan and execution. Specifically, during low-level skill execution, we use a vision question answering (VQA) model to regularly detect plan-execution misalignments. If certain misalignment occurs, our method will call the language model to re-plan in order to recover from misalignments. Experiments on various complex tasks including robot arms and humanoid robots demonstrate that our method can lead to higher task success rates and shorter task completion times. Videos of DoReMi are available at https://sites.google.com/view/doremi-paper.	['Jianyu Chen', 'Zheyuan Jiang', 'Lihan Zha', 'Yen-Jen Wang', 'Yanjiang Guo']	2023-07-01	null	null	null	null	['question-answering']	['natural-language-processing']	[ 4.30119902e-01 3.49355489e-01 2.27178037e-02 -2.00678512e-01 -4.83481735e-01 -6.39161587e-01 5.51135242e-01 -3.05593442e-02 9.22425389e-02 4.76219684e-01 -2.75655203e-02 -3.64329636e-01 -1.67995319e-01 -5.18751204e-01 -8.48684669e-01 1.18704729e-01 1.27393782e-01 5.61652720e-01 3.63166690e-01 -3.33498895e-01 5.97630382e-01 1.93438008e-01 -1.69444942e+00 3.08621764e-01 9.20678854e-01 5.08762836e-01 7.68091679e-01 7.41024494e-01 4.88230173e-04 1.33206928e+00 -5.05130112e-01 3.06104243e-01 3.10317487e-01 -4.64137405e-01 -1.15676475e+00 3.61550480e-01 -8.90973061e-02 -4.11195517e-01 -2.81705052e-01 1.31780291e+00 -1.70873582e-01 2.20810007e-02 2.46969715e-01 -1.75688422e+00 -3.46116990e-01 6.54219985e-01 -1.94351226e-01 -1.76085189e-01 8.88037026e-01 5.09952128e-01 5.73645353e-01 -3.59894395e-01 4.73025233e-01 1.32670116e+00 4.11141157e-01 6.67697608e-01 -7.77275443e-01 -3.94376129e-01 3.79521072e-01 2.46797174e-01 -1.47581732e+00 -5.04278660e-01 4.71805900e-01 -6.61883235e-01 1.27680242e+00 -5.85728921e-02 3.52059364e-01 9.95853841e-01 7.30271757e-01 5.52379847e-01 9.89001334e-01 -5.22856295e-01 2.27409869e-01 -1.99617043e-01 1.49753783e-02 1.05227637e+00 5.00537276e-01 2.69193739e-01 -6.64636493e-01 -4.88772877e-02 9.07376707e-01 5.13647348e-02 -1.65111244e-01 -4.02897924e-01 -1.57543933e+00 1.72428235e-01 -3.92935798e-02 2.10727394e-01 -6.68046057e-01 2.97576457e-01 2.62874275e-01 4.46898758e-01 -4.56165045e-01 1.03982925e+00 -2.16183677e-01 -4.02404755e-01 -3.56604159e-01 3.71216148e-01 9.48488712e-01 1.41689610e+00 8.39449406e-01 -3.96256447e-02 -1.40526697e-01 1.70415446e-01 2.19862908e-01 3.82168472e-01 5.84364176e-01 -1.38823557e+00 5.15343487e-01 9.78554308e-01 5.83667040e-01 -1.12207115e+00 -4.41135913e-01 8.29560086e-02 -1.90367833e-01 2.71315217e-01 1.17718808e-01 1.25983268e-01 -7.86858678e-01 1.50352609e+00 2.26802275e-01 3.63929927e-01 3.81731659e-01 1.04779947e+00 1.35917291e-01 4.78471071e-01 -1.28699541e-01 1.07377991e-02 1.45078862e+00 -9.74490821e-01 -8.83078992e-01 -1.01468349e+00 8.50571334e-01 -4.90160346e-01 1.20050919e+00 6.66553676e-01 -9.29861426e-01 -6.37566686e-01 -1.23347926e+00 1.89594015e-01 1.62452713e-01 2.03390718e-01 4.43020850e-01 -3.86438593e-02 -8.37965012e-01 5.37813544e-01 -1.11323893e+00 -3.69790673e-01 -1.77995581e-02 1.74204841e-01 -2.65485942e-01 -3.01874280e-01 -8.18347871e-01 9.72787499e-01 7.70797372e-01 3.17827344e-01 -1.52807856e+00 -1.40481383e-01 -1.07143700e+00 -3.21446151e-01 1.15361834e+00 -5.85738420e-01 2.11481547e+00 -8.02571118e-01 -1.58177602e+00 5.73579311e-01 -5.15653975e-02 -5.29567480e-01 1.90283537e-01 -5.15999377e-01 -1.21103399e-01 -2.65470762e-02 2.70453811e-01 3.51994246e-01 7.36291409e-01 -1.26424122e+00 -6.85689151e-01 -4.18255031e-01 4.39935982e-01 3.47122401e-01 3.07354778e-01 -1.48359478e-01 -3.51205945e-01 5.13430722e-02 2.53511429e-01 -1.24200439e+00 -1.60641685e-01 -3.84056211e-01 -4.41525698e-01 -1.70400351e-01 4.06391829e-01 -6.51736081e-01 1.14775968e+00 -1.86347413e+00 2.42069960e-01 3.62726748e-02 1.52160525e-01 1.61524966e-01 -3.76616269e-01 6.81864619e-01 2.52344191e-01 -1.58717126e-01 -1.55693635e-01 6.40010685e-02 6.04721904e-02 5.10757864e-01 -5.63450634e-01 2.28725359e-01 1.70504257e-01 7.57347047e-01 -1.04225361e+00 -2.81201035e-01 1.69781536e-01 -3.50380599e-01 -4.36910838e-01 5.08841813e-01 -8.23978603e-01 5.75616956e-01 -8.58520150e-01 6.96517885e-01 6.09708093e-02 -1.91851884e-01 5.50334871e-01 2.64353514e-01 5.32634556e-02 4.15923715e-01 -1.16148186e+00 1.90028906e+00 -5.40735543e-01 3.87603462e-01 1.21577054e-01 -8.63074958e-01 9.62211728e-01 3.06255966e-01 -3.08839623e-02 -7.08440781e-01 1.75013006e-01 1.32788464e-01 1.13071956e-01 -8.77303302e-01 5.49682915e-01 3.80657203e-02 -2.55156785e-01 4.74450260e-01 -4.41436321e-01 -7.03300238e-01 1.16006508e-01 -4.47683930e-02 1.23394895e+00 5.43972313e-01 7.02746630e-01 1.21621422e-01 4.80671942e-01 7.56647229e-01 5.78800917e-01 9.89026904e-01 -3.58527154e-01 2.31302738e-01 4.43162024e-01 -3.77643377e-01 -7.99717665e-01 -5.98557413e-01 8.74906003e-01 9.14409697e-01 4.89741653e-01 -6.61035419e-01 -8.69684637e-01 -2.71926582e-01 -4.18226480e-01 1.19154382e+00 -7.52944127e-02 -5.87863207e-01 -5.74294329e-01 1.81326047e-01 5.65143645e-01 5.00915170e-01 4.74405408e-01 -1.49400520e+00 -1.41557992e+00 1.64951563e-01 -5.36467612e-01 -1.27618992e+00 -3.45504552e-01 -8.20349306e-02 -8.54476690e-01 -1.56942225e+00 2.01151729e-01 -8.18216980e-01 7.84806907e-01 4.00379032e-01 9.96001899e-01 3.77917200e-01 -7.26755485e-02 7.53986597e-01 -5.00649869e-01 -4.51294988e-01 -9.08782899e-01 -4.28381413e-01 2.47159228e-01 -5.01663923e-01 3.25223535e-01 2.88754236e-02 -1.22437850e-01 3.35446447e-01 -7.66417861e-01 3.69868547e-01 5.72567523e-01 6.37084246e-01 6.78996623e-01 4.42721426e-01 1.36609390e-01 -4.62686449e-01 1.04475880e+00 -1.44882664e-01 -8.53413343e-01 4.02590364e-01 -8.64386439e-01 1.09657049e-01 5.28679967e-01 -2.61128247e-01 -8.86087596e-01 8.66205022e-02 5.24087071e-01 -5.22004902e-01 -5.19785047e-01 7.32232034e-01 -2.41047908e-02 5.00835180e-01 7.48806834e-01 3.84816498e-01 2.13079721e-01 -1.57821894e-01 9.74846408e-02 6.28454566e-01 8.23788464e-01 -9.18412149e-01 6.78170919e-01 2.39307731e-01 -2.50318766e-01 -4.04340595e-01 -7.10044265e-01 -2.82431573e-01 -3.93180370e-01 -2.43575826e-01 7.33776689e-01 -9.48397458e-01 -6.37004972e-01 4.84858513e-01 -1.34007049e+00 -7.89631128e-01 3.39008383e-02 3.41365606e-01 -1.11575961e+00 2.70446658e-01 -2.94588119e-01 -1.08057594e+00 1.30380795e-03 -1.43433082e+00 1.05198824e+00 1.84687793e-01 -5.51073730e-01 -4.04698223e-01 -7.52619430e-02 4.11090732e-01 -4.78823632e-02 9.49425623e-02 7.14206219e-01 -5.43960869e-01 -6.33932054e-01 -1.42373279e-01 2.74662822e-01 1.48711100e-01 4.50196326e-01 -2.62288511e-01 -4.25021499e-01 -3.15160304e-01 8.44621807e-02 -4.55771118e-01 -5.23007363e-02 7.34662414e-02 1.07731605e+00 -5.29127955e-01 -3.28375816e-01 -5.43451756e-02 1.16013062e+00 5.41394234e-01 6.53893888e-01 5.57267904e-01 4.81450468e-01 8.10344756e-01 1.29925799e+00 4.77759510e-01 5.00841081e-01 5.82538128e-01 7.05391705e-01 6.38549805e-01 1.00781888e-01 -5.99566877e-01 7.42135346e-01 5.23337245e-01 2.22130433e-01 -6.52562380e-02 -1.47195780e+00 6.53511107e-01 -2.17911434e+00 -6.65360808e-01 1.16586484e-01 2.11439514e+00 7.23877311e-01 2.64948279e-01 -1.44414544e-01 -6.71150908e-02 6.16365969e-01 -2.25376874e-01 -6.81080997e-01 -3.80567342e-01 5.67470133e-01 -3.52677494e-01 1.79569468e-01 5.13682246e-01 -7.20744908e-01 1.18976784e+00 5.71676207e+00 2.51120597e-01 -7.55955637e-01 -4.74126041e-02 -2.80460894e-01 1.38349652e-01 -4.55500409e-02 2.74424314e-01 -4.98104215e-01 -1.16002222e-03 8.82594943e-01 -3.54988694e-01 7.26566374e-01 1.03273511e+00 2.58782744e-01 -3.24130476e-01 -1.41715419e+00 9.37166214e-01 3.83340381e-02 -1.06783938e+00 -1.98048696e-01 -2.81418473e-01 4.74632472e-01 -1.12306915e-01 -3.14992845e-01 4.98335987e-01 4.76789773e-01 -1.06573224e+00 1.17434609e+00 7.06359565e-01 4.15306091e-01 -4.96133238e-01 5.74317038e-01 1.09361839e+00 -9.92071867e-01 -2.93233901e-01 -2.80630380e-01 -5.61010301e-01 1.67294368e-01 -6.68532727e-03 -1.32684565e+00 7.09057033e-01 4.85152185e-01 4.95182037e-01 -3.03656012e-01 5.48601270e-01 -5.70706606e-01 1.08548939e-01 -4.51422250e-03 6.54218793e-02 6.65951669e-02 -2.06517175e-01 6.96092427e-01 5.19162416e-01 3.07902694e-01 1.40528321e-01 6.58939898e-01 9.36293244e-01 5.63732207e-01 -6.09049380e-01 -1.05587721e+00 -5.98387659e-01 7.16934621e-01 5.80549061e-01 -5.07243693e-01 -3.51785094e-01 -1.43309459e-01 1.05338395e+00 2.63054788e-01 4.25049454e-01 -8.13340008e-01 -7.61296675e-02 7.80678153e-01 -3.64791341e-02 -4.35616374e-02 -6.97378099e-01 -1.56542808e-01 -8.63750875e-01 1.83504015e-01 -1.31838143e+00 2.67122742e-02 -1.23813462e+00 -6.92403257e-01 4.95428383e-01 2.45368481e-01 -1.22982073e+00 -5.73621571e-01 -5.83706200e-01 -3.37854743e-01 5.29306412e-01 -1.25026190e+00 -6.13359869e-01 -4.61364448e-01 3.96315306e-01 8.34798753e-01 -1.85058117e-01 7.50754297e-01 -4.64468449e-01 -4.48399574e-01 -1.30273074e-01 -7.86129296e-01 -2.15553030e-01 4.61580545e-01 -8.92751932e-01 2.84600407e-01 1.00661898e+00 -2.14134175e-02 8.01631153e-01 9.96303618e-01 -1.11322665e+00 -1.88310325e+00 -1.15696943e+00 7.48223126e-01 -5.52546561e-01 6.10138059e-01 1.13650627e-01 -9.04871106e-01 1.25435615e+00 -2.35933773e-02 -5.54293394e-01 1.89984530e-01 -3.94399554e-01 -2.94803262e-01 1.37765065e-01 -9.14921761e-01 8.82597864e-01 1.25589800e+00 -7.23836422e-01 -1.35705149e+00 5.59899747e-01 7.91754603e-01 -7.93918729e-01 -5.39896727e-01 2.80682951e-01 3.49996090e-01 -8.34334910e-01 6.12525344e-01 -6.64218724e-01 4.82694209e-01 -8.91618431e-01 -2.08833963e-01 -1.13566458e+00 -1.17206722e-01 -8.19010735e-01 -2.61323005e-01 6.61347866e-01 6.12021945e-02 -6.28007770e-01 1.96329013e-01 7.57582247e-01 -4.12792027e-01 -3.24240685e-01 -5.97873271e-01 -1.12517738e+00 -5.52053511e-01 -6.99626982e-01 6.17799878e-01 5.23031056e-01 3.62050444e-01 3.04694623e-01 -3.42004061e-01 6.51851118e-01 3.38682443e-01 2.31633335e-01 9.54258084e-01 -9.56723332e-01 -2.57466644e-01 -2.55412370e-01 -8.08595717e-02 -1.15509689e+00 5.85230827e-01 -6.52920365e-01 9.03380156e-01 -1.75767672e+00 1.19849173e-02 -2.55392104e-01 -2.83969156e-02 9.17445242e-01 2.32493043e-01 -4.66193557e-01 3.39692831e-01 4.04721439e-01 -8.45122874e-01 4.30676818e-01 1.34093893e+00 -1.13693491e-01 -4.17374611e-01 -2.26730645e-01 -7.46623158e-01 8.36459160e-01 1.17965710e+00 -3.61017287e-01 -7.47649133e-01 -7.37509131e-01 4.65363503e-01 4.34262961e-01 5.14521480e-01 -1.20342219e+00 4.54926342e-01 -7.22004473e-01 -4.49922353e-01 -2.90435642e-01 1.38314649e-01 -9.59698439e-01 3.58210623e-01 8.24584365e-01 -3.99131656e-01 3.93455684e-01 3.48218799e-01 7.49593318e-01 -1.47188038e-01 -3.99228930e-01 2.32745394e-01 -3.06093633e-01 -1.32032466e+00 -9.06012431e-02 -7.72741437e-01 -1.79796413e-01 1.40301919e+00 -3.69802088e-01 -4.26038593e-01 -2.33236700e-01 -5.13571262e-01 6.99140012e-01 6.22795165e-01 8.00246477e-01 8.48554790e-01 -9.55148757e-01 -3.48096371e-01 1.43605456e-01 4.57668513e-01 2.77953714e-01 3.09782121e-02 8.08706462e-01 -4.81938452e-01 5.77438593e-01 -2.82506675e-01 -5.21114886e-01 -9.90524054e-01 5.84709227e-01 5.00530839e-01 -6.60144836e-02 -7.52834499e-01 5.16043425e-01 1.95360616e-01 -5.98524213e-01 1.90062255e-01 -6.94255948e-01 -7.96820968e-02 -7.24776208e-01 6.28732204e-01 1.70289829e-01 -3.79986726e-02 -3.52232128e-01 -7.90477216e-01 4.02195424e-01 1.27477914e-01 -1.11454338e-01 8.10347855e-01 -4.35664862e-01 -9.14461017e-02 5.31370103e-01 2.84156024e-01 -4.59058702e-01 -1.44324446e+00 -2.24014938e-01 4.14174020e-01 -2.73568898e-01 -1.01965621e-01 -8.46600413e-01 -2.81859756e-01 5.31242907e-01 1.38146922e-01 2.39421893e-02 8.84466887e-01 9.71628353e-02 5.57684481e-01 9.28231061e-01 1.05398357e+00 -1.20375144e+00 4.95474398e-01 8.81175280e-01 1.18088472e+00 -9.98560965e-01 -1.80777729e-01 -2.97851205e-01 -1.04193580e+00 8.66879702e-01 1.21762276e+00 -3.44773121e-02 -3.16588767e-02 3.02940607e-01 6.60771206e-02 -4.11052078e-01 -9.57339942e-01 -1.67515144e-01 -1.75056070e-01 9.27334368e-01 -1.42025515e-01 2.11582571e-01 7.99944103e-02 6.08333290e-01 -2.95418054e-01 4.02477145e-01 9.75754678e-01 1.54549038e+00 -8.42655063e-01 -7.98381984e-01 -6.22955739e-01 1.68306246e-01 1.23682931e-01 3.22779924e-01 -5.73808372e-01 5.50901055e-01 -1.73968464e-01 1.41667366e+00 9.58869830e-02 -5.31385660e-01 6.71747983e-01 2.25351065e-01 7.39318967e-01 -1.13160992e+00 -8.89209211e-02 -3.33274543e-01 3.33457172e-01 -1.13920689e+00 -2.92888910e-01 -5.09058952e-01 -1.86895549e+00 -3.79532985e-02 2.09371746e-02 -1.76935613e-01 4.07403141e-01 1.22922492e+00 5.81572592e-01 7.82246768e-01 7.19329193e-02 -6.72526777e-01 -9.89989936e-01 -7.28919208e-01 -3.15952033e-01 2.12471440e-01 4.43375409e-01 -8.54795694e-01 4.28317934e-02 3.41459215e-01]	[4.430872440338135, 0.8886126279830933]
9b28c874-9fcc-43f9-9ae9-6d8e62d67322	e-lmc-extended-linear-model-of	2203.00525	null	https://arxiv.org/abs/2203.00525v2	https://arxiv.org/pdf/2203.00525v2.pdf	E-LMC: Extended Linear Model of Coregionalization for Spatial Field Prediction	Physical simulations based on partial differential equations typically generate spatial fields results, which are utilized to calculate specific properties of a system for engineering design and optimization. Due to the intensive computational burden of the simulations, a surrogate model mapping the low-dimensional inputs to the spatial fields are commonly built based on a relatively small dataset. To resolve the challenge of predicting the whole spatial field, the popular linear model of coregionalization (LMC) can disentangle complicated correlations within the high-dimensional spatial field outputs and deliver accurate predictions. However, LMC fails if the spatial field cannot be well approximated by a linear combination of base functions with latent processes. In this paper, we present the Extended Linear Model of Coregionalization (E-LMC) by introducing an invertible neural network to linearize the highly complex and nonlinear spatial fields so that the LMC can easily generalize to nonlinear problems while preserving the traceability and scalability. Several real-world applications demonstrate that E-LMC can exploit spatial correlations effectively, showing a maximum improvement of about 40% over the original LMC and outperforming the other state-of-the-art spatial field models.	['Wei W. Xing', 'Yichen Meng', 'Xueying Zhang', 'Shihong Wang']	2022-03-01	null	null	null	null	['physical-simulations']	['miscellaneous']	[-1.01894408e-01 -3.74856710e-01 3.24871913e-02 -7.65094161e-02 -5.20014763e-01 -5.84028900e-01 5.44592857e-01 -1.97936967e-01 2.17254922e-01 9.57081914e-01 2.17482768e-04 -4.99603271e-01 -7.33595848e-01 -9.05120432e-01 -8.17768931e-01 -8.69042039e-01 -1.29976228e-01 1.57292351e-01 8.62940997e-02 -2.13677049e-01 2.40330309e-01 7.51105845e-01 -1.18496728e+00 1.98763371e-01 1.00865221e+00 1.09690654e+00 5.24915457e-01 5.08086681e-01 5.59394248e-02 6.76679254e-01 -1.88701391e-01 4.21893537e-01 9.72195715e-02 -1.74931258e-01 -4.63171989e-01 -3.08040708e-01 1.69066519e-01 -1.87981784e-01 -5.80352545e-01 7.30125308e-01 2.71351933e-01 2.18010768e-01 8.56358886e-01 -1.13476658e+00 -9.38366890e-01 1.09221786e-01 -2.28659540e-01 -1.16078846e-01 7.37377107e-02 2.55004793e-01 6.35336995e-01 -1.07838953e+00 3.65377396e-01 1.28666306e+00 9.50981021e-01 -1.05834872e-01 -1.74846959e+00 -5.41404426e-01 2.75840998e-01 -7.64463171e-02 -1.72074807e+00 1.19070306e-01 8.03121924e-01 -8.66243899e-01 8.79777253e-01 1.35529757e-01 5.56433678e-01 8.44110072e-01 7.60502756e-01 3.26319754e-01 1.30623507e+00 -5.88254817e-02 4.45469141e-01 -1.11673981e-01 -9.36243087e-02 2.63801515e-01 -9.30148289e-02 4.45557415e-01 -6.19623542e-01 -2.59325176e-01 1.30652559e+00 2.14984491e-01 -2.16088653e-01 -4.51755434e-01 -1.30598474e+00 4.82709736e-01 7.60376096e-01 1.63373902e-01 -5.75747132e-01 3.64352643e-01 -2.27728516e-01 8.63337666e-02 4.73171800e-01 9.65465128e-01 -3.46538305e-01 -3.55023472e-03 -1.12214470e+00 4.48989421e-01 6.09220147e-01 9.45474267e-01 9.14771497e-01 2.04534084e-01 -2.59555459e-01 5.25498927e-01 3.10381770e-01 9.88738894e-01 -7.86613598e-02 -1.08856726e+00 3.92755657e-01 6.01427972e-01 3.93878222e-01 -1.07715809e+00 -5.24124801e-01 -6.99520230e-01 -1.25138307e+00 2.10393369e-01 2.58186698e-01 -2.75023639e-01 -8.31513762e-01 1.69623113e+00 -6.72249123e-02 3.63712281e-01 -1.19288325e-01 7.88615644e-01 2.02772483e-01 1.27719271e+00 1.64283156e-01 2.23983824e-02 7.35259354e-01 -5.11813104e-01 -7.32241988e-01 -1.89113751e-01 4.53281164e-01 -5.90225697e-01 9.70739901e-01 7.56463334e-02 -8.67398322e-01 -5.49464822e-01 -1.05720484e+00 2.10931674e-01 -6.15759850e-01 2.02282846e-01 6.74919665e-01 5.44355102e-02 -1.15173864e+00 9.03959572e-01 -1.00842381e+00 -1.89358797e-02 2.48981595e-01 3.79964948e-01 -2.42288083e-01 2.71452847e-03 -1.22862852e+00 7.82260597e-01 2.22256463e-02 5.00002921e-01 -8.10599148e-01 -1.56534374e+00 -6.85723901e-01 1.74528211e-01 2.87527025e-01 -4.28776473e-01 7.06378698e-01 -1.47016212e-01 -1.59556425e+00 -1.78605467e-01 -1.82463318e-01 1.12514608e-01 2.61373729e-01 -2.82579273e-01 -4.31351036e-01 -2.31963649e-01 2.73665279e-01 3.31715524e-01 4.67342645e-01 -1.23757327e+00 -2.52822965e-01 -8.61375965e-03 -7.42565989e-02 -4.86322045e-02 -1.66231841e-01 -5.11497617e-01 -2.51227975e-01 -8.08285475e-01 2.79359102e-01 -1.02869022e+00 -5.69210112e-01 1.67621791e-01 -3.89842391e-01 2.86430031e-01 9.75651622e-01 -7.79845417e-01 1.35383570e+00 -2.04520845e+00 3.25760990e-01 2.70033330e-01 2.26132497e-01 7.55395368e-02 -5.69390580e-02 7.76786327e-01 -9.49723423e-02 7.62147382e-02 -4.49373513e-01 -9.86777470e-02 -2.00140998e-02 1.50262490e-01 -7.55438507e-01 4.73555148e-01 4.90663379e-01 1.16359174e+00 -9.56059158e-01 -6.55286312e-02 4.97542232e-01 6.06206179e-01 -5.58711767e-01 9.87498760e-02 -1.99229255e-01 7.26152539e-01 -7.50762820e-01 3.59098107e-01 8.58256459e-01 -7.55495846e-01 1.35061860e-01 -2.98399746e-01 -4.28205162e-01 5.67430668e-02 -1.29928851e+00 1.55450070e+00 -9.21121061e-01 7.16936886e-01 2.74172984e-02 -8.68770540e-01 9.06043053e-01 1.59073859e-01 7.02460349e-01 -8.23911309e-01 -3.85695606e-01 3.61857325e-01 -2.67120004e-01 -8.34895298e-02 3.40505987e-01 -3.85368735e-01 -3.33266020e-01 4.64801699e-01 -7.40859061e-02 -3.26592833e-01 -3.38897318e-01 -8.43496397e-02 9.02546883e-01 4.57450837e-01 -9.97538343e-02 -8.68656993e-01 4.98570651e-01 -2.72241123e-02 5.66242754e-01 7.08135784e-01 3.32284600e-01 5.93527198e-01 5.57462633e-01 -3.77611995e-01 -1.10919499e+00 -1.31750238e+00 -3.69671345e-01 5.10393739e-01 4.20964152e-01 -1.46835372e-01 -5.80191433e-01 -1.09612837e-01 3.96802396e-01 6.98459148e-01 -5.66973269e-01 -2.39744708e-01 -7.36210406e-01 -8.39533269e-01 3.55061948e-01 8.67376864e-01 4.53683406e-01 -6.84417903e-01 -3.34965318e-01 3.97652030e-01 1.27766997e-01 -1.12475646e+00 -1.98008284e-01 2.01065004e-01 -6.70671642e-01 -4.76780325e-01 -7.98173189e-01 -4.89924490e-01 5.25468349e-01 -9.75811705e-02 8.29859793e-01 -2.84274101e-01 -1.13208771e-01 -1.11654729e-01 2.01169431e-01 -3.37837338e-02 -1.29982010e-01 1.74520463e-01 2.49710605e-01 8.99250358e-02 -1.45588145e-01 -1.01449728e+00 -5.59119225e-01 5.64979494e-01 -7.28790522e-01 3.56894135e-01 6.61511481e-01 8.50973248e-01 6.64619148e-01 2.15304032e-01 6.86340749e-01 -5.69527626e-01 5.03178596e-01 -6.17928982e-01 -9.07633662e-01 3.84390235e-01 -6.23448312e-01 5.45101047e-01 9.96372581e-01 -6.13451004e-01 -1.33938432e+00 3.11373137e-02 2.51004040e-01 -5.59083462e-01 1.84164479e-01 7.35171854e-01 -1.71734363e-01 -2.57115930e-01 5.13863325e-01 1.55243397e-01 -1.25180840e-01 -5.92401266e-01 4.06047910e-01 3.88523668e-01 4.74434882e-01 -7.81518221e-01 9.54029799e-01 4.27023888e-01 7.76739538e-01 -7.80715168e-01 -6.33597076e-01 -2.19391227e-01 -8.75541091e-01 -3.74401242e-01 6.24019027e-01 -8.19660962e-01 -7.58796215e-01 4.98984486e-01 -1.11627281e+00 -7.18055844e-01 -3.12737256e-01 6.12961948e-01 -6.91229105e-01 -1.72071144e-01 -4.12667722e-01 -8.44797313e-01 4.25584525e-01 -1.17512870e+00 1.30601776e+00 -1.26284165e-02 -2.08727390e-01 -1.37208343e+00 5.86695448e-02 -5.05958736e-01 8.91093850e-01 3.50546062e-01 1.35797429e+00 3.40021908e-01 -9.62810338e-01 -9.88112018e-02 -2.31995076e-01 8.51526856e-02 2.20723569e-01 -2.17966229e-01 -8.70182633e-01 -1.39798045e-01 7.47377872e-02 1.73149303e-01 5.23428082e-01 7.83842564e-01 1.11783552e+00 -1.02834426e-01 -5.89003265e-01 1.04961777e+00 1.66612339e+00 1.03903696e-01 6.33736372e-01 -4.71712165e-02 8.28102589e-01 5.68479538e-01 4.54563707e-01 2.59755522e-01 7.28145018e-02 6.95578933e-01 9.16651860e-02 -3.32714766e-01 -9.32756737e-02 -6.91020548e-01 -8.37547705e-03 8.52024078e-01 5.07198460e-02 -1.91039771e-01 -1.26216280e+00 5.26587009e-01 -2.04249096e+00 -6.85183942e-01 -1.25981107e-01 1.98438001e+00 4.84917700e-01 -2.12764785e-01 -4.35313016e-01 -2.52734363e-01 6.67047203e-01 2.11282849e-01 -8.83141696e-01 7.70412199e-03 -3.88103873e-01 1.10069998e-01 8.88595879e-01 7.33953893e-01 -9.74256635e-01 7.39630818e-01 7.45408487e+00 7.91492879e-01 -1.48286927e+00 -1.08232595e-01 6.64418995e-01 1.55780166e-02 -3.14593732e-01 1.94151819e-01 -8.07661176e-01 4.70275342e-01 1.02361822e+00 -2.58001536e-01 5.86303592e-01 6.43676102e-01 4.27959710e-01 7.19711259e-02 -1.09887409e+00 9.07429993e-01 -4.38584775e-01 -1.65906239e+00 -8.26043561e-02 3.85399222e-01 1.05748069e+00 -1.78760856e-01 5.41534066e-01 2.23822564e-01 3.02328408e-01 -1.31956363e+00 7.10428119e-01 1.14690244e+00 1.13288224e+00 -1.79715991e-01 3.92016560e-01 5.57369232e-01 -1.27250898e+00 -2.08626941e-01 -3.03363055e-01 -3.01697552e-01 6.14737570e-01 5.37588418e-01 -3.74821723e-01 5.06434441e-01 6.07606888e-01 1.00498915e+00 -2.92941779e-01 7.08456874e-01 8.77213180e-02 6.18931055e-01 -5.67499101e-01 9.47222300e-03 3.13461751e-01 -4.48244125e-01 4.89505500e-01 8.10773849e-01 7.68883586e-01 6.38723224e-02 -2.69536264e-02 1.60113239e+00 3.88035744e-01 -2.98537910e-01 -5.01491725e-01 -1.28755614e-01 3.63658637e-01 8.51530552e-01 -3.65358710e-01 1.38800755e-01 -2.97277331e-01 3.85265231e-01 9.89085808e-02 9.93060052e-01 -9.06934917e-01 -5.74892238e-02 7.64944077e-01 6.86529875e-01 3.14935684e-01 -6.02062464e-01 -6.72684491e-01 -8.26843202e-01 1.56409442e-01 -4.39124823e-01 -3.78130138e-01 -1.00769806e+00 -1.42617643e+00 5.51602840e-01 3.99208695e-01 -1.42464280e+00 -2.96140224e-01 -1.02359700e+00 -4.17985678e-01 1.67189622e+00 -1.41299021e+00 -1.22590470e+00 -1.86077043e-01 2.13848680e-01 -5.02077118e-02 -2.92523876e-02 6.30774677e-01 2.10767224e-01 -3.49834800e-01 2.64607556e-02 5.38125157e-01 -3.94092411e-01 4.83675778e-01 -1.22670257e+00 4.97370780e-01 7.84801722e-01 -5.46358705e-01 7.04086363e-01 6.17728114e-01 -8.84392738e-01 -1.52704275e+00 -1.34549725e+00 6.42788053e-01 -5.11659920e-01 8.33820939e-01 -5.09357810e-01 -1.19166744e+00 4.96918708e-01 -2.31474981e-01 2.73657054e-01 3.41032624e-01 -4.55244668e-02 7.23903393e-03 -2.59312630e-01 -6.63054228e-01 7.04090893e-01 9.95487571e-01 -8.79597366e-01 -6.44703954e-02 2.81398028e-01 6.28365457e-01 -2.60905921e-01 -1.20021141e+00 6.57696366e-01 4.31435376e-01 -4.73297417e-01 1.18425071e+00 -4.99032706e-01 6.16706014e-01 -5.90087295e-01 -3.42520744e-01 -1.36579752e+00 -6.38747215e-01 -7.31726706e-01 -1.30775809e-01 9.95652378e-01 5.34968078e-01 -7.61691451e-01 6.36384785e-01 7.97596514e-01 3.97622027e-02 -7.93412507e-01 -8.27738464e-01 -9.46256697e-01 6.20011628e-01 -5.24429262e-01 9.73787189e-01 7.94330180e-01 -1.95996672e-01 1.46857426e-01 -4.65542048e-01 5.98118246e-01 5.34139931e-01 2.64953703e-01 4.41822678e-01 -1.15841794e+00 -3.51044148e-01 -4.15391207e-01 -1.99451149e-01 -1.52880752e+00 2.22436830e-01 -6.83133185e-01 1.69331536e-01 -1.42436981e+00 -1.42722383e-01 -8.59665573e-01 -2.59825915e-01 1.17235579e-01 -8.80106911e-02 -1.28271997e-01 4.05715257e-02 3.51509243e-01 -1.38021354e-02 7.92525113e-01 1.47969365e+00 -4.01963256e-02 -3.45725000e-01 -6.19805530e-02 -4.96111095e-01 4.66937989e-01 4.50248778e-01 -2.31695116e-01 -6.07389927e-01 -4.23882008e-01 3.77042592e-01 5.08506000e-01 6.09701812e-01 -1.14732730e+00 5.44773996e-01 -6.38981938e-01 3.92208695e-01 -6.61930025e-01 5.58073521e-01 -7.08235443e-01 6.50541127e-01 5.04224114e-02 -3.11440647e-01 1.62812099e-01 5.09291649e-01 5.09710670e-01 -4.16630059e-01 4.64388788e-01 5.73636651e-01 2.14629889e-01 -6.18167639e-01 4.24421787e-01 -3.68346274e-01 -2.61271000e-01 8.73724699e-01 5.78624383e-02 -3.62654448e-01 -4.74536479e-01 -5.91880977e-01 2.22266465e-01 4.91410881e-01 1.69236764e-01 4.12454903e-01 -1.62958086e+00 -4.43584919e-01 5.86627662e-01 -1.19225867e-01 2.41581023e-01 6.35590374e-01 7.75054991e-01 -6.75441802e-01 7.61980236e-01 -9.64071378e-02 -8.56601775e-01 -2.62906909e-01 6.01415396e-01 5.87892175e-01 -3.72157902e-01 -6.45917833e-01 3.47866863e-01 8.34470689e-01 -6.25224769e-01 -4.29044694e-01 -6.12084270e-01 2.57667094e-01 -5.20779312e-01 9.04123709e-02 4.04235065e-01 -1.43424362e-01 -8.35215926e-01 -3.49443614e-01 9.51233447e-01 6.40608907e-01 -1.76880434e-01 1.49024618e+00 -1.40422121e-01 -2.06794608e-02 6.25731349e-01 1.30924308e+00 -2.95137137e-01 -1.83719921e+00 -3.99816543e-01 -3.20623904e-01 -3.33782107e-01 5.21960318e-01 -6.58812404e-01 -8.99520397e-01 1.11010432e+00 4.35263872e-01 1.42528415e-01 1.08300471e+00 -2.16066822e-01 5.25552511e-01 3.48490298e-01 3.50303560e-01 -9.71347809e-01 -2.05854625e-01 5.95380962e-01 1.16685414e+00 -6.53509796e-01 1.21563770e-01 -7.18605578e-01 -4.73643124e-01 8.87962520e-01 3.28570306e-01 -3.42165023e-01 1.49166548e+00 7.29532003e-01 -1.92825347e-01 -1.16221778e-01 -6.25059068e-01 4.31749552e-01 5.16783774e-01 4.26586509e-01 7.72581547e-02 1.54432565e-01 3.64598334e-01 6.02990031e-01 -7.68853724e-02 2.92979274e-02 4.23445553e-02 7.95523345e-01 7.25721493e-02 -9.12124932e-01 -4.10360485e-01 2.96424925e-01 1.36507273e-01 -8.10889602e-02 1.88204259e-01 8.18312049e-01 -1.11248262e-01 6.16920292e-01 2.25368783e-01 -4.04144228e-01 5.93889773e-01 -1.86011076e-01 6.54104576e-02 -3.00702393e-01 -1.21986106e-01 7.93625042e-02 -5.11994362e-01 -7.17629373e-01 -1.61269382e-01 -6.74935937e-01 -8.99698138e-01 -3.88408303e-01 -1.27963305e-01 -9.42112356e-02 4.64616507e-01 9.56759155e-01 6.49341881e-01 8.50368202e-01 5.43743730e-01 -1.12290204e+00 -5.15372455e-01 -8.15886915e-01 -9.43170726e-01 2.63186932e-01 5.20094991e-01 -1.11772513e+00 -3.11246425e-01 1.47406030e-02]	[6.6085100173950195, 3.403987407684326]
f558dada-8845-43c7-83fe-448ba8b78a09	listening-to-chaotic-whispers-a-deep-learning	1712.02136	null	http://arxiv.org/abs/1712.02136v3	http://arxiv.org/pdf/1712.02136v3.pdf	Listening to Chaotic Whispers: A Deep Learning Framework for News-oriented Stock Trend Prediction	Stock trend prediction plays a critical role in seeking maximized profit from stock investment. However, precise trend prediction is very difficult since the highly volatile and non-stationary nature of stock market. Exploding information on Internet together with advancing development of natural language processing and text mining techniques have enable investors to unveil market trends and volatility from online content. Unfortunately, the quality, trustworthiness and comprehensiveness of online content related to stock market varies drastically, and a large portion consists of the low-quality news, comments, or even rumors. To address this challenge, we imitate the learning process of human beings facing such chaotic online news, driven by three principles: sequential content dependency, diverse influence, and effective and efficient learning. In this paper, to capture the first two principles, we designed a Hybrid Attention Networks to predict the stock trend based on the sequence of recent related news. Moreover, we apply the self-paced learning mechanism to imitate the third principle. Extensive experiments on real-world stock market data demonstrate the effectiveness of our approach.	['Tie-Yan Liu', 'Weiqing Liu', 'Ziniu Hu', 'Xuanzhe Liu', 'Jiang Bian']	2017-12-06	null	null	null	null	['stock-trend-prediction']	['time-series']	[-9.54227924e-01 -5.70160925e-01 -4.71166074e-01 -3.87764499e-02 2.28046209e-01 -5.07531524e-01 5.37496030e-01 1.32213548e-01 -2.83163518e-01 7.59367466e-01 5.76813579e-01 -1.94670871e-01 1.37328357e-01 -9.60871696e-01 -4.26571101e-01 -3.84300768e-01 -2.43984297e-01 8.86222944e-02 5.57628512e-01 -5.35639763e-01 7.43944705e-01 1.48495615e-01 -1.42952406e+00 -1.90607786e-01 7.95661867e-01 1.39415836e+00 1.74943298e-01 1.43331960e-01 -8.94336581e-01 1.59060836e+00 -5.15766799e-01 -6.90952659e-01 2.06568792e-01 -4.13199663e-01 -2.69486010e-01 -5.90129681e-02 -2.85481662e-01 -6.47310078e-01 -3.98186475e-01 1.44832170e+00 1.95753619e-01 -2.43997216e-01 4.17323947e-01 -1.15681100e+00 -9.46534336e-01 1.10580349e+00 -8.56578350e-01 8.68773997e-01 -8.76889974e-02 2.45694891e-02 1.30862081e+00 -4.43639785e-01 6.09376013e-01 7.35286951e-01 4.95593876e-01 3.07866156e-01 -4.63331699e-01 -9.69408929e-01 5.94359219e-01 1.32336006e-01 -7.02707112e-01 -2.85305083e-02 1.11165369e+00 -5.64534009e-01 4.20418471e-01 -1.07352272e-01 1.30304146e+00 1.00472963e+00 7.03679264e-01 9.56772447e-01 9.02961314e-01 3.02191168e-01 2.49292403e-01 4.71822768e-01 1.88097149e-01 3.72157425e-01 5.01217067e-01 -4.74574156e-02 -8.08687866e-01 -1.27638355e-01 6.29595697e-01 4.25725698e-01 -3.12718838e-01 4.46340628e-02 -1.10739422e+00 8.62820029e-01 3.39560032e-01 4.01539773e-01 -7.88096011e-01 -1.65618569e-01 7.04142749e-01 8.37816298e-01 9.80187535e-01 3.27800751e-01 -6.87970161e-01 -4.62155044e-01 -8.22144687e-01 3.44082236e-01 1.09345675e+00 6.51735961e-01 3.49924564e-01 4.72231537e-01 3.10808599e-01 3.73364210e-01 4.09963220e-01 5.38670540e-01 1.28566241e+00 -3.31988007e-01 3.53607744e-01 6.35685027e-01 2.91282117e-01 -1.41987586e+00 -4.37770456e-01 -8.74168754e-01 -1.00744629e+00 1.34184295e-02 1.66815072e-01 -4.62021291e-01 -3.03479970e-01 1.34360433e+00 7.72625729e-02 4.18277048e-02 -3.67265791e-02 9.06967580e-01 4.17326957e-01 1.08613396e+00 -2.45604932e-01 -8.06065023e-01 1.11090624e+00 -8.46270442e-01 -1.14683938e+00 -3.14689726e-02 3.88342887e-02 -6.40250862e-01 6.20962024e-01 4.10048336e-01 -1.00266135e+00 -2.33270958e-01 -1.14058387e+00 4.50640053e-01 -2.48900250e-01 -5.18311679e-01 5.36111355e-01 3.09442282e-01 -7.06350207e-01 6.93292558e-01 -4.52894092e-01 3.31176758e-01 3.75900060e-01 -2.43478455e-02 2.96584159e-01 5.34059584e-01 -1.55451453e+00 5.94963551e-01 5.57496905e-01 1.34320138e-02 -1.44264549e-01 -7.21756220e-01 -3.03651303e-01 1.55628383e-01 5.26460946e-01 -3.75478268e-01 1.45795524e+00 -1.31787097e+00 -1.73927331e+00 1.68653652e-01 2.93952376e-01 -6.65156960e-01 7.90601194e-01 -2.40289867e-01 -5.34776807e-01 2.41496470e-02 -1.09704636e-01 -1.75985366e-01 1.02288342e+00 -5.31667531e-01 -9.64911163e-01 -1.66681141e-01 -4.09654289e-01 1.24189042e-01 -7.26588488e-01 6.57853857e-02 1.08935021e-01 -8.42776597e-01 4.64038663e-02 -4.00071204e-01 2.61437893e-02 -2.95186132e-01 1.94657035e-02 -3.88893068e-01 6.93311989e-01 -7.31720269e-01 1.50648141e+00 -1.96450543e+00 -3.53205264e-01 1.91102833e-01 4.92247254e-01 2.42918301e-02 2.66072571e-01 5.47217011e-01 1.83606565e-01 2.44654417e-01 2.08577827e-01 2.25155711e-01 -1.43807335e-03 -2.01997325e-01 -1.12412345e+00 2.28054479e-01 -3.30748642e-03 9.81322527e-01 -9.34280992e-01 -1.71865314e-01 -4.81770277e-01 -9.09080654e-02 -2.37492010e-01 2.81886250e-01 -5.04261672e-01 4.54986282e-02 -9.92236137e-01 6.35500908e-01 2.71213144e-01 -6.52733982e-01 -2.43851393e-01 2.39984915e-01 -5.86433291e-01 3.69807780e-01 -9.49028730e-01 8.59137356e-01 -2.51772944e-02 6.39489055e-01 -2.86018431e-01 -7.65236855e-01 1.18789840e+00 2.02961504e-01 5.59661627e-01 -9.13431048e-01 2.42078930e-01 4.24778640e-01 -1.57399978e-02 -4.03651983e-01 6.82756543e-01 -3.68688971e-01 1.32288858e-01 1.01087356e+00 -3.20900083e-01 1.59905955e-01 2.34537140e-01 3.00847292e-01 7.36684740e-01 -2.53525645e-01 5.27695537e-01 -1.93697333e-01 2.70044297e-01 3.74027751e-02 6.37184858e-01 3.05779696e-01 -3.19384873e-01 1.56529695e-01 1.05959702e+00 -9.06534851e-01 -1.03836834e+00 -4.57451433e-01 2.38210529e-01 7.82401323e-01 1.18698537e-01 -1.65109292e-01 -3.17550451e-01 -4.19167221e-01 6.22837022e-02 5.69916010e-01 -5.15998781e-01 -1.58844739e-02 -3.22717190e-01 -9.76261973e-01 -1.01763189e-01 2.70489663e-01 7.07085073e-01 -1.36589348e+00 -6.03187442e-01 6.40106797e-01 4.33539823e-02 -7.20857143e-01 -6.38580620e-01 -2.94144958e-01 -7.63347149e-01 -8.58490467e-01 -9.68744695e-01 -3.53388190e-01 2.39531860e-01 3.10295403e-01 1.04388475e+00 7.01497942e-02 3.47741693e-01 1.09390263e-03 -4.24043030e-01 -9.57220972e-01 -3.16140950e-01 1.90215424e-01 1.33632123e-01 2.91072369e-01 2.46688202e-01 -5.81180453e-01 -5.87197781e-01 -2.82117985e-02 -8.97749722e-01 -1.54857993e-01 5.42793334e-01 7.92294204e-01 5.57179525e-02 8.31294298e-01 1.06602728e+00 -8.52252007e-01 1.17149353e+00 -9.78419900e-01 -1.07086337e+00 -6.15624003e-02 -1.26476216e+00 -6.80487528e-02 6.60393894e-01 -6.30394399e-01 -1.20874763e+00 -5.06392300e-01 7.56672546e-02 -1.69458732e-01 6.30146623e-01 1.27111101e+00 4.70221609e-01 4.76269603e-01 6.55052364e-02 7.63170362e-01 3.17180842e-01 -1.71273693e-01 -8.16414505e-02 6.82681739e-01 2.95110524e-01 1.35601580e-01 9.45262492e-01 3.95969689e-01 -5.69597900e-01 -4.63131607e-01 -1.03164423e+00 -2.52786100e-01 -2.86988337e-02 -3.60685170e-01 2.77583450e-01 -9.69058156e-01 -8.66930842e-01 1.02852249e+00 -9.35816467e-01 8.69514197e-02 -1.70345455e-01 4.58793521e-01 -1.86669260e-01 2.58090585e-01 -9.08794463e-01 -1.15724409e+00 -7.14754522e-01 -4.52411860e-01 1.60395682e-01 4.93304133e-01 -1.52368350e-02 -9.64630008e-01 3.66530329e-01 -7.53097460e-02 8.28968406e-01 -1.18818678e-01 7.69452572e-01 -1.14151859e+00 -7.74281323e-01 -6.45321846e-01 -2.37144217e-01 1.62404865e-01 2.39888310e-01 1.37467235e-01 -4.13345277e-01 1.36595964e-01 5.99619985e-01 -2.44737953e-01 8.65069270e-01 2.51208842e-01 4.66447294e-01 -8.13704073e-01 2.30016068e-01 3.83775905e-02 1.13224912e+00 3.70119989e-01 3.17117006e-01 6.45477355e-01 4.01094586e-01 7.79320478e-01 4.25602108e-01 9.62556064e-01 5.47981203e-01 -1.31672770e-01 3.37456912e-01 6.28363132e-01 6.66495860e-01 -5.32170296e-01 5.12272060e-01 1.54748178e+00 -3.64689752e-02 -2.13919446e-01 -6.90856636e-01 2.63920665e-01 -1.91945517e+00 -1.45202470e+00 1.71891138e-01 1.68082440e+00 9.63980615e-01 8.66434216e-01 3.85454386e-01 -8.62995908e-02 6.64121747e-01 4.68579352e-01 -8.63289833e-01 3.18774313e-01 -3.51394325e-01 -7.93099284e-01 5.12225032e-01 -1.08918317e-01 -7.49783397e-01 6.48646057e-01 5.74304008e+00 7.19564974e-01 -1.51230669e+00 -2.83308625e-01 8.68055224e-01 -5.78384027e-02 -5.96289277e-01 -5.63811779e-01 -8.92533481e-01 1.16462898e+00 8.96244645e-01 -9.05402422e-01 2.39788368e-01 1.08781791e+00 2.31549501e-01 1.71268210e-01 -2.65885115e-01 8.16321611e-01 -5.53588616e-03 -1.78767037e+00 1.23539101e-02 -6.03292659e-02 8.21910381e-01 2.40469217e-01 2.62551785e-01 2.48570025e-01 2.56749451e-01 -1.99560925e-01 1.14615691e+00 6.93747759e-01 -1.68709055e-01 -7.86891580e-01 9.78281617e-01 7.62055695e-01 -1.13921893e+00 -3.64448249e-01 -4.40488011e-01 -2.49967396e-01 1.39688179e-01 7.86136210e-01 -2.40667775e-01 2.99609095e-01 7.56564856e-01 1.21488094e+00 -1.17111512e-01 1.01205909e+00 -7.67942816e-02 7.76911736e-01 -1.93834186e-01 -8.32103610e-01 4.19986755e-01 -4.73072946e-01 5.20947874e-01 6.57130718e-01 4.30648118e-01 2.17384368e-01 -1.57619819e-01 8.01551521e-01 -3.06832522e-01 3.91930640e-01 -4.27662253e-01 -6.68922305e-01 3.37684810e-01 1.02462864e+00 -1.03329778e+00 -4.01738435e-01 -7.45543122e-01 4.95103717e-01 2.92122513e-01 1.00977480e-01 -3.75209153e-01 -2.36670122e-01 3.98830831e-01 2.29395881e-01 3.54511470e-01 -1.56256661e-01 -2.84108698e-01 -1.47741961e+00 2.26229072e-01 -7.78408468e-01 2.79552341e-01 -5.03285408e-01 -1.78403676e+00 7.43177176e-01 -3.95384878e-01 -1.37247860e+00 -1.96786299e-01 -3.51041496e-01 -1.02496207e+00 5.18258750e-01 -1.91851830e+00 -3.86578858e-01 1.89648256e-01 2.40360603e-01 7.51502573e-01 -6.32651627e-01 1.81595404e-02 -4.20393376e-03 -5.93136013e-01 -1.01636469e-01 3.18606317e-01 1.38101667e-01 3.87378812e-01 -1.14411557e+00 8.13976049e-01 7.08892465e-01 -1.17528550e-02 4.14968312e-01 7.25105882e-01 -1.11179578e+00 -1.27104688e+00 -8.35745037e-01 9.06277776e-01 -7.49246823e-03 1.68044579e+00 9.02415738e-02 -1.22814918e+00 4.30542111e-01 3.36441398e-01 -4.32135314e-01 3.94424587e-01 -1.56214446e-01 -3.84664565e-01 -2.38526687e-01 -5.08578181e-01 6.70484126e-01 5.34352481e-01 -1.13912798e-01 -7.95536399e-01 2.00318798e-01 1.02790380e+00 -7.75375292e-02 -4.51369584e-01 -1.08200721e-01 4.82137293e-01 -9.47614968e-01 3.16609621e-01 -3.98616761e-01 6.39860749e-01 9.98220146e-02 4.06787515e-01 -1.45348203e+00 -3.17458808e-01 -1.09684801e+00 -3.90999138e-01 1.06510854e+00 3.37459326e-01 -1.05297244e+00 9.20059085e-01 6.70038164e-01 2.18480617e-01 -7.34084427e-01 -5.66966772e-01 -5.81779957e-01 -8.77040997e-02 -7.20918644e-03 8.32607925e-01 1.05824637e+00 3.63534063e-01 2.83677399e-01 -8.72291088e-01 -2.29998142e-01 4.56583202e-01 5.59946179e-01 4.15719807e-01 -1.51149905e+00 -9.11473762e-03 -9.87616003e-01 -1.37336746e-01 -1.11684895e+00 9.50955003e-02 -3.30939561e-01 -2.78043896e-01 -9.50245261e-01 9.15681347e-02 -2.48126965e-02 -5.61181605e-01 -1.35915294e-01 -2.13676423e-01 -3.40656281e-01 4.23000008e-02 8.84424984e-01 -6.11779094e-01 8.74052227e-01 1.23384523e+00 -5.81325591e-02 -2.34802037e-01 4.19448793e-01 -7.50945687e-01 7.60393679e-01 9.93083179e-01 -3.83012146e-01 -3.97507250e-01 9.73933414e-02 1.15122354e+00 1.91875115e-01 -1.60408989e-01 -4.39163089e-01 7.23972559e-01 -3.24999720e-01 1.54793113e-01 -8.91146839e-01 -1.37580663e-01 -8.47627580e-01 -9.09070820e-02 7.74936676e-01 -3.11191142e-01 4.52181846e-01 -2.07295612e-01 8.83612633e-01 -5.82455277e-01 -1.80826008e-01 2.86535412e-01 -4.01042819e-01 -7.57236540e-01 6.30064487e-01 -7.05329597e-01 7.38167390e-02 6.95355833e-01 9.99073908e-02 -4.44029838e-01 -8.55164707e-01 6.87175319e-02 4.62393105e-01 4.02534716e-02 6.52557254e-01 6.09136283e-01 -1.16284156e+00 -8.18756878e-01 2.36013591e-01 -2.59945124e-01 -2.61079460e-01 1.73248872e-01 7.30470896e-01 -6.03623092e-01 3.19393545e-01 -3.49043518e-01 2.28609443e-01 -5.58521628e-01 7.89471030e-01 9.47776809e-02 -3.09708506e-01 -6.83455408e-01 4.84773099e-01 -2.31363907e-01 2.35508770e-01 2.75139689e-01 -2.94986576e-01 -6.20067954e-01 7.84549475e-01 9.90081906e-01 3.34273487e-01 -3.89430702e-01 -2.28822991e-01 2.87682414e-01 2.82444149e-01 -6.07186198e-01 -2.96395812e-02 1.59808505e+00 -4.69636559e-01 -3.08104992e-01 9.71330106e-01 8.44085097e-01 -3.26408632e-02 -1.28824866e+00 -6.35127008e-01 5.41197240e-01 -2.29945123e-01 2.65395939e-01 -3.40680271e-01 -1.29159832e+00 3.24540973e-01 1.55503256e-03 1.12366390e+00 7.76510537e-01 -3.52991849e-01 1.25342810e+00 5.99458218e-01 1.79922745e-01 -1.21780860e+00 2.50570029e-01 5.49633026e-01 6.68627977e-01 -1.30715346e+00 4.19523232e-02 8.34584311e-02 -9.26293254e-01 1.20548129e+00 5.17491698e-01 -1.48598522e-01 1.19185138e+00 1.43334031e-01 4.41056877e-01 -2.11655408e-01 -1.20535278e+00 2.35633969e-01 1.40732855e-01 -1.62925377e-01 1.30970612e-01 -3.59257966e-01 -2.29862243e-01 8.18877459e-01 -4.22628015e-01 8.74242932e-02 8.40804398e-01 8.57481062e-01 -9.12103772e-01 -4.89375263e-01 -1.00879751e-01 6.79245114e-01 -1.01139963e+00 -1.31733909e-01 -2.59663552e-01 5.11764109e-01 -6.54187739e-01 4.65737820e-01 2.48253793e-01 -3.72106433e-01 4.85311262e-02 -5.66419251e-02 -5.37171006e-01 -1.08620137e-01 -4.92428690e-01 2.51734197e-01 -3.88203382e-01 -1.72615051e-01 -2.03023359e-01 -8.65018606e-01 -1.07038140e+00 -5.04037023e-01 -2.11162969e-01 4.64630127e-01 6.88881040e-01 9.05960023e-01 1.91458449e-01 2.48133391e-01 1.28404260e+00 -3.24585497e-01 -1.14084029e+00 -8.06754112e-01 -1.02025270e+00 -2.99712624e-02 6.17122531e-01 -3.16578150e-01 -6.06687665e-01 3.73300537e-02]	[4.411365985870361, 4.271608829498291]
60152d69-b4a7-4d93-8d59-a4fe5876776f	multimodal-fusion-with-deep-neural-networks	1907.03196	null	https://arxiv.org/abs/1907.03196v1	https://arxiv.org/pdf/1907.03196v1.pdf	Multimodal Fusion with Deep Neural Networks for Audio-Video Emotion Recognition	This paper presents a novel deep neural network (DNN) for multimodal fusion of audio, video and text modalities for emotion recognition. The proposed DNN architecture has independent and shared layers which aim to learn the representation for each modality, as well as the best combined representation to achieve the best prediction. Experimental results on the AVEC Sentiment Analysis in the Wild dataset indicate that the proposed DNN can achieve a higher level of Concordance Correlation Coefficient (CCC) than other state-of-the-art systems that perform early fusion of modalities at feature-level (i.e., concatenation) and late fusion at score-level (i.e., weighted average) fusion. The proposed DNN has achieved CCCs of 0.606, 0.534, and 0.170 on the development partition of the dataset for predicting arousal, valence and liking, respectively.	['Marco Pedersoli', 'Patrick Cardinal', 'Mohammed Senoussaoui', 'Juan D. S. Ortega', 'Eric Granger', 'Alessandro L. Koerich']	2019-07-06	null	null	null	null	['video-emotion-recognition']	['computer-vision']	[-7.30530769e-02 -3.81500661e-01 5.61759770e-02 -7.93375134e-01 -6.21946812e-01 -2.52445430e-01 4.01126236e-01 2.13534534e-01 -4.05239135e-01 5.69660008e-01 2.19369248e-01 4.25734669e-01 -1.38245121e-01 -4.00063843e-01 -2.26859361e-01 -7.49626279e-01 9.69109908e-02 -1.40981898e-01 -4.96977389e-01 -3.95638883e-01 -1.65827319e-01 2.77920306e-01 -1.87265277e+00 6.54324532e-01 6.31055057e-01 1.79300237e+00 -3.55216235e-01 6.19317591e-01 1.04699410e-01 7.73698211e-01 -6.15073204e-01 -7.37573981e-01 -1.14493281e-01 -3.58818412e-01 -3.49612296e-01 -3.07174802e-01 2.64803678e-01 -2.51022279e-01 -1.78405672e-01 9.31802273e-01 8.55524361e-01 3.80112767e-01 5.37506044e-01 -1.41838825e+00 -3.56338292e-01 8.06493282e-01 -5.94283342e-01 -3.52820903e-02 2.95073390e-01 -2.83627689e-01 1.07421303e+00 -9.80354786e-01 2.93610156e-01 1.06211996e+00 4.27042693e-01 6.55144930e-01 -9.73292470e-01 -9.50376689e-01 -5.45018241e-02 3.09873760e-01 -1.33947337e+00 -5.01636565e-01 9.59474266e-01 -3.79815623e-02 1.01901102e+00 2.82555133e-01 6.17521107e-01 1.29782450e+00 2.53295392e-01 9.45595086e-01 9.54004943e-01 -1.32457882e-01 2.14055419e-01 2.77265966e-01 2.33022243e-01 1.75696224e-01 -9.30170417e-02 -2.30076090e-01 -1.17288792e+00 4.70375270e-02 2.78601617e-01 -5.32627106e-03 6.24035532e-03 2.47980580e-01 -7.86853015e-01 7.60135651e-01 3.11907053e-01 5.26842177e-01 -8.61885548e-01 -4.30600643e-02 7.55601764e-01 3.40503275e-01 5.71164668e-01 1.40229613e-01 -6.02062345e-01 -5.99649370e-01 -8.62301767e-01 7.39107430e-02 6.91925168e-01 2.86098212e-01 3.66680562e-01 3.28852117e-01 -2.11625755e-01 1.16819155e+00 3.48356396e-01 3.59878600e-01 4.74346817e-01 -7.52235949e-01 1.65310830e-01 5.98819256e-01 -2.73197919e-01 -1.06274760e+00 -5.92884362e-01 -4.73423839e-01 -1.15668082e+00 8.92388597e-02 -1.63804203e-01 -7.73116231e-01 -7.89693892e-01 2.20822716e+00 7.55043998e-02 2.19179094e-01 7.14866221e-01 9.26310897e-01 1.47510314e+00 8.08063984e-01 2.65314192e-01 -1.45224661e-01 1.20989251e+00 -6.82432413e-01 -1.06373549e+00 -5.96729219e-02 3.45094144e-01 -6.58969104e-01 4.48249459e-01 8.09563935e-01 -1.13312149e+00 -6.41118884e-01 -1.22373736e+00 1.73439771e-01 -3.99867356e-01 4.07267869e-01 6.64473534e-01 7.75016904e-01 -1.11402094e+00 4.46233153e-01 -4.55612093e-01 -2.51876086e-01 2.70352930e-01 5.28280199e-01 -5.45700312e-01 1.90847188e-01 -1.49061537e+00 8.18221986e-01 5.21392703e-01 4.12206829e-01 -7.38501966e-01 -5.42835236e-01 -7.46386349e-01 2.98245847e-01 -3.11419129e-01 -4.35802132e-01 1.10978401e+00 -1.47213554e+00 -1.66103590e+00 4.92207497e-01 1.34367824e-01 -4.20595497e-01 -1.31554335e-01 -4.02785987e-01 -9.24276650e-01 1.61895946e-01 -5.90241075e-01 1.07801795e+00 4.85982776e-01 -1.20085931e+00 -7.17799962e-01 -3.31488967e-01 -4.01333347e-02 6.25908911e-01 -9.70522940e-01 1.27837464e-01 -1.64159581e-01 -3.23245257e-01 -1.43056706e-01 -6.06430829e-01 9.92357582e-02 -4.31212008e-01 -1.12111822e-01 -4.51925427e-01 8.76098692e-01 -8.39350820e-01 1.37329280e+00 -2.49274182e+00 3.52403194e-01 3.65787059e-01 1.27904817e-01 3.61087382e-01 -2.64334738e-01 4.02599871e-01 -4.36469644e-01 -1.67011663e-01 2.45713621e-01 -6.29034936e-01 9.86275449e-02 1.13526791e-01 3.27685565e-01 2.39633635e-01 2.17545807e-01 5.85215092e-01 -5.42927742e-01 -8.31374153e-02 3.32629502e-01 9.25928414e-01 -3.68973643e-01 3.75237584e-01 1.21978208e-01 1.00746453e-01 1.16262985e-02 6.43030643e-01 7.98611224e-01 2.73478180e-01 2.86931604e-01 -5.55695891e-01 2.19581053e-02 -2.05174740e-02 -1.30813444e+00 1.87690842e+00 -4.70921099e-01 7.37211049e-01 2.22029373e-01 -7.20701158e-01 1.36100006e+00 7.18602121e-01 5.93246341e-01 -7.25648284e-01 7.35450506e-01 1.04244567e-01 4.62590270e-02 -3.12120914e-01 6.62836611e-01 -2.81628728e-01 -4.32914853e-01 2.46118680e-01 6.98019087e-01 4.63930279e-01 -1.76746119e-02 1.42664209e-01 6.64517939e-01 -3.11928749e-01 -2.21385434e-02 1.63638845e-01 6.53555334e-01 -6.73590302e-01 5.72038651e-01 2.66118854e-01 -2.19558418e-01 3.21618885e-01 6.50889754e-01 -2.32671961e-01 -7.17308462e-01 -7.16963351e-01 -2.92559713e-02 1.27547204e+00 -7.48301446e-02 -4.14582163e-01 -7.65415847e-01 -1.99278787e-01 -3.39739829e-01 6.40913606e-01 -6.52000844e-01 -6.23619497e-01 2.21310586e-01 -6.95535898e-01 7.82676399e-01 8.36208880e-01 6.95967257e-01 -1.13410079e+00 -3.88872147e-01 8.94665271e-02 -4.04088169e-01 -1.09963834e+00 2.67716259e-01 3.75398278e-01 -5.74077249e-01 -5.34987688e-01 -5.12963951e-01 -6.68618321e-01 1.83056980e-01 -2.26853609e-01 9.21576738e-01 -3.60653222e-01 3.15478444e-01 3.39907736e-01 -5.54984987e-01 -5.58478892e-01 4.40694019e-02 -1.65324554e-01 8.50267261e-02 6.32829726e-01 7.20686078e-01 -5.85674107e-01 -4.65740472e-01 -6.35981858e-02 -1.00235105e+00 -9.18751508e-02 4.42425042e-01 8.68739843e-01 3.19402397e-01 1.39678508e-01 8.27354372e-01 -2.22830907e-01 1.22501814e+00 -5.98593235e-01 1.15485219e-02 1.89758435e-01 -2.93328762e-01 -4.57070827e-01 3.32838982e-01 -4.39471543e-01 -1.32241583e+00 3.38596962e-02 -2.62240261e-01 -5.96315980e-01 -4.55922604e-01 9.25909281e-01 -1.50485933e-01 1.93247929e-01 3.31277400e-01 -1.14511274e-01 -1.29104570e-01 -3.51388574e-01 3.11914176e-01 1.08245730e+00 5.53442717e-01 -2.78342217e-01 -1.34374037e-01 -6.43892139e-02 -1.18219346e-01 -4.85262841e-01 -6.90894902e-01 -3.74061972e-01 -4.70679432e-01 -5.46883464e-01 1.14802122e+00 -1.33668506e+00 -1.01833856e+00 8.82558763e-01 -1.04281521e+00 2.92994261e-01 -9.92221311e-02 6.89185917e-01 -1.35849535e-01 -2.47418255e-01 -6.47163451e-01 -1.08002615e+00 -7.68320262e-01 -8.08033645e-01 8.85420024e-01 7.03903079e-01 -4.18711901e-01 -8.98071349e-01 1.03799067e-01 3.38937342e-01 4.42327291e-01 6.03605151e-01 6.07084632e-01 -1.17435825e+00 4.61077660e-01 -5.59455395e-01 -1.05595045e-01 1.05485904e+00 -1.07426941e-01 1.85795024e-01 -1.45791006e+00 1.18561843e-02 -3.91750634e-02 -6.63219213e-01 7.93340385e-01 4.53705341e-01 1.00600588e+00 1.37677819e-01 3.87841612e-01 3.41336340e-01 1.38331807e+00 4.95049775e-01 6.67402983e-01 -2.00319700e-02 5.01717687e-01 3.77877176e-01 4.25075978e-01 8.64816427e-01 3.45980853e-01 2.34610438e-01 6.98219240e-01 -2.48594508e-01 3.25336099e-01 3.63696516e-01 4.35802132e-01 9.57975388e-01 -1.02438163e-02 -4.92288351e-01 -6.92848146e-01 3.95323515e-01 -1.91612768e+00 -8.13854516e-01 1.51889604e-02 1.69271147e+00 3.59705001e-01 -4.11700755e-02 1.68995664e-01 4.65869546e-01 6.69166625e-01 8.91677812e-02 -3.48988920e-01 -1.17810559e+00 -4.27040756e-01 3.85250628e-01 -2.26002797e-01 8.48360732e-02 -1.11864531e+00 5.59332967e-01 5.65301323e+00 7.51517594e-01 -1.26576221e+00 6.40384108e-02 7.55878866e-01 -4.33404565e-01 -7.99093116e-03 -7.56635964e-01 -4.47549939e-01 3.32047611e-01 1.52623475e+00 -7.28909392e-03 4.68528271e-01 5.61633527e-01 -5.29183336e-02 -9.92106199e-02 -9.32821155e-01 1.43823147e+00 3.44492793e-01 -8.35752249e-01 -2.98829339e-02 -3.29151809e-01 8.45698476e-01 -1.04292044e-02 3.91204566e-01 4.74518180e-01 -1.31992742e-01 -1.00453079e+00 4.86246496e-01 7.83767939e-01 4.76069748e-01 -1.54491389e+00 1.55093193e+00 1.10821068e-01 -9.54790354e-01 -3.35347444e-01 8.62376839e-02 -1.44956380e-01 9.50260833e-02 4.52704638e-01 -2.84357578e-01 7.80000687e-01 7.89148390e-01 8.80550921e-01 -3.20751011e-01 6.56455874e-01 5.62967360e-02 4.17852938e-01 -2.67838597e-01 -2.75805473e-01 3.47100347e-01 -1.62496995e-02 1.22666270e-01 1.34566844e+00 5.26418567e-01 1.73443705e-01 -1.88942596e-01 2.68312395e-01 -4.03665006e-01 3.01140636e-01 -2.73778528e-01 -1.71052635e-01 2.75712669e-01 1.78803289e+00 -3.47521812e-01 -4.43462223e-01 -2.91613549e-01 6.86888695e-01 7.77605223e-03 1.56267464e-01 -9.25600708e-01 -6.19638681e-01 9.48195457e-01 -8.81761432e-01 4.52894837e-01 1.97529957e-01 -4.44817007e-01 -8.33236933e-01 -1.49076059e-01 -8.87592614e-01 5.34707248e-01 -1.04059601e+00 -1.34343767e+00 9.41946626e-01 -1.77372783e-01 -1.26750338e+00 -1.98872209e-01 -6.78725541e-01 -5.83135605e-01 9.91982996e-01 -1.04358494e+00 -1.00835764e+00 -2.93042719e-01 8.27833951e-01 3.69433835e-02 -3.63809347e-01 1.03942597e+00 7.12675214e-01 -6.96528614e-01 8.26551735e-01 2.68751830e-02 6.25259802e-02 6.87527299e-01 -9.35534656e-01 -5.38954675e-01 5.77041090e-01 -1.57863677e-01 3.63931686e-01 5.19419014e-01 -4.69814241e-02 -1.28452933e+00 -8.13835561e-01 8.52285981e-01 -1.81112997e-02 3.18787664e-01 -2.92941928e-01 -5.56737065e-01 1.04611896e-01 7.23398149e-01 -2.88240999e-01 1.31649792e+00 3.81362647e-01 -2.83938438e-01 -7.31498778e-01 -1.48469019e+00 2.20592484e-01 1.95315823e-01 -5.64805686e-01 -1.53111473e-01 -1.10082924e-01 5.22258937e-01 -3.93044323e-01 -1.29541671e+00 5.91194808e-01 1.03050601e+00 -1.19673967e+00 5.68953872e-01 -3.59800786e-01 7.77126014e-01 1.39414638e-01 -7.12431371e-01 -1.45935249e+00 -2.01224685e-01 -1.41049638e-01 -3.60638320e-01 1.66751361e+00 4.63887364e-01 -1.48598477e-01 3.39662075e-01 5.44075072e-01 -1.48239136e-01 -1.10804570e+00 -1.01168120e+00 -1.67466804e-01 -3.62146318e-01 -7.70731807e-01 6.06024504e-01 8.84561718e-01 1.05456948e-01 6.40286446e-01 -7.07472444e-01 -3.99815179e-02 2.10503832e-01 -3.60376179e-01 2.70130128e-01 -1.42158556e+00 3.01121354e-01 -4.93576914e-01 -6.85300708e-01 -2.60231525e-01 1.51464239e-01 -6.14386618e-01 -1.69758826e-01 -1.53721249e+00 2.30619222e-01 2.69721985e-01 -1.21916950e+00 6.14845753e-01 2.81980544e-01 4.77337778e-01 3.26784432e-01 -6.16469026e-01 -7.77844548e-01 6.46656692e-01 8.68131280e-01 2.25628000e-02 -3.62150252e-01 -3.75258625e-01 -1.13187802e+00 6.57616496e-01 1.02968490e+00 -1.58078939e-01 -4.23220456e-01 -3.34257573e-01 4.17799234e-01 1.66318610e-01 1.36481807e-01 -1.30735731e+00 1.21255785e-01 2.23816589e-01 8.37066650e-01 -8.81617665e-01 9.49491680e-01 -9.68374074e-01 1.83623582e-01 1.34116009e-01 -5.10511696e-01 8.96353647e-02 4.71333057e-01 1.98479623e-01 -6.48821294e-01 9.44571123e-02 6.41391933e-01 3.38556051e-01 -7.08897054e-01 1.44885793e-01 -2.97890604e-01 -5.58629334e-01 9.08827603e-01 -5.81405014e-02 -1.26974955e-01 -6.97916746e-01 -9.84847486e-01 1.18680477e-01 -2.95522660e-01 8.41487825e-01 9.06143546e-01 -1.72644389e+00 -7.06600487e-01 1.83782801e-01 6.67126626e-02 -5.20829558e-01 6.81228399e-01 8.01441252e-01 1.24635845e-01 2.43111566e-01 -6.47321343e-01 -5.03471613e-01 -1.62906778e+00 -1.16892517e-01 3.90547514e-01 -2.70639747e-01 1.94550812e-01 1.09181297e+00 -2.91448474e-01 -4.14709240e-01 4.56256419e-01 -1.34328529e-01 -7.77863920e-01 6.94423139e-01 6.01179302e-01 3.90307575e-01 3.67833287e-01 -9.42157149e-01 -5.49048424e-01 3.20728600e-01 8.57667699e-02 -2.97267318e-01 1.73494828e+00 -3.83045673e-02 -2.77053088e-01 7.23884463e-01 1.30873501e+00 -5.72437286e-01 -7.52231061e-01 -7.59831443e-02 -6.44470215e-01 6.87937438e-02 5.23484111e-01 -1.31436038e+00 -1.41104960e+00 9.59380150e-01 1.13951361e+00 1.38520658e-01 1.79223180e+00 -3.04481119e-01 7.94199824e-01 3.59233588e-01 -2.71590978e-01 -1.34307480e+00 6.48548231e-02 6.76882982e-01 6.55936480e-01 -1.01294649e+00 -2.25749910e-01 3.76194805e-01 -1.07716227e+00 1.30882740e+00 8.06454301e-01 8.12417716e-02 8.42379928e-01 1.56602561e-01 1.86953694e-01 -2.41862714e-01 -1.29948187e+00 5.85069880e-02 6.08897150e-01 3.42867911e-01 7.58367002e-01 2.85522580e-01 -3.36808860e-01 1.26002634e+00 -1.85630843e-02 -4.57494594e-02 2.25371718e-01 8.10520172e-01 -2.51172811e-01 -7.70391703e-01 -2.13408843e-01 4.91963714e-01 -8.00237477e-01 4.76763546e-02 -5.08483887e-01 3.35401446e-01 4.21261340e-01 1.34850907e+00 2.22092405e-01 -1.20112038e+00 3.89996856e-01 3.21244657e-01 1.61302418e-01 -1.00672282e-01 -1.04842675e+00 2.51576573e-01 3.57946485e-01 -5.50876617e-01 -9.59667325e-01 -4.95923519e-01 -9.86403227e-01 -3.09555411e-01 -2.96724558e-01 4.16205823e-02 1.04386103e+00 8.68521154e-01 4.10118669e-01 8.28910887e-01 8.15957725e-01 -9.96622264e-01 -3.79629657e-02 -1.26915264e+00 -8.47700059e-01 3.48839939e-01 3.84279191e-01 -4.06826138e-01 -3.12848121e-01 -2.27161601e-01]	[13.327082633972168, 5.132423400878906]
c4cf9916-7d26-4b65-8239-1b50b532440a	mining-implicit-relevance-feedback-from-user	2006.07581	null	https://arxiv.org/abs/2006.07581v2	https://arxiv.org/pdf/2006.07581v2.pdf	Mining Implicit Relevance Feedback from User Behavior for Web Question Answering	Training and refreshing a web-scale Question Answering (QA) system for a multi-lingual commercial search engine often requires a huge amount of training examples. One principled idea is to mine implicit relevance feedback from user behavior recorded in search engine logs. All previous works on mining implicit relevance feedback target at relevance of web documents rather than passages. Due to several unique characteristics of QA tasks, the existing user behavior models for web documents cannot be applied to infer passage relevance. In this paper, we make the first study to explore the correlation between user behavior and passage relevance, and propose a novel approach for mining training data for Web QA. We conduct extensive experiments on four test datasets and the results show our approach significantly improves the accuracy of passage ranking without extra human labeled data. In practice, this work has proved effective to substantially reduce the human labeling cost for the QA service in a global commercial search engine, especially for languages with low resources. Our techniques have been deployed in multi-language services.	['Jian Pei', 'Feixiang Cheng', 'Ming Gong', 'Daxin Jiang', 'Shining Bo', 'Linjun Shou']	2020-06-13	null	null	null	null	['passage-ranking']	['natural-language-processing']	[ 7.27551430e-02 7.62378722e-02 -2.39241943e-01 -2.80528843e-01 -1.53248000e+00 -7.19729602e-01 6.16452217e-01 2.81186998e-01 -7.17742980e-01 7.32578516e-01 -4.80072722e-02 -8.65082085e-01 -2.70095080e-01 -6.00415885e-01 -2.91507363e-01 -8.69769230e-02 7.71931782e-02 8.70146930e-01 8.36651862e-01 -7.90046632e-01 5.15362024e-01 -6.53911307e-02 -1.48899090e+00 6.07576489e-01 1.29843128e+00 6.69035017e-01 5.19446552e-01 9.76150155e-01 -4.07582670e-01 1.00681269e+00 -7.33864486e-01 -5.12812674e-01 7.42887780e-02 -4.18103725e-01 -1.56588638e+00 -2.00443208e-01 4.87389147e-01 -5.09174578e-02 4.84263636e-02 9.14278269e-01 5.86277246e-01 2.56331503e-01 5.30678272e-01 -7.54945040e-01 -7.81490505e-01 4.40248847e-01 -1.58128187e-01 7.31248319e-01 7.23602474e-01 -2.28276268e-01 1.36679089e+00 -8.65207791e-01 4.21242177e-01 8.78450751e-01 3.03079724e-01 3.99955094e-01 -9.09374118e-01 -9.77977365e-02 -4.38531190e-02 6.93179786e-01 -1.27793646e+00 -4.38073128e-01 3.80756319e-01 -1.36649072e-01 1.12214029e+00 8.19143414e-01 8.61141905e-02 6.12299204e-01 -2.82168567e-01 7.94531345e-01 9.59414363e-01 -1.16499817e+00 1.63250845e-02 5.74960649e-01 5.79999030e-01 6.99961722e-01 1.66295141e-01 -3.86278778e-01 -3.05219889e-01 -5.83697736e-01 1.02539383e-01 -3.93371820e-01 -1.27256572e-01 4.16414216e-02 -4.86769348e-01 9.68969226e-01 7.35940970e-03 4.58970070e-01 -2.33082727e-01 -4.97399181e-01 2.46602833e-01 6.87041163e-01 5.02123296e-01 7.25816965e-01 -7.83430576e-01 -1.93702757e-01 -5.78808427e-01 2.66042590e-01 1.24167371e+00 8.76593113e-01 9.14132357e-01 -7.88743496e-01 -3.82544547e-01 1.31597686e+00 3.62872362e-01 5.13556957e-01 9.24639940e-01 -9.26350832e-01 5.54162979e-01 7.18138516e-01 4.49069858e-01 -6.86121464e-01 -5.06569929e-02 -1.67226076e-01 9.69667658e-02 -3.31806481e-01 4.34588164e-01 9.82789174e-02 -6.87216997e-01 1.09910345e+00 2.97470182e-01 -6.80258930e-01 -6.23744056e-02 7.79887557e-01 6.44327343e-01 4.75847393e-01 2.90676862e-01 -3.47565919e-01 1.64434969e+00 -1.20459104e+00 -7.66819358e-01 -3.36793661e-01 1.04353976e+00 -1.20063519e+00 1.89898014e+00 3.88567328e-01 -1.04760146e+00 -2.96285778e-01 -5.95720828e-01 -3.35942805e-01 -4.63608474e-01 -5.48518263e-02 6.85946763e-01 6.97510421e-01 -1.02187216e+00 3.01975340e-01 -4.44097042e-01 -5.35690486e-01 -4.23258096e-01 2.13982880e-01 1.62732214e-01 -2.18443647e-01 -1.66002190e+00 9.53701496e-01 7.44905323e-02 -2.95181036e-01 -2.21393630e-01 -1.95377514e-01 -5.81717491e-01 1.44670889e-01 6.27615273e-01 -4.83063817e-01 2.04648232e+00 -8.33807468e-01 -1.35642111e+00 6.85239553e-01 -5.47525942e-01 -2.19229385e-01 1.73281610e-01 -3.90522301e-01 -5.99718153e-01 1.80059269e-01 5.16148269e-01 1.33451642e-02 3.73376608e-01 -9.74858403e-01 -1.07107592e+00 -1.95719540e-01 3.35239112e-01 5.77692211e-01 -6.67747915e-01 2.35765144e-01 -8.95070851e-01 -6.60996288e-02 -2.42117405e-01 -6.01438403e-01 -2.79326797e-01 -7.14713156e-01 1.58130795e-01 -8.83084416e-01 4.65485126e-01 -9.40588951e-01 1.68380177e+00 -1.48012972e+00 -5.33004701e-01 1.71420380e-01 -1.69375867e-01 3.60057533e-01 -1.08267672e-01 6.17462337e-01 4.02179629e-01 4.38868284e-01 1.53144926e-01 8.33356660e-03 -4.91837077e-02 1.60870865e-01 -2.10361123e-01 -2.00673833e-01 -2.75336981e-01 1.03456771e+00 -1.28687453e+00 -9.95303571e-01 -3.79063308e-01 -4.68152948e-02 -3.83085519e-01 2.89431751e-01 -4.49347913e-01 -2.15823770e-01 -7.07848549e-01 6.52934372e-01 9.92681608e-02 -6.90057456e-01 2.40618974e-01 2.73992538e-01 4.25667912e-02 9.69014645e-01 -5.73875368e-01 1.67598259e+00 -7.48669386e-01 4.38203543e-01 -2.07268864e-01 -6.56862497e-01 5.12819350e-01 5.73926568e-01 3.34906846e-01 -1.71471620e+00 -2.83690125e-01 3.84092063e-01 -1.21565104e-01 -9.82000709e-01 8.23435903e-01 3.25216919e-01 1.68456957e-02 7.66941667e-01 -2.26328045e-01 1.67823181e-01 8.72780740e-01 3.71547490e-01 1.25607407e+00 -1.46200269e-01 2.54494607e-01 -3.66486609e-01 1.00814056e+00 5.25365055e-01 -5.84899122e-03 1.03480065e+00 -1.08467177e-01 5.75205870e-02 -4.66371998e-02 -1.67350769e-01 -8.31479430e-01 -7.61560619e-01 3.01416256e-02 1.67865896e+00 1.80600658e-01 -5.63410163e-01 -7.96348393e-01 -1.05220819e+00 -5.01451612e-01 8.09950948e-01 -1.60789400e-01 -5.31990789e-02 -6.27304733e-01 -5.61106741e-01 1.75328270e-01 2.81252023e-02 1.43393010e-01 -1.28170812e+00 -2.05290109e-01 3.02308321e-01 -9.95770276e-01 -1.07166076e+00 -6.40785336e-01 -9.13263410e-02 -9.34268832e-01 -1.17997146e+00 -4.21524346e-01 -1.01566589e+00 4.30901676e-01 4.55394000e-01 1.72385526e+00 6.06347144e-01 -3.36707562e-01 5.66443682e-01 -8.63326192e-01 -1.86195835e-01 -3.16903681e-01 5.67091584e-01 -2.13166311e-01 -6.57587588e-01 1.05574560e+00 -1.32520214e-01 -8.05530250e-01 6.13544047e-01 -8.88557196e-01 -4.96660262e-01 6.67313516e-01 7.84710169e-01 4.31330293e-01 3.59109752e-02 8.54014397e-01 -1.15190709e+00 1.51109302e+00 -6.12330019e-01 -4.66615975e-01 8.46571565e-01 -1.36709607e+00 2.31369019e-01 4.85652417e-01 -2.80211985e-01 -1.25300896e+00 -4.52516079e-01 -3.37000579e-01 5.54933310e-01 -1.30643677e-02 7.66410112e-01 2.58937985e-01 5.39971627e-02 1.15694642e+00 1.87040925e-01 -4.31784928e-01 -8.07075083e-01 3.63320559e-01 1.07703042e+00 1.60510585e-01 -6.04205370e-01 5.36329687e-01 6.03392795e-02 -6.61835015e-01 -7.94134796e-01 -1.10965419e+00 -1.35738516e+00 -3.27184677e-01 6.88525140e-02 3.69377464e-01 -4.32647943e-01 -6.93080604e-01 -2.98871368e-01 -7.40039289e-01 -3.38516384e-01 -1.93056211e-01 1.82348371e-01 -4.35439140e-01 9.53890741e-01 -7.12727845e-01 -9.10213768e-01 -9.12164688e-01 -8.23613465e-01 7.56648719e-01 2.29513139e-01 -3.92709374e-01 -1.12846518e+00 5.33339262e-01 9.49918389e-01 3.88746589e-01 -8.84310901e-01 9.80698943e-01 -7.94936061e-01 -4.53621894e-01 -4.91825193e-01 1.14504471e-02 1.74940005e-01 4.37501967e-01 -4.51576918e-01 -7.76750922e-01 -1.72998875e-01 1.18007652e-01 -6.58837438e-01 3.56167704e-01 -1.88705251e-01 1.06628513e+00 -6.01171851e-01 -5.41437715e-02 -3.48336130e-01 1.26128256e+00 1.58307731e-01 5.34917295e-01 9.24298823e-01 -2.47315597e-02 6.14549756e-01 1.28631532e+00 7.75346681e-02 4.76406753e-01 8.13863218e-01 7.59012671e-03 1.11516424e-01 -4.20814976e-02 -2.73635328e-01 4.27730113e-01 1.00196934e+00 -1.50366172e-01 -4.15903419e-01 -7.15310037e-01 7.89318204e-01 -1.76792753e+00 -1.04479027e+00 -1.06947392e-01 2.45291018e+00 1.35502422e+00 8.47746432e-02 1.72790159e-02 3.55428830e-02 2.54157454e-01 -5.29400587e-01 -1.14012383e-01 -4.10698444e-01 3.20291370e-01 4.51537549e-01 2.90535003e-01 9.29377615e-01 -7.68111825e-01 9.14733708e-01 6.65946817e+00 8.61799657e-01 -5.43302298e-01 3.94383848e-01 3.60545993e-01 1.18295595e-01 -4.86668229e-01 3.53925936e-02 -8.86034012e-01 2.56608218e-01 1.24958849e+00 -4.57802624e-01 4.65504825e-01 1.10132682e+00 1.11280307e-01 -2.13422835e-01 -9.30399597e-01 5.86192250e-01 1.40349418e-01 -8.83829474e-01 1.51043814e-02 -3.06797214e-02 6.81732714e-01 1.74620673e-01 -1.80975832e-02 8.01963210e-01 5.60040772e-01 -7.08876550e-01 9.04171765e-02 2.24007532e-01 5.62823713e-01 -5.49426377e-01 7.29711711e-01 8.70677054e-01 -7.73262799e-01 4.31485288e-02 -4.91853327e-01 7.72135183e-02 3.01489651e-01 3.83237094e-01 -1.40640581e+00 4.33542252e-01 1.08835161e+00 -1.27816603e-01 -9.84860837e-01 1.02871335e+00 -2.42083803e-01 7.24339664e-01 -3.03689241e-01 -4.43041891e-01 1.85716957e-01 -1.26369476e-01 2.02731460e-01 1.17184579e+00 8.54685232e-02 -4.21160981e-02 5.26725538e-02 3.05779904e-01 -3.90713960e-02 6.29843712e-01 -2.77224690e-01 -1.43659860e-01 2.14999691e-01 1.45358658e+00 -5.15126526e-01 -5.77243686e-01 -6.60324037e-01 1.07150400e+00 5.65834224e-01 3.81317794e-01 -3.60397190e-01 -4.80346382e-01 1.95967436e-01 3.01718920e-01 8.83115306e-02 -2.01604739e-01 6.74285963e-02 -1.19499958e+00 3.73305291e-01 -1.12570119e+00 8.79323125e-01 -4.61971223e-01 -1.35388196e+00 7.70846665e-01 -1.03851825e-01 -9.42055702e-01 -8.49186063e-01 -5.02648950e-01 -1.83395728e-01 1.22276938e+00 -1.88565528e+00 -5.37718117e-01 -2.38385081e-01 7.35805154e-01 1.08122051e+00 5.13684601e-02 9.33483779e-01 6.09958172e-01 5.83703332e-02 7.29135394e-01 1.15231022e-01 -1.40196785e-01 8.74053895e-01 -1.35855877e+00 5.37291110e-01 7.26121485e-01 5.83120525e-01 1.12433863e+00 7.08344817e-01 -6.48521662e-01 -1.20026290e+00 -4.86389458e-01 1.62023830e+00 -1.01265669e+00 7.28051186e-01 -7.64026642e-02 -1.12803447e+00 3.89251262e-01 4.43168730e-01 -5.16962290e-01 9.81121898e-01 6.25482559e-01 -1.42418012e-01 7.80516863e-02 -8.13512862e-01 6.57574177e-01 8.21367681e-01 -7.70611584e-01 -9.00064111e-01 9.08792436e-01 7.06909716e-01 -3.37661132e-02 -7.52542377e-01 4.26914424e-01 2.35349193e-01 -4.65812027e-01 1.10337090e+00 -7.15392232e-01 1.33255363e-01 -2.79595733e-01 7.60239288e-02 -9.18924987e-01 -4.27685112e-01 -2.56713778e-01 -5.32988846e-01 9.78012562e-01 1.01955998e+00 -4.66457695e-01 8.19528282e-01 8.88029039e-01 2.88466811e-02 -5.46992540e-01 -6.27457678e-01 -7.46101022e-01 -7.77332708e-02 -2.51184762e-01 1.92115292e-01 6.07641816e-01 3.91002357e-01 8.84831548e-01 -4.49711867e-02 5.09884693e-02 5.14933407e-01 8.79092216e-02 4.51873273e-01 -1.12244201e+00 -5.19226611e-01 -2.57243633e-01 3.66349369e-01 -1.39210522e+00 -2.62200624e-01 -7.38314331e-01 2.05684721e-01 -1.72704065e+00 3.37368488e-01 -6.62360489e-01 -3.64319235e-01 3.20283532e-01 -6.97520077e-01 1.54634520e-01 -4.10989910e-01 4.60223913e-01 -1.10699034e+00 2.08291203e-01 1.09533453e+00 -2.29804382e-01 -2.63896674e-01 3.88153374e-01 -8.70209813e-01 3.61070842e-01 7.81064212e-01 -5.13558567e-01 -9.75687385e-01 -6.08466446e-01 5.96042216e-01 -6.05956838e-02 -1.86293423e-01 -4.34222668e-01 4.95820254e-01 -2.22015545e-01 -1.28921652e-02 -5.27700424e-01 6.45722002e-02 -8.68619502e-01 -5.64482212e-01 2.00391963e-01 -4.80512291e-01 2.22621247e-01 -1.27448425e-01 4.29136753e-01 -3.46071303e-01 -9.38404322e-01 1.51181519e-01 -4.64344144e-01 -1.02606082e+00 2.52025276e-01 -5.22980750e-01 4.62853014e-01 4.70583975e-01 1.10339396e-01 -3.70380610e-01 -5.23044348e-01 -4.34491575e-01 4.97739851e-01 2.38690853e-01 7.49211848e-01 2.80774385e-01 -1.08812666e+00 -4.52713698e-01 -2.41362542e-01 4.42874134e-01 -3.60505909e-01 -1.37317041e-02 5.26807070e-01 -4.69130635e-01 1.05619991e+00 4.12794501e-01 -4.18553114e-01 -1.48703098e+00 5.84330797e-01 1.21250823e-01 -7.28324771e-01 -2.94495016e-01 6.26512229e-01 -1.75284550e-01 -5.10060310e-01 1.39123961e-01 2.16189876e-01 -7.44476020e-01 -1.74625859e-01 9.41168308e-01 2.25537941e-01 7.54083753e-01 -1.87212169e-01 8.43570828e-02 1.18533395e-01 -6.73422158e-01 -5.33148289e-01 7.51401007e-01 -5.07416248e-01 -1.75833836e-01 3.29096407e-01 1.09628391e+00 1.46198481e-01 -4.28470224e-01 -6.10293031e-01 7.42757320e-01 -5.47732472e-01 -1.66698620e-01 -1.12597620e+00 -3.97127211e-01 6.19970798e-01 4.86957759e-01 6.17392004e-01 1.09272254e+00 2.43676558e-01 7.24330306e-01 1.11128235e+00 6.71728194e-01 -1.53532445e+00 1.53930709e-01 6.86180592e-01 6.26085937e-01 -1.48297346e+00 -1.76157787e-01 -2.63280153e-01 -4.99338418e-01 8.13833237e-01 7.45134175e-01 6.11159265e-01 3.83231699e-01 -3.96385163e-01 5.37030935e-01 -4.84809965e-01 -9.15427148e-01 -3.97381485e-01 5.81385911e-01 3.56223255e-01 9.68139470e-01 -1.81247726e-01 -9.43024218e-01 3.74910027e-01 -3.50004047e-01 -8.52684975e-02 1.91350907e-01 1.11683953e+00 -7.91425109e-01 -1.62590647e+00 -1.16506472e-01 4.91907179e-01 -8.18091869e-01 -5.22728860e-01 -4.11462426e-01 5.60509920e-01 -4.62888509e-01 1.42042947e+00 -2.33272836e-01 -6.01910539e-02 5.18009067e-01 4.19226021e-01 2.33826935e-01 -8.63533676e-01 -5.69196999e-01 -2.15346947e-01 4.41233158e-01 -5.10116756e-01 -2.87851661e-01 -5.22621036e-01 -9.02663529e-01 -2.18333420e-03 -7.18805850e-01 1.13309550e+00 6.44989491e-01 9.09694552e-01 1.73356161e-01 7.45932907e-02 6.22919559e-01 1.27140462e-01 -8.93936694e-01 -1.30657804e+00 -2.68798292e-01 6.65246785e-01 -4.10595164e-02 -2.35228390e-01 -3.53103638e-01 1.91288903e-01]	[11.619684219360352, 7.950410842895508]
0ebb491e-f863-4d28-9ab2-bd2c1ba2eba7	heterogeneous-separation-consistency-training	2204.11032	null	https://arxiv.org/abs/2204.11032v3	https://arxiv.org/pdf/2204.11032v3.pdf	Heterogeneous Separation Consistency Training for Adaptation of Unsupervised Speech Separation	Recently, supervised speech separation has made great progress. However, limited by the nature of supervised training, most existing separation methods require ground-truth sources and are trained on synthetic datasets. This ground-truth reliance is problematic, because the ground-truth signals are usually unavailable in real conditions. Moreover, in many industry scenarios, the real acoustic characteristics deviate far from the ones in simulated datasets. Therefore, the performance usually degrades significantly when applying the supervised speech separation models to real applications. To address these problems, in this study, we propose a novel separation consistency training, termed SCT, to exploit the real-world unlabeled mixtures for improving cross-domain unsupervised speech separation in an iterative manner, by leveraging upon the complementary information obtained from heterogeneous (structurally distinct but behaviorally complementary) models. SCT follows a framework using two heterogeneous neural networks (HNNs) to produce high confidence pseudo labels of unlabeled real speech mixtures. These labels are then updated, and used to refine the HNNs to produce more reliable consistent separation results for real mixture pseudo-labeling. To maximally utilize the large complementary information between different separation networks, a cross-knowledge adaptation is further proposed. Together with simulated dataset, those real mixtures with high confidence pseudo labels are then used to update the HNN separation models iteratively. In addition, we find that combing the heterogeneous separation outputs by a simple linear fusion can further slightly improve the final system performance.	['Yanhua Long', 'Jiangyu Han']	2022-04-23	null	null	null	null	['speech-separation']	['speech']	[ 3.20870757e-01 -1.07142448e-01 -2.61499256e-01 -3.23489934e-01 -1.05308473e+00 -5.73166728e-01 3.66940558e-01 -3.24844956e-01 -1.96425319e-02 8.57092857e-01 1.05696999e-01 -1.35802701e-01 -1.62839890e-01 -1.41972393e-01 -5.30995607e-01 -1.06033671e+00 3.25241506e-01 5.47400296e-01 1.39542565e-01 2.67109200e-02 -2.87253857e-01 2.72486717e-01 -1.48974180e+00 2.16546416e-01 1.45459628e+00 7.95346916e-01 3.51982296e-01 3.63595724e-01 -2.59903997e-01 5.12956202e-01 -8.92455935e-01 -1.46889016e-01 2.98859686e-01 -5.98381460e-01 -2.02603355e-01 5.10501027e-01 3.90336402e-02 5.92104830e-02 -1.40937299e-01 1.43708885e+00 5.24636626e-01 2.92691380e-01 6.04990482e-01 -1.24771690e+00 -4.45107669e-01 1.05133224e+00 -6.13409936e-01 -2.36609608e-01 -6.62074089e-02 2.91099455e-02 5.07414818e-01 -8.35710883e-01 1.81219857e-02 1.20145690e+00 4.35124934e-01 5.79868615e-01 -1.24264097e+00 -1.19588768e+00 4.22146052e-01 -3.70740369e-02 -1.34966254e+00 -1.04217219e+00 1.15112901e+00 -1.60342768e-01 4.75064665e-01 2.17176735e-01 3.33227932e-01 1.27747941e+00 -6.57540381e-01 8.15686285e-01 1.06968856e+00 -4.43052709e-01 2.92226881e-01 4.05776054e-01 1.47768348e-01 3.37060034e-01 4.51516956e-01 -5.69121400e-03 -3.93866807e-01 -2.93440353e-02 4.50443774e-01 1.60582308e-02 -6.50111318e-01 -3.24993938e-01 -1.15545046e+00 2.60174453e-01 7.79763535e-02 5.94744146e-01 -3.18515718e-01 -5.65620959e-01 1.60925940e-01 5.44897281e-02 3.79939288e-01 3.51197541e-01 -2.90595144e-01 1.78743154e-01 -1.21282136e+00 -3.69079381e-01 7.64072478e-01 1.03092217e+00 7.93321073e-01 5.49199522e-01 1.27985343e-01 1.40387237e+00 5.94403148e-01 8.55089426e-01 8.74017358e-01 -9.39577878e-01 7.44861603e-01 3.45830500e-01 9.65845063e-02 -8.30865383e-01 -5.72846234e-02 -9.44198191e-01 -9.57920909e-01 -3.23548242e-02 3.81038308e-01 -2.21193999e-01 -1.16370225e+00 1.82511151e+00 2.92806119e-01 6.53786123e-01 6.77628398e-01 9.49967206e-01 5.56013882e-01 6.97500825e-01 -1.26446322e-01 -5.10230124e-01 8.56764495e-01 -1.20743525e+00 -9.73185480e-01 -4.73858088e-01 2.10654885e-01 -7.74676681e-01 6.63470626e-01 5.93382418e-01 -7.78791726e-01 -7.96686471e-01 -1.48054945e+00 7.68630624e-01 -1.83408573e-01 4.27917004e-01 1.16589293e-01 8.62173140e-01 -6.71649635e-01 2.54671395e-01 -7.27011800e-01 -2.38283593e-02 1.46709695e-01 4.39859658e-01 -2.46388718e-01 -9.61121842e-02 -1.21706510e+00 4.42600548e-01 6.10943854e-01 4.17021841e-01 -9.77480829e-01 -2.27543548e-01 -9.16952431e-01 -2.70755310e-02 7.22165883e-01 -2.70715266e-01 1.29576361e+00 -1.14451659e+00 -1.82467067e+00 2.14547515e-01 -8.25480595e-02 -2.58553654e-01 1.62021607e-01 1.39325619e-01 -1.18871725e+00 3.16444300e-02 1.26512557e-01 3.97173971e-01 1.21067536e+00 -2.06875730e+00 -6.87114418e-01 -2.37878665e-01 -4.70604092e-01 1.92068473e-01 -3.23672771e-01 -3.24200481e-01 -6.22141361e-01 -6.35378182e-01 5.89564443e-01 -9.82787430e-01 -2.29151711e-01 -7.32166469e-01 -7.22172081e-01 2.91104704e-01 8.15805137e-01 -7.20427573e-01 9.91499782e-01 -2.38578439e+00 1.12293996e-01 3.40449810e-01 1.95629895e-01 7.35754013e-01 -1.43779188e-01 -7.03597814e-02 -2.31189385e-01 -9.97887477e-02 -5.51513910e-01 -6.12577200e-01 1.37108669e-01 1.82750732e-01 -2.52141178e-01 2.53749251e-01 6.53769001e-02 3.75996649e-01 -9.40540373e-01 -4.19085979e-01 1.90253258e-01 1.98019415e-01 -4.23025489e-02 3.89053166e-01 -4.05640453e-02 6.28715754e-01 -2.00831518e-01 6.51488364e-01 8.78634691e-01 -1.78563744e-01 4.93969768e-01 -4.75496322e-01 2.30080456e-01 -4.45049517e-02 -1.42953062e+00 1.48411751e+00 -4.06209648e-01 3.92036498e-01 5.96541703e-01 -1.25776362e+00 1.00512421e+00 6.16790354e-01 3.61652911e-01 -3.73264700e-01 2.02281922e-01 4.83726501e-01 2.60984480e-01 -2.80736208e-01 2.08741143e-01 -3.80277365e-01 1.33683071e-01 1.93298236e-01 1.34310812e-01 -2.08096191e-01 2.56069768e-02 -2.73828954e-02 5.57568491e-01 -1.53646260e-01 3.39499209e-03 2.35067397e-01 8.22946668e-01 -1.99842736e-01 9.17181194e-01 4.97773618e-01 -3.16449225e-01 6.26924515e-01 -1.75348356e-01 4.91999894e-01 -7.09478915e-01 -1.21311677e+00 9.35076475e-02 5.06309748e-01 3.99590939e-01 -5.63807487e-02 -9.20783699e-01 -7.28125572e-01 -2.49432087e-01 6.62262082e-01 -5.75280599e-02 -4.70644563e-01 -2.49820441e-01 -7.71962345e-01 7.57696033e-01 4.65496868e-01 6.87718630e-01 -7.02461898e-01 3.52286071e-01 2.74010926e-01 -3.15293998e-01 -1.40660894e+00 -4.12818700e-01 4.88123357e-01 -5.76515198e-01 -7.12669194e-01 -8.21838677e-01 -8.16721976e-01 6.65669262e-01 7.26786733e-01 5.31983495e-01 -3.07536274e-01 5.28018653e-01 5.68379276e-02 -2.89686114e-01 -1.94335192e-01 -8.59497905e-01 -1.17126100e-01 6.62193477e-01 4.55218136e-01 2.53151387e-01 -7.42117941e-01 3.26818484e-03 5.34927189e-01 -8.84774208e-01 8.90509412e-02 8.00570011e-01 9.15049374e-01 4.52583194e-01 7.90642083e-01 1.05161989e+00 -6.83175445e-01 5.33935726e-01 -5.19646883e-01 -4.19368446e-01 4.08498585e-01 -5.74140608e-01 1.18479475e-01 9.99861896e-01 -9.46828902e-01 -1.49719226e+00 -1.35902032e-01 7.52174854e-02 -9.89515305e-01 -4.19234961e-01 4.99197245e-01 -8.84990335e-01 1.91708252e-01 4.94063586e-01 2.97754139e-01 -9.23334211e-02 -4.92398530e-01 2.56852418e-01 1.10964596e+00 7.75876164e-01 -7.15640008e-01 9.40717697e-01 1.57222256e-01 -5.50804198e-01 -8.29596400e-01 -8.31910551e-01 -4.52733308e-01 -4.18965459e-01 -2.82867085e-02 5.75380564e-01 -1.10849166e+00 -1.23348303e-01 7.49422789e-01 -7.61239707e-01 -3.28738242e-01 -9.88526121e-02 1.07419968e+00 -2.31101424e-01 6.24131620e-01 -4.82642531e-01 -1.15304565e+00 8.27243328e-02 -1.40825295e+00 8.32618952e-01 4.62719232e-01 1.21376134e-01 -6.74753726e-01 -1.54476717e-01 7.19224811e-01 1.16815463e-01 -4.15007740e-01 6.03019416e-01 -1.12495041e+00 -3.28311950e-01 -1.29741400e-01 -1.16410293e-01 8.13183367e-01 7.26264000e-01 1.36593180e-02 -1.28428161e+00 -1.75572038e-01 3.08273375e-01 -4.27545309e-02 5.62474132e-01 3.20772737e-01 7.92073548e-01 2.72953939e-02 -5.13428509e-01 3.45154256e-01 7.82299340e-01 6.78253174e-01 4.83483896e-02 -2.88161896e-02 8.60770643e-01 6.53476715e-01 5.88917851e-01 7.51845688e-02 5.48089780e-02 5.21025896e-01 8.74923263e-03 -1.36934981e-01 -3.13792139e-01 -1.76830053e-01 6.05989277e-01 1.55442894e+00 2.72671610e-01 -4.91147310e-01 -6.96402967e-01 4.18272465e-01 -1.82841539e+00 -7.84490585e-01 1.74674258e-01 2.27034116e+00 9.02622044e-01 3.53344649e-01 -5.27556092e-02 3.47116321e-01 1.27823567e+00 7.50056282e-02 -7.76216328e-01 3.94636333e-01 -3.14171970e-01 -2.31739417e-01 3.21089625e-01 3.78138393e-01 -9.62903500e-01 6.31982982e-01 5.49560452e+00 1.11720824e+00 -1.21285760e+00 1.71757221e-01 5.17476797e-01 8.10372382e-02 -3.14223856e-01 -9.60168540e-02 -6.04268909e-01 6.31215513e-01 1.05514348e+00 2.51152590e-02 5.75001538e-01 5.18738329e-01 2.00011358e-01 1.26288995e-01 -8.44296098e-01 1.22586882e+00 1.81165725e-01 -6.70225203e-01 -5.94406389e-02 -9.16996971e-02 8.07643175e-01 -1.99377030e-01 1.38883322e-01 4.93776500e-01 6.31642342e-01 -6.60172701e-01 7.05577016e-01 4.19131577e-01 4.75499868e-01 -6.50306165e-01 7.01482356e-01 7.79930353e-01 -1.16661549e+00 -1.41603678e-01 -1.02574110e-01 5.61195016e-01 2.83213615e-01 8.94358158e-01 -9.12317395e-01 9.78709280e-01 2.22316399e-01 5.67640841e-01 -1.85357049e-01 8.34893465e-01 -2.96227187e-01 9.07171309e-01 -3.09057027e-01 3.77327204e-01 -6.76287115e-02 -4.71897393e-01 5.76295912e-01 1.10182118e+00 4.14895564e-01 -5.02619222e-02 2.94827193e-01 7.92100072e-01 -3.36440057e-02 -1.09854110e-01 -3.99527848e-01 -3.69785577e-01 7.47048199e-01 1.18786263e+00 -7.69387305e-01 -6.13644242e-01 -3.29924524e-01 8.59554708e-01 1.29916683e-01 8.53530347e-01 -9.82910037e-01 -1.02596387e-01 4.44407731e-01 -4.31081086e-01 7.24006966e-02 -2.49501079e-01 -2.79530091e-03 -1.41827929e+00 6.76623806e-02 -1.29340398e+00 -4.12237048e-02 -6.63053691e-01 -1.52209413e+00 8.07408869e-01 4.63321544e-02 -1.58663964e+00 -2.78091133e-01 -2.67431408e-01 -4.15950000e-01 8.86530995e-01 -1.25767100e+00 -9.72574472e-01 -2.21488655e-01 3.47187281e-01 6.76313460e-01 -5.49403131e-01 4.89855379e-01 5.74922383e-01 -1.14351404e+00 6.99429154e-01 3.32076967e-01 8.67999420e-02 9.17279959e-01 -9.40351367e-01 6.78695738e-02 1.05203784e+00 3.52066189e-01 6.21924400e-01 6.38477266e-01 -6.74718618e-01 -1.01269150e+00 -1.10260224e+00 4.08481248e-03 1.49751335e-01 6.02918684e-01 -2.35647768e-01 -1.15797985e+00 3.23361129e-01 4.78900522e-02 -6.40430674e-02 8.97284985e-01 -9.67950448e-02 -3.23363453e-01 -3.42304349e-01 -1.03570640e+00 6.65133536e-01 9.56298172e-01 -6.18474483e-01 -6.60868585e-01 -3.33254635e-02 9.70456421e-01 -1.69215396e-01 -6.29348099e-01 5.53316653e-01 2.95599520e-01 -7.31031179e-01 9.85749364e-01 -3.27123463e-01 -1.90906689e-01 -8.15051019e-01 -5.49902856e-01 -1.70804894e+00 -1.68754578e-01 -3.92256677e-01 -2.02826679e-01 1.74450374e+00 6.56446338e-01 -8.58729839e-01 6.64644539e-01 5.45555472e-01 -4.61072564e-01 -1.38660938e-01 -7.54990697e-01 -1.13305390e+00 -4.41115797e-01 -4.78952765e-01 8.92653584e-01 1.17179728e+00 5.22880629e-02 4.63171631e-01 -3.55888039e-01 7.78199971e-01 7.06508279e-01 8.94267112e-02 7.97159731e-01 -1.09349239e+00 -6.29315436e-01 -4.65378881e-01 -1.25840411e-01 -1.20104098e+00 6.00514531e-01 -7.93462813e-01 5.67547441e-01 -1.15113664e+00 -2.90295836e-02 -6.81711018e-01 -4.86397266e-01 2.04712957e-01 -3.78560603e-01 7.93966502e-02 1.09337106e-01 3.49710047e-01 -5.07057190e-01 7.70162582e-01 1.13006246e+00 -6.20761991e-01 -3.83120000e-01 1.59255832e-01 -6.94838345e-01 6.82901204e-01 8.19474876e-01 -4.25132990e-01 -8.22975278e-01 -1.84958532e-01 -6.33452237e-01 1.83584422e-01 -1.00315392e-01 -1.47973800e+00 3.49523187e-01 -2.39002764e-01 2.53826022e-01 -4.31409270e-01 6.64810479e-01 -1.18736696e+00 4.69293416e-01 -1.85499191e-02 -1.03839472e-01 -7.68876851e-01 1.99497715e-01 7.41732061e-01 -6.09264076e-01 -3.82894367e-01 6.48556948e-01 1.41991854e-01 -4.19167161e-01 1.38129711e-01 -2.62022525e-01 -2.14564070e-01 8.67281318e-01 -3.27952057e-01 -1.81408167e-01 -4.74912494e-01 -8.47084045e-01 1.40910491e-01 3.47583950e-01 2.87650108e-01 2.67710447e-01 -1.40244734e+00 -4.28718418e-01 3.95035505e-01 -2.04119291e-02 3.06553751e-01 3.76329184e-01 8.71486306e-01 2.22628847e-01 1.82912126e-01 1.94418758e-01 -6.50615394e-01 -1.03939605e+00 9.06897426e-01 4.88003522e-01 2.87095588e-02 -1.59170344e-01 5.89853227e-01 4.06539530e-01 -5.71290553e-01 3.15245360e-01 -4.11243767e-01 -9.92911533e-02 -3.55389155e-02 3.95183116e-01 2.25411728e-01 2.00791597e-01 -7.61899173e-01 -1.78063601e-01 3.71526986e-01 5.10161482e-02 -4.65496123e-01 8.91058683e-01 -3.18865418e-01 1.84598178e-01 6.78974867e-01 9.03114617e-01 5.53873301e-01 -1.21622646e+00 -4.73606318e-01 -8.01302567e-02 -3.00075948e-01 -1.68311208e-01 -7.01720178e-01 -1.34220052e+00 6.98952556e-01 4.72384810e-01 3.49024653e-01 1.25963020e+00 -8.66677836e-02 7.14566171e-01 3.50999355e-01 3.04609984e-01 -1.05448675e+00 -7.17359781e-02 1.05319306e-01 4.05859143e-01 -1.28995025e+00 -3.72491241e-01 -6.93712890e-01 -6.79449022e-01 8.14934731e-01 8.55062485e-01 4.25900042e-01 6.05338931e-01 4.38103884e-01 4.16912854e-01 1.15298852e-01 -3.39597911e-01 -2.36487970e-01 2.60184735e-01 7.42288172e-01 6.57445937e-02 2.81426162e-01 3.04192752e-01 1.12650895e+00 9.08573568e-02 -3.59337866e-01 2.39377365e-01 5.24840653e-01 -4.99301970e-01 -1.37991655e+00 -8.60090017e-01 3.75346243e-01 -9.20504797e-03 1.32028326e-01 -2.55000442e-01 2.25512221e-01 2.37935349e-01 1.53937912e+00 -3.63665462e-01 -7.27344453e-01 2.55340874e-01 2.38950804e-01 1.96866751e-01 -6.52883887e-01 -4.75971401e-02 5.28940678e-01 1.98706433e-01 1.23787381e-01 -6.69487536e-01 -5.34367383e-01 -1.40608704e+00 4.69852611e-02 -8.65633547e-01 6.29927218e-01 5.35889745e-01 1.09597075e+00 2.00396046e-01 8.17217231e-01 7.45464265e-01 -9.58600998e-01 -6.16383314e-01 -1.19098020e+00 -8.28069448e-01 3.65094095e-01 3.99831921e-01 -9.06808138e-01 -7.14545608e-01 1.39554307e-01]	[14.790277481079102, 5.950778484344482]
c376d664-c191-473b-a9bc-46c5de17e891	take-a-prior-from-other-tasks-for-severe-blur	2302.06898	null	https://arxiv.org/abs/2302.06898v1	https://arxiv.org/pdf/2302.06898v1.pdf	Take a Prior from Other Tasks for Severe Blur Removal	Recovering clear structures from severely blurry inputs is a challenging problem due to the large movements between the camera and the scene. Although some works apply segmentation maps on human face images for deblurring, they cannot handle natural scenes because objects and degradation are more complex, and inaccurate segmentation maps lead to a loss of details. For general scene deblurring, the feature space of the blurry image and corresponding sharp image under the high-level vision task is closer, which inspires us to rely on other tasks (e.g. classification) to learn a comprehensive prior in severe blur removal cases. We propose a cross-level feature learning strategy based on knowledge distillation to learn the priors, which include global contexts and sharp local structures for recovering potential details. In addition, we propose a semantic prior embedding layer with multi-level aggregation and semantic attention transformation to integrate the priors effectively. We introduce the proposed priors to various models, including the UNet and other mainstream deblurring baselines, leading to better performance on severe blur removal. Extensive experiments on natural image deblurring benchmarks and real-world images, such as GoPro and RealBlur datasets, demonstrate our method's effectiveness and generalization ability.	['Yanning Zhang', 'Sung-Eui Yoon', 'Qingsen Yan', 'Jinqiu Sun', 'Yu Zhu', 'Danna Xue', 'Pei Wang']	2023-02-14	null	null	null	null	['deblurring']	['computer-vision']	[ 2.75058478e-01 -3.91076744e-01 5.21559119e-02 -3.71677577e-01 -3.03182989e-01 -3.57185543e-01 3.68466854e-01 -8.04636240e-01 -2.61960411e-03 6.65584981e-01 8.42639863e-01 2.07934558e-01 -7.24530816e-02 -2.98949897e-01 -6.65873110e-01 -8.91535759e-01 5.81488371e-01 -2.05374226e-01 2.54837751e-01 -4.86999042e-02 2.34774366e-01 2.06975266e-01 -1.24727476e+00 1.09615743e-01 1.19452441e+00 8.48791480e-01 4.89258170e-01 4.42815542e-01 1.49368271e-01 7.69715488e-01 -4.56172675e-01 -2.92723835e-01 2.55296946e-01 -4.54269171e-01 -7.88328528e-01 3.94992620e-01 9.19030190e-01 -9.30081785e-01 -8.29577684e-01 1.57351744e+00 4.06246871e-01 1.76044598e-01 6.11602068e-01 -8.59223008e-01 -1.42007601e+00 2.33307824e-01 -1.09690642e+00 5.83358824e-01 2.69937336e-01 4.63848621e-01 5.66157222e-01 -1.10354161e+00 2.68681139e-01 1.53924155e+00 6.52485251e-01 5.74452102e-01 -1.12866318e+00 -6.14349842e-01 1.70004159e-01 6.93912745e-01 -1.26370573e+00 -8.65705252e-01 8.85564864e-01 -3.67524207e-01 3.27398539e-01 2.70132959e-01 2.47875169e-01 1.15230870e+00 6.53158724e-02 8.85927260e-01 1.27164662e+00 5.71595505e-02 -3.23570333e-02 -9.97853875e-02 3.28016311e-01 4.27271634e-01 6.33880854e-01 2.68067986e-01 -2.55574077e-01 1.38223857e-01 1.13928127e+00 4.50129718e-01 -1.16569698e+00 -1.64810553e-01 -1.06832922e+00 3.25853646e-01 7.01171041e-01 1.87510997e-01 -3.79385382e-01 1.03486530e-01 -1.83841586e-02 -1.51379965e-02 6.39060318e-01 3.87184799e-01 -2.87505269e-01 1.54169887e-01 -1.06542301e+00 3.67415510e-02 3.40648353e-01 9.52908814e-01 7.91385293e-01 -5.00058793e-02 -5.33525407e-01 1.19860768e+00 2.74809718e-01 3.77718151e-01 4.59961683e-01 -9.99749303e-01 3.92219305e-01 2.85493076e-01 3.54352295e-01 -1.19672620e+00 6.61698133e-02 -5.34222722e-01 -1.24378717e+00 -5.67177385e-02 2.34748736e-01 -7.64548630e-02 -1.17571521e+00 1.59622288e+00 1.32808849e-01 9.19351459e-01 -4.59891697e-03 1.69134390e+00 8.25656533e-01 6.62746727e-01 -3.22028041e-01 -2.83249319e-01 1.46627450e+00 -1.34004521e+00 -1.00811815e+00 -6.54869974e-01 -2.61487812e-01 -8.58029664e-01 9.37251627e-01 2.42501974e-01 -1.10498559e+00 -7.47524559e-01 -9.04258847e-01 -4.93186831e-01 9.17081386e-02 1.03712820e-01 5.08129537e-01 3.29578429e-01 -1.14152002e+00 4.22576159e-01 -7.30351210e-01 -2.75812924e-01 8.76966953e-01 -5.64021058e-02 -1.76997349e-01 -7.46851027e-01 -1.24259138e+00 9.26549852e-01 1.98254958e-01 6.63818240e-01 -1.08881545e+00 -7.20126271e-01 -1.07644224e+00 2.34871805e-01 3.00379127e-01 -8.72236609e-01 9.96785998e-01 -7.70951331e-01 -1.31639075e+00 4.65079993e-01 -3.02800238e-01 5.87416403e-02 5.54534972e-01 -6.64622247e-01 -2.92610586e-01 1.47182688e-01 8.83380696e-02 4.27600712e-01 1.40627670e+00 -1.50566876e+00 -3.86137873e-01 -3.93293828e-01 6.61881045e-02 3.89975369e-01 -3.29580367e-01 1.79413464e-02 -7.76318550e-01 -1.06592309e+00 7.23315403e-02 -4.95587140e-01 -6.46752641e-02 -8.81704912e-02 -2.92148083e-01 1.08213179e-01 1.12728012e+00 -1.19540119e+00 1.17160070e+00 -2.12878966e+00 5.26506782e-01 -5.55252910e-01 5.06105185e-01 3.78834546e-01 -2.93871760e-01 -1.89731210e-01 -1.96356550e-01 -1.43701077e-01 -5.66886425e-01 -2.75798380e-01 -2.68221140e-01 3.19130160e-02 -5.39942384e-01 6.66102767e-01 2.22213432e-01 1.12166595e+00 -1.01567733e+00 -1.81335762e-01 3.28295290e-01 6.04395032e-01 -4.44836676e-01 5.35778165e-01 1.16737209e-01 4.93682653e-01 -3.89222801e-01 6.53117001e-01 1.31597435e+00 -4.12362337e-01 -5.64184010e-01 -6.51324272e-01 1.36479750e-01 -2.00407282e-01 -8.62590611e-01 1.82900655e+00 -4.15195644e-01 7.38722503e-01 4.00635600e-01 -6.66989982e-01 4.04945374e-01 8.28432366e-02 1.10669089e-02 -2.35234916e-01 8.81157666e-02 4.84213280e-03 -2.46434972e-01 -9.15183306e-01 5.19522369e-01 -5.08893840e-02 3.73171002e-01 2.37203702e-01 -6.77098930e-02 -3.03274572e-01 -3.57249796e-01 1.46822914e-01 8.68640840e-01 1.57928959e-01 -8.06791186e-02 -3.12930405e-01 7.09686160e-01 -5.55035591e-01 6.62149191e-01 4.17674422e-01 -4.26701099e-01 1.13499653e+00 9.70416889e-02 -2.48108879e-01 -6.73253059e-01 -9.43154156e-01 -1.63284764e-01 8.12051773e-01 9.49683487e-01 5.99025115e-02 -1.07623839e+00 -5.37184119e-01 -1.23353697e-01 5.78491867e-01 -6.20735288e-01 -4.88305926e-01 -5.31953454e-01 -1.06398439e+00 8.41491297e-02 3.84239048e-01 1.00360715e+00 -7.72918344e-01 -4.99009453e-02 -8.85402262e-02 -5.52331626e-01 -1.24323308e+00 -1.27980745e+00 -5.29983699e-01 -7.14039564e-01 -1.10229635e+00 -1.21338165e+00 -9.57708299e-01 8.54577422e-01 9.61283803e-01 7.71246731e-01 4.19096276e-02 -3.20494086e-01 2.08968595e-01 -1.75394267e-01 2.08920240e-01 -6.81892335e-02 -4.42803204e-01 -7.12597324e-03 3.55648786e-01 1.17291220e-01 -5.45742035e-01 -1.19432926e+00 4.01858032e-01 -1.16899824e+00 2.20938250e-01 8.81342888e-01 1.18240118e+00 -3.65609080e-02 2.17855215e-01 3.80119681e-01 -5.54652095e-01 7.68496096e-01 -5.10930181e-01 -3.63655478e-01 3.20657820e-01 -5.33011913e-01 5.26720583e-02 3.29507023e-01 -4.61770862e-01 -1.61036360e+00 -3.78475100e-01 3.15778732e-01 -9.73268390e-01 -2.30855852e-01 1.66471615e-01 -5.02062201e-01 -1.04063608e-01 4.57749456e-01 5.94422340e-01 -4.18457463e-02 -7.78924763e-01 5.50291717e-01 8.74251008e-01 8.62028897e-01 -3.22028846e-01 9.14780438e-01 4.15417314e-01 -4.84892458e-01 -7.18735397e-01 -1.06703508e+00 -3.17005992e-01 -4.44459110e-01 7.50734434e-02 8.56171131e-01 -1.17476404e+00 -2.74316907e-01 1.10104537e+00 -1.47224057e+00 -1.49112329e-01 1.20549396e-01 3.98014575e-01 -3.39620292e-01 9.50585067e-01 -9.63989675e-01 -2.69822389e-01 -2.67814577e-01 -1.43431878e+00 1.18874002e+00 6.74129128e-01 3.20229441e-01 -8.29669356e-01 -2.33056948e-01 7.81812429e-01 7.05228150e-01 -1.74026862e-01 5.39296925e-01 1.81690857e-01 -9.13642347e-01 1.94150537e-01 -9.48023498e-01 7.41739213e-01 7.37265527e-01 -5.16198218e-01 -1.17496824e+00 -4.71394241e-01 5.70752025e-01 -1.00563511e-01 1.34885263e+00 5.75761080e-01 1.34763908e+00 -5.53020835e-01 -4.34543788e-01 9.30003762e-01 1.20318770e+00 -2.06803367e-01 8.06317210e-01 -1.55940324e-01 1.19713843e+00 4.85403925e-01 4.23518121e-01 1.44813424e-02 3.98044407e-01 6.87485516e-01 5.06360173e-01 -2.16912150e-01 -7.93019116e-01 -4.75756787e-02 4.56585020e-01 7.52756357e-01 3.50508839e-02 6.72929734e-02 -5.01873970e-01 6.68899417e-01 -1.83583331e+00 -9.13932621e-01 2.36570910e-02 1.81959450e+00 1.36324310e+00 -3.45273018e-01 -5.04535675e-01 -4.75500762e-01 1.04972303e+00 5.51945567e-01 -7.91452527e-01 3.46164793e-01 -2.19237104e-01 -1.70554444e-01 3.43611479e-01 8.17278802e-01 -1.09312809e+00 1.13142312e+00 5.61050940e+00 9.98450458e-01 -1.05843377e+00 2.84344703e-01 9.32751894e-01 3.28706652e-02 -2.10801303e-01 3.16200368e-02 -4.23683971e-01 9.20244455e-01 2.51672864e-01 -3.41723338e-02 9.58114684e-01 5.97427309e-01 3.57639730e-01 -3.92557494e-02 -1.00818384e+00 1.52234399e+00 3.57239097e-01 -1.16970956e+00 4.15900573e-02 -1.97214916e-01 1.14558721e+00 -8.94959420e-02 1.84330329e-01 1.22628957e-01 2.96556830e-01 -1.16921639e+00 5.67409098e-01 8.24593663e-01 8.70460570e-01 -3.05610627e-01 7.75255144e-01 1.61143064e-01 -8.24148953e-01 5.05855726e-03 -6.12699568e-01 3.08770500e-02 1.21137962e-01 7.30964839e-01 -2.16037065e-01 6.36333287e-01 8.91386151e-01 1.37512541e+00 -7.04088092e-01 1.20536935e+00 -3.74563277e-01 4.66847658e-01 2.48083308e-01 6.33699119e-01 -1.70279145e-01 -3.43239516e-01 6.68887734e-01 1.21461773e+00 2.70195335e-01 3.83169144e-01 -1.90814734e-01 1.12696028e+00 -2.19884828e-01 -5.40467024e-01 -2.15605929e-01 2.84294724e-01 3.92931819e-01 1.23485327e+00 -1.68917611e-01 -3.66246194e-01 -5.84034026e-01 1.63655233e+00 1.09035186e-01 8.99049342e-01 -8.78244340e-01 -4.78175849e-01 1.15044630e+00 -9.30327401e-02 3.42886686e-01 -7.59461522e-02 -2.54030406e-01 -1.74232090e+00 1.32279135e-02 -8.63899350e-01 7.70998225e-02 -1.17854643e+00 -1.66867268e+00 4.78109419e-01 -1.72549896e-02 -1.09397364e+00 3.74955118e-01 -4.70713735e-01 -7.12437987e-01 1.16733384e+00 -1.99360156e+00 -1.02171659e+00 -8.93318594e-01 5.79149604e-01 9.46367860e-01 1.59388348e-01 -6.01837747e-02 3.50431740e-01 -8.31752539e-01 3.64618480e-01 3.03966980e-02 1.17993183e-01 1.04393137e+00 -1.03279591e+00 2.96690524e-01 1.28268588e+00 -3.83457094e-01 7.10434616e-01 7.92292237e-01 -6.44215047e-01 -1.43767273e+00 -1.36493456e+00 1.79791957e-01 -6.44820750e-01 5.41630626e-01 -1.08799025e-01 -1.32129931e+00 5.93056083e-01 5.92541456e-01 3.65884453e-01 -4.30753008e-02 -3.43367189e-01 -3.62442464e-01 -2.33139113e-01 -9.02675629e-01 5.32646298e-01 1.20477140e+00 -5.40729463e-01 -9.66298938e-01 1.30728945e-01 9.24664259e-01 -5.78892469e-01 -5.37297964e-01 5.19475579e-01 2.38089412e-01 -1.06086636e+00 1.15803790e+00 -3.51622611e-01 6.91035748e-01 -6.28291845e-01 1.96508780e-01 -1.59901464e+00 -6.81872189e-01 -7.99621403e-01 -4.78201896e-01 1.31683111e+00 -3.34472030e-01 -7.02196240e-01 3.89070302e-01 5.44941247e-01 -2.40533754e-01 -4.69864458e-01 -6.71702087e-01 -5.24342597e-01 -1.37928918e-01 6.71152174e-02 7.96723008e-01 1.12913597e+00 -4.10162061e-01 4.52111512e-01 -7.70418704e-01 5.02805531e-01 8.54223907e-01 3.50730002e-01 4.75602597e-01 -8.99374843e-01 -2.00110599e-01 -5.78648150e-01 -1.92486793e-01 -1.76623631e+00 1.07607313e-01 -5.91671884e-01 3.18157583e-01 -1.55041528e+00 7.23376215e-01 -1.35902554e-01 -3.46465409e-01 2.52344131e-01 -8.91689301e-01 2.06653893e-01 -1.35788947e-01 6.34709358e-01 -4.39412564e-01 1.03534126e+00 1.82759786e+00 -3.85105908e-01 -2.92455312e-02 -3.55684996e-01 -1.11123955e+00 7.10446298e-01 2.31883109e-01 -1.09566063e-01 -4.91287738e-01 -1.02154791e+00 -3.31582934e-01 1.42123904e-02 7.18339205e-01 -6.98096097e-01 3.66882384e-01 -4.43428457e-01 5.99424005e-01 -2.39244133e-01 3.49456191e-01 -7.73667693e-01 -4.67623509e-02 1.05346233e-01 -9.87739041e-02 -4.91234094e-01 7.84933865e-02 9.74773228e-01 -3.73641312e-01 3.70490504e-03 1.04556584e+00 3.63957584e-02 -7.30751336e-01 6.24380827e-01 2.69837052e-01 5.83281294e-02 7.60203183e-01 -2.74324864e-01 -7.65066862e-01 -4.43572491e-01 -4.43596065e-01 2.71400839e-01 7.39563346e-01 7.44433701e-01 8.78679037e-01 -1.24468625e+00 -8.95846069e-01 2.99594522e-01 -1.24578901e-01 2.75193483e-01 6.64206505e-01 1.08706725e+00 -4.13983107e-01 1.70424163e-01 -2.49971926e-01 -5.24024725e-01 -9.70995069e-01 6.18424237e-01 4.77433294e-01 3.15398604e-01 -6.30326867e-01 1.13061202e+00 1.07087529e+00 2.10524902e-01 2.33072918e-02 -4.82545018e-01 -1.87731922e-01 -3.10203105e-01 8.46387386e-01 3.89281929e-01 -2.22063214e-01 -7.30427146e-01 -2.56439567e-01 7.63354838e-01 -3.98307979e-01 3.75759572e-01 1.28145945e+00 -7.99637735e-01 -5.38239002e-01 -2.08048135e-01 1.10747433e+00 1.01412855e-01 -1.95316470e+00 -6.18164361e-01 -2.83138782e-01 -1.08767688e+00 4.08462763e-01 -6.89294517e-01 -1.46065462e+00 9.91356313e-01 6.01223886e-01 -1.52665645e-01 1.48077881e+00 1.56729892e-01 1.00928569e+00 -1.52314007e-01 1.11962773e-01 -6.08829737e-01 3.76120538e-01 3.58815461e-01 1.26331675e+00 -1.39824605e+00 8.00655335e-02 -5.76255977e-01 -5.53599298e-01 9.72082675e-01 8.15499067e-01 -1.22034907e-01 6.30398929e-01 -1.39389500e-01 -2.99838055e-02 -2.42631622e-02 -4.80597973e-01 -9.67618264e-03 4.24696952e-01 4.91288096e-01 -3.10445074e-02 -2.53851533e-01 -2.77589373e-02 9.20900643e-01 1.67759851e-01 9.39955264e-02 7.29864836e-01 2.96525985e-01 -4.77295905e-01 -5.08421838e-01 -5.98793805e-01 2.78246820e-01 -2.65870541e-01 -3.47854197e-01 -2.25893170e-01 1.46893486e-01 2.03677624e-01 1.05724716e+00 -1.42861038e-01 -2.77236372e-01 7.45568797e-02 -3.96805167e-01 5.19858837e-01 -6.10264421e-01 9.23066437e-02 1.80799291e-01 -4.48861241e-01 -5.75727224e-01 -3.06773007e-01 -5.95479965e-01 -7.13132083e-01 -2.05984145e-01 -4.70732957e-01 -9.91802737e-02 7.19526783e-02 8.33539963e-01 5.72405457e-01 4.67866719e-01 6.54458821e-01 -1.04810357e+00 -5.86264372e-01 -1.29729712e+00 -5.86653769e-01 5.83451092e-01 9.49268997e-01 -7.62493730e-01 -5.44602156e-01 4.04137164e-01]	[11.524763107299805, -2.6930649280548096]
737ff9a9-4af0-4d94-ba81-5895fecb3c01	datasets-for-portuguese-legal-semantic	2306.00007	null	https://arxiv.org/abs/2306.00007v1	https://arxiv.org/pdf/2306.00007v1.pdf	Datasets for Portuguese Legal Semantic Textual Similarity: Comparing weak supervision and an annotation process approaches	The Brazilian judiciary has a large workload, resulting in a long time to finish legal proceedings. Brazilian National Council of Justice has established in Resolution 469/2022 formal guidance for document and process digitalization opening up the possibility of using automatic techniques to help with everyday tasks in the legal field, particularly in a large number of texts yielded on the routine of law procedures. Notably, Artificial Intelligence (AI) techniques allow for processing and extracting useful information from textual data, potentially speeding up the process. However, datasets from the legal domain required by several AI techniques are scarce and difficult to obtain as they need labels from experts. To address this challenge, this article contributes with four datasets from the legal domain, two with documents and metadata but unlabeled, and another two labeled with a heuristic aiming at its use in textual semantic similarity tasks. Also, to evaluate the effectiveness of the proposed heuristic label process, this article presents a small ground truth dataset generated from domain expert annotations. The analysis of ground truth labels highlights that semantic analysis of domain text can be challenging even for domain experts. Also, the comparison between ground truth and heuristic labels shows that heuristic labels are useful.	['Daniel de Oliveira', 'Aline Paes', 'Paulo Roberto dos S. Corval', 'Daniel da Silva Junior']	2023-05-29	null	null	null	null	['semantic-textual-similarity', 'semantic-similarity']	['natural-language-processing', 'natural-language-processing']	[ 3.60801637e-01 6.04358017e-01 -2.20150575e-01 -2.60049820e-01 -9.51496780e-01 -9.40770984e-01 7.27254987e-01 8.04259002e-01 -6.32368684e-01 9.14378643e-01 3.80133271e-01 -4.13467675e-01 -6.13575995e-01 -7.05301285e-01 -1.92235932e-01 -3.07041198e-01 5.16807199e-01 9.63847637e-01 1.05886959e-01 -3.37192297e-01 7.19715297e-01 6.58943653e-01 -1.56638956e+00 5.58857322e-01 1.29774570e+00 7.08551884e-01 8.37303773e-02 -7.87336454e-02 -5.47107279e-01 1.09868443e+00 -7.78222978e-01 -9.75659251e-01 5.49939215e-01 -3.12389523e-01 -1.14159298e+00 -1.29716158e-01 5.27475059e-01 -2.27648318e-01 -1.31814048e-01 1.14119756e+00 5.68780363e-01 2.28618816e-01 8.44283283e-01 -1.08001423e+00 -5.34382999e-01 7.07377136e-01 -6.93261176e-02 3.29562843e-01 6.22221589e-01 -1.63028240e-01 9.96417344e-01 -4.70546484e-01 1.19119763e+00 1.11647069e+00 5.14739931e-01 2.36770824e-01 -8.10324788e-01 -4.37632680e-01 -3.22903246e-01 7.80904114e-01 -1.29287803e+00 -4.71031457e-01 7.57040262e-01 -9.20093656e-01 7.97844648e-01 -7.74690360e-02 3.56777936e-01 9.28796828e-01 -3.04174185e-01 3.88210148e-01 1.19931686e+00 -7.13788211e-01 5.03836334e-01 2.09393293e-01 4.01700705e-01 2.67650396e-01 6.23605847e-01 -2.75535971e-01 -1.96134120e-01 -3.26260000e-01 5.25329746e-02 -1.27419502e-01 -3.08497474e-02 -1.82265528e-02 -7.05726504e-01 8.85997951e-01 -2.15557784e-01 1.07333088e+00 -6.70773208e-01 -4.24008757e-01 8.45359743e-01 2.88210928e-01 1.66557550e-01 7.40293860e-01 -8.66769329e-02 -5.02462685e-01 -1.15944242e+00 4.75411862e-01 8.23647082e-01 8.93191695e-01 5.84006011e-01 -4.22538429e-01 -3.08037102e-01 8.04867446e-01 7.39154965e-02 6.15651071e-01 2.30892450e-01 -1.25181592e+00 8.68148029e-01 1.17363548e+00 3.37563157e-01 -1.29094040e+00 -2.50240654e-01 2.71652881e-02 -3.74567688e-01 2.26068899e-01 7.45355606e-01 1.77706525e-01 -6.93588078e-01 1.02876830e+00 2.64204264e-01 -4.84400243e-01 8.23925212e-02 6.00586355e-01 6.84388280e-01 1.98997959e-01 4.23948556e-01 -2.84795195e-01 1.39369798e+00 -4.34735149e-01 -1.11647904e+00 1.91154028e-03 7.11612523e-01 -9.45834279e-01 1.00578952e+00 4.27133232e-01 -1.03177691e+00 -2.38012195e-01 -7.53109157e-01 -2.41784737e-01 -7.33842790e-01 1.09691687e-01 2.80113488e-01 6.91255629e-01 -5.36635995e-01 7.08632171e-01 -1.83861524e-01 -6.53024554e-01 8.62785101e-01 -2.00170383e-01 -3.82012457e-01 -2.85631031e-01 -1.28168190e+00 1.33150744e+00 7.87433267e-01 -1.93495780e-01 -2.51879722e-01 -4.34648693e-01 -6.87888563e-01 7.06823915e-02 7.97751427e-01 -4.36750799e-02 8.81295085e-01 -4.27477777e-01 -8.94845843e-01 1.30495143e+00 2.05441453e-02 -6.41158223e-01 9.23694313e-01 -1.01852499e-01 -6.62782192e-01 5.36826849e-01 8.19220543e-01 1.61302298e-01 5.30935168e-01 -1.01887560e+00 -7.66205251e-01 -4.87526059e-01 2.75806695e-01 -2.00287968e-01 -3.65549862e-01 3.82373929e-01 -6.18014745e-02 -6.02783203e-01 -1.36591956e-01 -6.88987017e-01 7.66742900e-02 -3.91851187e-01 -1.76829509e-02 -3.25043708e-01 6.67028666e-01 -1.20849621e+00 1.43952429e+00 -1.88368475e+00 -2.76907235e-01 5.26117623e-01 2.27765754e-01 7.19139934e-01 3.26135486e-01 8.50016415e-01 1.23569913e-01 4.03335929e-01 -3.72161657e-01 3.79630834e-01 2.51240939e-01 4.12217289e-01 -3.64518553e-01 1.92611635e-01 -2.65505165e-01 6.49508238e-01 -1.12895381e+00 -1.00402510e+00 1.06917061e-01 1.84099691e-03 -2.14795694e-01 -2.19081059e-01 -5.57435490e-02 5.07239521e-01 -6.31158710e-01 7.12388992e-01 4.26085621e-01 -7.36347884e-02 4.75812942e-01 -2.76568651e-01 -5.32111824e-02 4.93530296e-02 -1.05842578e+00 1.76602113e+00 -3.92108560e-01 6.66871607e-01 -1.80149660e-01 -1.21933126e+00 1.01157296e+00 6.75803244e-01 6.40649796e-01 -1.20502281e+00 4.29130197e-01 5.86377621e-01 -5.78368232e-02 -1.10758936e+00 6.64821029e-01 -2.28720218e-01 6.91988990e-02 6.58411145e-01 -1.63807794e-01 -3.63288105e-01 9.26077962e-01 2.48818263e-01 1.04619491e+00 1.50203789e-02 3.71598810e-01 -8.24178830e-02 7.32334614e-01 4.87846047e-01 4.45576459e-01 4.74310011e-01 -2.74231255e-01 1.66446730e-01 4.42449450e-01 -5.96634448e-01 -1.30223978e+00 -4.37716872e-01 -3.46155226e-01 5.05915225e-01 -2.01617181e-02 -3.42724800e-01 -8.87071669e-01 -9.57837999e-01 -4.16961312e-02 9.64342535e-01 -5.32488167e-01 1.38453618e-01 -5.43285728e-01 9.66362283e-03 7.05823660e-01 2.36163154e-01 6.11151934e-01 -1.12285900e+00 -1.06231225e+00 3.95179212e-01 -5.57711005e-01 -1.47993994e+00 9.43212137e-02 -5.37618697e-01 -5.57081044e-01 -1.59661365e+00 -3.30810159e-01 -2.47243285e-01 4.70906675e-01 -9.93282199e-02 7.82286048e-01 2.14556456e-01 -4.89893347e-01 3.09538722e-01 -6.57656074e-01 -4.98168409e-01 -8.76313031e-01 1.83574006e-01 -3.09582502e-01 -2.11195752e-01 7.71835864e-01 -3.62393171e-01 -1.08014621e-01 2.00664490e-01 -1.20149815e+00 -3.76614362e-01 2.52109140e-01 4.50129509e-01 1.91944972e-01 2.25384548e-01 7.79755414e-01 -1.22897053e+00 1.14485681e+00 -4.91052479e-01 -5.05630672e-01 6.62389755e-01 -7.82983005e-01 8.84659030e-03 6.54515982e-01 8.40157941e-02 -1.32707512e+00 -5.69624007e-01 -6.25700876e-03 -2.64716595e-02 -3.33244562e-01 5.16913176e-01 2.09204685e-02 3.01243573e-01 1.15402603e+00 -4.52260077e-01 -7.29583353e-02 -5.30554771e-01 3.56514513e-01 1.14181745e+00 5.99811256e-01 -8.50354314e-01 7.92368293e-01 6.16171181e-01 7.33685046e-02 -4.78691876e-01 -1.18219650e+00 -5.06774306e-01 -9.62850094e-01 -5.50958097e-01 8.80197227e-01 -2.95868993e-01 -6.11083448e-01 -1.57036722e-01 -1.21294165e+00 7.70361125e-02 -6.71273053e-01 4.72174108e-01 -5.01860738e-01 7.16314435e-01 -1.58064067e-01 -8.56093526e-01 -3.32808018e-01 -8.75673234e-01 6.64298832e-01 4.99919988e-02 -6.48927450e-01 -9.63962853e-01 -1.12175517e-01 1.25627255e+00 3.35961550e-01 6.20550036e-01 1.16828740e+00 -1.09289718e+00 1.26217613e-02 -5.71188033e-01 -3.22846979e-01 4.92647797e-01 1.60397679e-01 -1.78367719e-01 -9.63350475e-01 1.19166814e-01 8.03227350e-02 -2.52433211e-01 2.49616578e-01 -1.14392012e-01 5.35792112e-01 -2.64158696e-01 -1.42254382e-01 -3.64839017e-01 1.27595150e+00 7.14137256e-01 7.01082528e-01 7.56320536e-01 3.53032887e-01 1.01526058e+00 1.22541428e+00 4.67053741e-01 2.15166554e-01 8.14283848e-01 -9.29349884e-02 3.12472194e-01 -2.12312117e-01 -1.35076299e-01 -3.60737920e-01 3.95744115e-01 -6.24737859e-01 1.29052535e-01 -1.43235874e+00 6.46088779e-01 -2.03932238e+00 -1.36034167e+00 -1.77445740e-01 1.99236882e+00 6.78805768e-01 7.83535913e-02 -3.57536785e-02 6.85206592e-01 8.49889874e-01 -2.16058925e-01 -1.88493744e-01 -4.99199629e-01 -1.50616929e-01 2.50711590e-01 4.19601381e-01 4.43508238e-01 -7.69553006e-01 9.00864482e-01 5.45269823e+00 1.15894067e+00 -5.74070454e-01 3.38603646e-01 8.99054036e-02 1.32747561e-01 -3.62805098e-01 1.94488555e-01 -4.33655441e-01 6.93386376e-01 8.23054910e-01 -5.94236612e-01 3.25558394e-01 7.94645965e-01 4.64549929e-01 -2.23496586e-01 -8.64404798e-01 1.03399253e+00 1.36054695e-01 -1.50893569e+00 1.26144424e-01 3.10622841e-01 8.00851822e-01 -4.73262936e-01 -5.85865140e-01 1.06885813e-01 1.92660704e-01 -9.29567754e-01 7.96617985e-01 5.54049611e-01 5.88842332e-01 -5.32158375e-01 1.10784376e+00 2.66738921e-01 -5.73461175e-01 -3.26045305e-01 -4.34601963e-01 -4.14289255e-03 4.85962927e-01 4.77265835e-01 -8.83771002e-01 8.63727033e-01 4.27122444e-01 6.11602902e-01 -4.73445445e-01 9.06480908e-01 -3.57716620e-01 2.68149316e-01 6.81037977e-02 2.11874336e-01 3.42761844e-01 -4.87664849e-01 5.21389425e-01 1.03289640e+00 3.29484344e-01 1.59410313e-01 4.22811396e-02 7.77860761e-01 -3.75433117e-02 6.00861132e-01 -8.28825235e-01 -4.47039269e-02 7.97526300e-01 9.68539894e-01 -9.93375301e-01 -6.03922367e-01 -1.45047829e-01 4.75080907e-01 9.79044959e-02 1.84777722e-01 -6.91949248e-01 -5.03257573e-01 3.24869491e-02 4.23371464e-01 -2.39744172e-01 8.85561481e-02 -4.09205258e-01 -8.17752302e-01 3.53761107e-01 -1.02860546e+00 5.82067251e-01 -6.89401269e-01 -1.09131694e+00 6.25720024e-01 3.95369858e-01 -1.30625641e+00 -3.70547324e-01 -4.72974539e-01 -1.27252797e-02 6.31350517e-01 -1.32331753e+00 -9.65439379e-01 -3.31618160e-01 4.62998539e-01 4.07166690e-01 -3.60720396e-01 5.92236757e-01 8.80356729e-01 -3.41677777e-02 4.50036265e-02 -2.11709961e-01 1.00870930e-01 7.18478501e-01 -9.47821498e-01 -1.63258150e-01 7.53218532e-01 -1.72070637e-02 5.93980730e-01 7.11944520e-01 -8.21427643e-01 -7.26317942e-01 -5.98586321e-01 1.32341158e+00 -4.15086657e-01 7.67027140e-01 2.72677749e-01 -1.00567722e+00 3.49819332e-01 1.79797158e-01 -6.69639647e-01 8.37171316e-01 -3.52955610e-01 -3.38376224e-01 1.63725600e-01 -1.65433705e+00 4.26564425e-01 1.13817549e+00 -7.86596715e-01 -1.26960945e+00 6.85917020e-01 -2.02289112e-02 -1.92990899e-01 -1.01986182e+00 6.18304610e-02 4.31153953e-01 -7.05837548e-01 7.26750672e-01 -6.18920863e-01 3.17616820e-01 -4.27638650e-01 -1.03264041e-01 -7.51051307e-01 2.43756190e-01 -3.89437348e-01 4.10210460e-01 1.56067288e+00 3.97297591e-01 -4.83132511e-01 5.34133077e-01 1.08622074e+00 -1.09643236e-01 -8.05214718e-02 -1.10240507e+00 -1.04553676e+00 2.03646608e-02 -6.17302179e-01 6.43564463e-01 1.43260896e+00 2.95195699e-01 1.36627063e-01 -4.07081805e-02 -2.85894334e-01 4.24333960e-01 7.84535334e-02 6.30397975e-01 -1.75908148e+00 3.13867539e-01 -4.02705163e-01 -5.83600581e-01 -1.52062863e-01 4.03109878e-01 -1.11922061e+00 -5.53091705e-01 -1.99892473e+00 -1.10348091e-02 -3.93127382e-01 2.91810393e-01 6.56021297e-01 1.87006727e-01 -3.56836729e-02 3.79936337e-01 5.35112619e-01 -4.00618374e-01 7.67077729e-02 1.13546991e+00 -3.15026313e-01 2.11439584e-03 -3.57469946e-01 -6.71117306e-01 8.39584827e-01 6.82816982e-01 -7.74079323e-01 -3.82037789e-01 -2.80539215e-01 5.00229537e-01 -2.32864276e-01 5.05337007e-02 -9.67601120e-01 3.38100463e-01 -2.18021035e-01 -3.60099196e-01 -3.27715933e-01 -4.20037098e-02 -1.27411997e+00 2.01012015e-01 3.02752137e-01 -2.00142428e-01 -2.04503670e-01 -7.64759555e-02 1.79519549e-01 -6.77596867e-01 -1.08524656e+00 5.34037650e-01 -2.00514570e-01 -7.21037686e-01 -2.69026071e-01 -4.67035949e-01 5.55413961e-01 1.31302774e+00 -6.95873439e-01 -5.82581758e-01 -1.82715580e-01 -8.33707094e-01 1.20593600e-01 5.14787793e-01 2.67799705e-01 1.06810823e-01 -1.14324117e+00 -6.76968575e-01 -4.53961909e-01 2.83676744e-01 -2.68455029e-01 3.64892423e-01 6.77467883e-01 -8.65156412e-01 5.70991814e-01 -4.73000497e-01 -1.41941980e-01 -1.22783637e+00 6.50080085e-01 -1.34255171e-01 -4.43602622e-01 -6.10550046e-01 -1.68459341e-01 -6.90294564e-01 -1.52924493e-01 -3.47446874e-02 -7.29601309e-02 -7.88816452e-01 7.71877944e-01 4.57973242e-01 9.44778442e-01 1.83230951e-01 -8.46432865e-01 -5.32222651e-02 8.39762628e-01 1.88852295e-01 -2.16571808e-01 1.39282680e+00 -3.54713425e-02 -3.08988988e-01 3.10971644e-02 7.75762677e-01 2.41730288e-01 -3.55156213e-01 -2.19980061e-01 8.24563026e-01 -8.34435701e-01 -3.28155279e-01 -8.34550261e-01 -7.17973113e-01 8.17993164e-01 3.42795789e-01 5.77843606e-01 8.01655710e-01 -8.27133730e-02 6.18566990e-01 6.11731768e-01 5.27507901e-01 -1.84160829e+00 -3.54337901e-01 3.16837400e-01 1.00723302e+00 -1.12085843e+00 3.70674096e-02 -5.62485635e-01 -7.66516566e-01 1.25083911e+00 -3.45110111e-02 4.80079591e-01 3.72232139e-01 3.14469486e-02 2.31537253e-01 -6.56188548e-01 1.18391335e-01 -3.47231299e-01 2.18515381e-01 5.78939438e-01 2.62632519e-01 -1.88056186e-01 -1.23304331e+00 3.19427550e-01 -2.61175215e-01 3.27014029e-01 5.80225587e-01 1.17226458e+00 -4.86680061e-01 -1.60772109e+00 -4.60629642e-01 4.43168968e-01 -8.59297931e-01 -9.18115303e-03 -5.29885948e-01 8.54225934e-01 4.30757552e-01 1.25550461e+00 -3.34664822e-01 1.75817803e-01 6.18527949e-01 3.24918896e-01 2.52322376e-01 -6.26467168e-01 -7.74279654e-01 -4.25137252e-01 5.85487247e-01 -2.29015321e-01 -8.46202612e-01 -7.02591479e-01 -1.40052760e+00 -1.35102659e-01 -1.05316870e-01 6.80338681e-01 7.88917720e-01 1.41436410e+00 1.58772320e-01 2.53592730e-01 1.22303687e-01 -1.77987322e-01 -3.37085992e-01 -1.00012326e+00 -5.59917748e-01 1.22624147e+00 -4.43907708e-01 -7.48741806e-01 -1.17896855e-01 4.35344547e-01]	[9.710095405578613, 9.19984245300293]
274a18ab-61fb-4888-9459-c3bc0f328ed9	pdnet-prior-model-guided-depth-enhanced	1803.08636	null	http://arxiv.org/abs/1803.08636v2	http://arxiv.org/pdf/1803.08636v2.pdf	PDNet: Prior-model Guided Depth-enhanced Network for Salient Object Detection	Fully convolutional neural networks (FCNs) have shown outstanding performance in many computer vision tasks including salient object detection. However, there still remains two issues needed to be addressed in deep learning based saliency detection. One is the lack of tremendous amount of annotated data to train a network. The other is the lack of robustness for extracting salient objects in images containing complex scenes. In this paper, we present a new architecture$ - $PDNet, a robust prior-model guided depth-enhanced network for RGB-D salient object detection. In contrast to existing works, in which RGB-D values of image pixels are fed directly to a network, the proposed architecture is composed of a master network for processing RGB values, and a sub-network making full use of depth cues and incorporate depth-based features into the master network. To overcome the limited size of the labeled RGB-D dataset for training, we employ a large conventional RGB dataset to pre-train the master network, which proves to contribute largely to the final accuracy. Extensive evaluations over five benchmark datasets demonstrate that our proposed method performs favorably against the state-of-the-art approaches.	['Thomas H. Li', 'Ge Li', 'Chunbiao Zhu', 'Xing Cai', 'Kan Huang']	2018-03-23	null	null	null	null	['rgb-d-salient-object-detection']	['computer-vision']	[ 5.24167061e-01 1.69502854e-01 -4.58883187e-05 -4.58480299e-01 -4.68349099e-01 5.04841171e-02 2.06129268e-01 -9.82470717e-03 -6.07982814e-01 6.37304604e-01 2.16485150e-02 -1.14630342e-01 3.00799280e-01 -6.35065854e-01 -7.17468143e-01 -7.53862202e-01 1.56166434e-01 -3.03736091e-01 1.02990878e+00 -3.96218300e-01 4.12686825e-01 5.80294132e-01 -1.89811003e+00 1.31087705e-01 8.24978352e-01 1.49877107e+00 7.01446593e-01 2.67743528e-01 -1.34841010e-01 9.62410808e-01 -2.60384709e-01 -3.93521562e-02 4.51894879e-01 -3.41648817e-01 -8.09852719e-01 1.09736636e-01 3.73724431e-01 -5.20320356e-01 -3.86640280e-01 1.18824220e+00 6.24528289e-01 1.41724318e-01 2.53710866e-01 -1.25235462e+00 -5.12068748e-01 3.72274041e-01 -7.47001469e-01 5.66409409e-01 1.81396231e-01 3.22028026e-02 8.13319266e-01 -1.11725247e+00 5.31322122e-01 1.08401370e+00 5.12214184e-01 5.93530118e-01 -6.67923093e-01 -5.42396426e-01 2.96697438e-01 2.22951636e-01 -1.04603815e+00 -2.58338988e-01 1.51228797e+00 -7.06479624e-02 8.00717175e-01 5.13802581e-02 9.15940046e-01 5.82918942e-01 -4.87665832e-02 1.25743556e+00 1.11255336e+00 -4.95647907e-01 3.56412053e-01 1.67341411e-01 -2.03654040e-02 8.90779793e-01 2.33150452e-01 1.35320753e-01 -6.43452585e-01 1.29784107e-01 1.01876044e+00 2.11481065e-01 -2.35359684e-01 -8.39148641e-01 -1.11089742e+00 7.26507008e-01 1.28550816e+00 2.57601053e-01 -5.23305714e-01 1.46756425e-01 2.04353899e-01 -2.95559555e-01 3.48537683e-01 3.72906029e-01 -3.30031246e-01 2.97915310e-01 -1.15062428e+00 8.55113566e-02 1.41300932e-01 9.05258298e-01 9.48837936e-01 2.82418072e-01 -3.90511705e-03 5.41152298e-01 4.17827219e-01 3.24539661e-01 4.92885381e-01 -6.53567135e-01 3.76243293e-01 1.11340547e+00 1.23386182e-01 -1.17304361e+00 -5.69169939e-01 -5.11352479e-01 -6.47594631e-01 5.61092794e-01 2.62327850e-01 9.28085446e-02 -1.05883682e+00 1.54134166e+00 4.86587822e-01 7.31967464e-02 1.22076966e-01 1.49488699e+00 1.22999895e+00 3.41943562e-01 8.76893774e-02 1.36117935e-01 1.08772933e+00 -1.15120053e+00 -3.75249714e-01 -6.48917675e-01 3.05770691e-02 -7.46258497e-01 1.09934556e+00 -2.35979632e-02 -1.30056405e+00 -6.83168471e-01 -1.29638958e+00 -4.97919708e-01 -5.39081812e-01 1.75120726e-01 8.98990154e-01 2.69118011e-01 -1.33496690e+00 3.36782277e-01 -7.03398705e-01 -3.34593654e-01 7.71684110e-01 4.72707421e-01 -2.28653342e-01 -1.72803625e-02 -1.16617560e+00 8.50038052e-01 7.09560633e-01 4.61836696e-01 -1.03273511e+00 -3.38255763e-01 -9.98233497e-01 -1.41723361e-03 3.22880298e-01 -4.88085747e-01 1.04828036e+00 -1.05601490e+00 -1.21604002e+00 9.17344749e-01 -2.05264196e-01 -3.30605179e-01 3.76774043e-01 -1.48804069e-01 1.48399975e-02 5.51476479e-01 1.60274386e-01 1.12696111e+00 9.14265454e-01 -1.55877447e+00 -1.00719059e+00 -3.33151042e-01 3.73512387e-01 3.23221743e-01 -3.09928775e-01 3.90071124e-02 -6.41884625e-01 -6.07632458e-01 6.00636363e-01 -7.26399720e-01 -3.91153961e-01 2.86529064e-01 -5.02077758e-01 -8.38964209e-02 1.04227793e+00 -4.05705988e-01 7.40116835e-01 -1.96868467e+00 3.00767291e-02 -1.77760959e-01 3.47832590e-01 5.17207921e-01 6.15181960e-02 -1.07498415e-01 -1.91522054e-02 -3.09244424e-01 -3.90477896e-01 -3.66173416e-01 -2.17089787e-01 -2.09191013e-02 -2.41739184e-01 4.90254670e-01 5.45977712e-01 1.05894375e+00 -1.00491965e+00 -7.75943935e-01 5.93747735e-01 4.76435393e-01 -2.86191672e-01 1.80013970e-01 -1.62654266e-01 2.28529096e-01 -6.44594967e-01 1.04304290e+00 7.47579396e-01 -2.55996287e-01 -4.86931503e-01 -1.86918542e-01 -3.08314711e-01 1.85169846e-01 -1.13236463e+00 1.92073369e+00 6.50771484e-02 6.65452778e-01 1.07783578e-01 -1.01821053e+00 1.03623021e+00 -4.85370494e-02 2.52736211e-01 -9.39183891e-01 3.61062765e-01 3.73673022e-01 -2.55702555e-01 -3.79956514e-01 7.32950032e-01 4.64256704e-02 7.99830034e-02 3.13066244e-01 1.59747358e-02 -1.94319233e-01 -5.12728170e-02 1.41358003e-01 6.94745243e-01 1.71165109e-01 7.07780942e-02 -2.63731629e-01 6.86153650e-01 3.15277964e-01 6.73131049e-01 5.60962319e-01 -6.72528327e-01 8.62871110e-01 1.97476342e-01 -6.42284036e-01 -8.95073354e-01 -9.09285069e-01 8.68330970e-02 1.15929258e+00 8.37804437e-01 1.86185926e-01 -5.88820994e-01 -6.16519690e-01 -1.68969586e-01 2.97465563e-01 -7.81004071e-01 -2.24415883e-01 -5.83534002e-01 -7.10876346e-01 2.91918814e-01 7.25713134e-01 1.16378438e+00 -1.41213322e+00 -1.30494416e+00 1.98073149e-01 -9.09669697e-02 -1.19175756e+00 -1.22914158e-01 3.64322007e-01 -1.08487678e+00 -1.12931740e+00 -7.92488277e-01 -1.21994519e+00 8.38345230e-01 8.34623098e-01 9.39398527e-01 1.02324016e-01 -4.62519109e-01 9.47589800e-02 -3.09202999e-01 -7.45974660e-01 2.10048407e-01 1.35992423e-01 -3.01029444e-01 -2.47013688e-01 3.80732685e-01 -2.94384927e-01 -1.17479634e+00 2.38383114e-01 -9.80734408e-01 3.21197242e-01 9.81586814e-01 6.41729891e-01 6.60908818e-01 -1.79682627e-01 5.04672825e-01 -3.84652466e-01 1.92077607e-01 -1.18112348e-01 -4.42694962e-01 1.89590454e-02 -3.25796366e-01 -1.68072641e-01 2.24573582e-01 -9.65988263e-03 -9.70548570e-01 5.00756562e-01 -1.43770009e-01 -2.66431212e-01 -1.39497593e-01 2.06039399e-01 -5.94669729e-02 -4.09725279e-01 4.24638003e-01 4.45437700e-01 2.64312290e-02 -3.46773177e-01 1.78456664e-01 4.54642326e-01 6.75567448e-01 -2.89092939e-02 8.16352487e-01 6.80262029e-01 1.63345486e-02 -6.93914771e-01 -1.12993956e+00 -6.39057577e-01 -8.08374226e-01 -2.75987923e-01 8.69750559e-01 -1.16230416e+00 -3.59749764e-01 5.65269053e-01 -1.00644433e+00 -2.70505846e-01 -1.81806594e-01 3.11201006e-01 -5.31249940e-01 6.06657378e-02 -4.14543867e-01 -7.55687654e-01 -6.38555348e-01 -1.10427773e+00 1.13677597e+00 7.44093299e-01 3.13110411e-01 -7.89735615e-01 -3.21649492e-01 3.08008134e-01 6.05880320e-01 5.47233343e-01 6.84867382e-01 -1.29043043e-01 -7.67455280e-01 -1.17203452e-01 -7.49152660e-01 3.00896466e-01 1.18622720e-01 -3.39298487e-01 -1.18441200e+00 -2.08035886e-01 9.32263881e-02 -5.36634445e-01 1.03239453e+00 5.41921556e-01 1.20247400e+00 1.40856236e-01 -3.87205303e-01 6.22709751e-01 1.64491868e+00 1.34162232e-03 4.36576784e-01 7.26851583e-01 7.87174046e-01 4.98512328e-01 7.62494147e-01 3.23834509e-01 4.63650405e-01 3.74996871e-01 1.06169176e+00 -8.33848894e-01 -3.43697071e-01 -1.73487291e-01 4.71988544e-02 3.46578687e-01 -9.39439908e-02 1.87166378e-01 -8.09601545e-01 8.47219765e-01 -1.88573802e+00 -7.93197513e-01 1.35411909e-02 1.76127338e+00 8.57851088e-01 4.02714193e-01 1.67764753e-01 3.19155753e-01 5.34907341e-01 2.83173263e-01 -8.84215653e-01 1.52782789e-02 -2.66541392e-01 3.96429338e-02 5.01326263e-01 1.03320628e-01 -1.34953666e+00 1.17015326e+00 6.04070663e+00 4.42927212e-01 -1.46162844e+00 -3.09798960e-02 5.39559960e-01 -1.05727538e-01 -8.24075639e-02 -2.78132170e-01 -7.82625258e-01 2.33504996e-01 2.72070616e-01 2.65591174e-01 -3.28792706e-02 1.17063439e+00 2.73770571e-01 -5.92307031e-01 -6.74351573e-01 9.69793379e-01 2.18171135e-01 -1.25393820e+00 -3.54736522e-02 -2.92462498e-01 8.71252596e-01 2.78283864e-01 1.91710338e-01 9.20245275e-02 5.55533022e-02 -8.34805667e-01 9.25233543e-01 2.30842322e-01 3.36804211e-01 -7.97004223e-01 9.77763593e-01 1.91886768e-01 -1.21036160e+00 -1.43081814e-01 -8.23955119e-01 -1.65871069e-01 -1.31773785e-01 6.99614286e-01 -6.97049320e-01 3.43576819e-01 1.11412382e+00 8.18784773e-01 -8.90229702e-01 1.21842206e+00 -3.93887162e-01 2.60075063e-01 -2.03682005e-01 -2.26018652e-01 6.49570763e-01 1.86806396e-01 2.82184035e-01 1.03700042e+00 4.18339809e-03 5.80568053e-02 1.04567334e-01 9.31756616e-01 3.28641273e-02 -1.07093543e-01 -5.10266662e-01 3.30953449e-01 2.07241803e-01 1.45468533e+00 -1.00642550e+00 -3.18954736e-01 -4.55318511e-01 1.02312970e+00 4.74778891e-01 3.12966585e-01 -7.59761751e-01 -5.76840758e-01 4.03303772e-01 5.59625365e-02 6.29315317e-01 -1.69373482e-01 -4.48175967e-01 -8.80804598e-01 4.54287082e-02 -4.18743610e-01 1.15817808e-01 -1.00744414e+00 -7.88375735e-01 7.69580722e-01 -1.53433740e-01 -1.24408031e+00 -1.66368075e-02 -7.03004181e-01 -6.35538638e-01 9.22516406e-01 -2.29997325e+00 -1.25262284e+00 -7.30125189e-01 8.40490699e-01 5.04947424e-01 9.10240635e-02 3.50031525e-01 -5.33406110e-03 -5.80859005e-01 1.68463871e-01 -1.94693759e-01 2.95636117e-01 3.49992037e-01 -1.28067946e+00 1.77966535e-01 1.15938020e+00 -1.73688561e-01 5.25424123e-01 6.47353113e-01 -3.84051979e-01 -1.38882816e+00 -1.21840537e+00 5.79387963e-01 -1.69511661e-01 2.64955789e-01 -3.92918229e-01 -7.47797668e-01 3.34781259e-01 2.23458707e-01 4.51973945e-01 3.65331024e-01 -4.19328362e-01 5.74920662e-02 -1.30330473e-01 -1.16090405e+00 4.64113444e-01 1.01050150e+00 -4.06566054e-01 -8.85154545e-01 -4.71340679e-02 8.23433399e-01 -4.80476737e-01 -4.00410205e-01 5.43856621e-01 3.85556906e-01 -1.25295937e+00 1.12977910e+00 -2.00089723e-01 7.08240092e-01 -6.14143312e-01 -1.01013608e-01 -1.01952171e+00 -9.48293954e-02 -1.40763909e-01 -3.75960581e-02 1.06042957e+00 1.93921998e-01 -2.66701013e-01 1.12323451e+00 6.46002889e-01 -3.70250672e-01 -1.00759935e+00 -7.97983706e-01 -2.27484494e-01 -3.59914243e-01 -2.87198156e-01 4.43335861e-01 6.69123769e-01 -2.16957048e-01 1.48858413e-01 -9.96839106e-02 1.52741477e-01 8.30652654e-01 5.47614098e-01 5.61094284e-01 -1.20586228e+00 2.85859227e-01 -4.64103758e-01 -4.93683457e-01 -1.04671347e+00 -1.45873740e-01 -6.13849938e-01 4.16393340e-01 -1.91977155e+00 2.26358205e-01 -5.72896183e-01 -6.63684070e-01 6.27336442e-01 -4.62636203e-01 5.96012175e-01 1.64372191e-01 1.56183556e-01 -8.69621813e-01 8.55969608e-01 1.45386302e+00 -1.00689016e-01 -3.81244838e-01 -1.21702403e-01 -8.79788160e-01 9.96825993e-01 9.29287076e-01 -2.79864579e-01 -4.74211663e-01 -4.20995295e-01 -1.04467668e-01 -3.10994238e-01 6.53211534e-01 -1.22935021e+00 4.47134554e-01 -3.08860615e-02 8.42426479e-01 -1.07044065e+00 3.50792617e-01 -8.51409137e-01 -5.83834112e-01 5.15760541e-01 -1.15604334e-01 -1.22099392e-01 4.33212280e-01 3.97417635e-01 -4.24037069e-01 -1.26607433e-01 8.82451355e-01 -3.71077895e-01 -1.41161370e+00 3.10274512e-01 3.85436267e-02 -7.81327114e-02 1.08539557e+00 -5.87671220e-01 -3.42051834e-01 4.10820059e-02 -2.96114445e-01 1.57185137e-01 5.72803020e-01 4.32126135e-01 1.12872815e+00 -1.20045722e+00 -2.47169197e-01 3.39088231e-01 9.47985426e-02 5.38296759e-01 1.64443776e-01 7.02254176e-01 -6.81715608e-01 4.59910959e-01 -6.97875500e-01 -6.82171762e-01 -9.33662832e-01 6.20802879e-01 2.02025950e-01 1.23118594e-01 -5.08773625e-01 1.06118906e+00 1.90743089e-01 -1.95898771e-01 4.05268252e-01 -5.28349161e-01 -3.68670195e-01 -3.89255658e-02 5.23001432e-01 6.69165105e-02 -3.42914760e-02 -7.66558051e-01 -5.93044221e-01 6.24684155e-01 -5.64104021e-02 1.36981979e-01 1.57210696e+00 -3.17833513e-01 -1.57455608e-01 2.49650016e-01 1.06071031e+00 -4.93877441e-01 -1.74214649e+00 -4.50728685e-01 -5.40274791e-02 -4.92217690e-01 4.35134351e-01 -5.71596324e-01 -1.42711461e+00 9.85973299e-01 8.29206526e-01 -5.37471101e-02 1.40607643e+00 9.28479712e-03 8.10939610e-01 2.27863908e-01 4.72556591e-01 -1.21583116e+00 4.08651173e-01 2.86289990e-01 9.10494030e-01 -1.79594994e+00 1.45894021e-01 -4.52050686e-01 -6.41288817e-01 9.16118085e-01 9.12277222e-01 -3.10787082e-01 5.41808367e-01 -1.13350362e-01 2.40360200e-01 -2.92983741e-01 -2.54443794e-01 -6.55238569e-01 2.99642414e-01 6.21443391e-01 1.78488538e-01 -2.95749873e-01 1.11832879e-01 3.09578925e-01 -9.75784957e-02 1.30623057e-02 3.51569414e-01 1.31006634e+00 -8.93690348e-01 -5.04384160e-01 -2.91912258e-01 2.35488430e-01 -3.69377315e-01 -9.32919905e-02 -4.18001205e-01 8.68269801e-01 1.70433104e-01 9.51938152e-01 -6.11597784e-02 -2.30007350e-01 2.50357449e-01 -2.05018878e-01 1.92918390e-01 -5.42593539e-01 -4.71458793e-01 -7.35268667e-02 -4.27770734e-01 -7.19417334e-01 -1.02048707e+00 -4.65742797e-01 -1.39840329e+00 7.95438215e-02 -2.48185799e-01 -2.07979217e-01 7.41121650e-01 8.46815825e-01 2.09860280e-01 6.15631878e-01 5.74877322e-01 -1.37344491e+00 -1.59473151e-01 -8.00237000e-01 -5.93026936e-01 2.51124740e-01 6.85689390e-01 -8.91983271e-01 -2.37832040e-01 -7.54875988e-02]	[9.695270538330078, -0.7330524325370789]
eb5ec060-2f81-457c-b24f-24e871144f58	gpu-activity-prediction-using-representation	1703.09146	null	http://arxiv.org/abs/1703.09146v1	http://arxiv.org/pdf/1703.09146v1.pdf	GPU Activity Prediction using Representation Learning	GPU activity prediction is an important and complex problem. This is due to the high level of contention among thousands of parallel threads. This problem was mostly addressed using heuristics. We propose a representation learning approach to address this problem. We model any performance metric as a temporal function of the executed instructions with the intuition that the flow of instructions can be identified as distinct activities of the code. Our experiments show high accuracy and non-trivial predictive power of representation learning on a benchmark.	['Sek Chai', 'Mohamed Amer', 'David Zhang', 'Timothy Shields', 'Aswin Raghavan']	2017-03-27	null	null	null	null	['activity-prediction', 'activity-prediction']	['computer-vision', 'time-series']	[ 2.06558462e-02 -4.99960572e-01 -6.93580568e-01 -2.03308105e-01 -5.35767794e-01 -5.08730650e-01 6.24965668e-01 4.79810178e-01 -7.79573917e-02 6.68936789e-01 6.17154717e-01 -4.63992625e-01 5.72896563e-02 -8.03108156e-01 -7.10502386e-01 -5.87337375e-01 -4.94203538e-01 3.51556569e-01 5.64404190e-01 -8.09521675e-02 8.18999827e-01 5.31298399e-01 -1.76434362e+00 8.56419384e-01 3.95152599e-01 8.41624558e-01 7.67699815e-03 1.09265924e+00 -2.74647415e-01 1.37872159e+00 -7.14957297e-01 2.89416075e-01 1.37427300e-01 -5.42066276e-01 -7.87623286e-01 -2.62523681e-01 1.14122801e-01 -1.10392598e-02 -2.22651288e-01 7.17313409e-01 -1.13606684e-01 -8.08975939e-03 6.40437186e-01 -1.35384107e+00 4.85508412e-01 3.12491000e-01 -1.01468861e+00 8.44441891e-01 3.27590972e-01 -7.27963224e-02 1.05332541e+00 -3.63860697e-01 5.01690626e-01 7.82367826e-01 4.78861123e-01 -2.77614947e-02 -1.42120504e+00 -4.14971471e-01 -5.10850959e-02 3.57507735e-01 -1.06385481e+00 -1.53506175e-01 5.41097999e-01 -8.63293886e-01 1.21148252e+00 4.64669973e-01 7.02760398e-01 9.66123104e-01 7.45975852e-01 8.24057460e-01 1.19381690e+00 -5.16627669e-01 4.74062622e-01 -3.73554438e-01 8.18374574e-01 8.99318814e-01 4.47687358e-01 3.29760969e-01 -7.68957973e-01 -5.60321152e-01 7.25046098e-01 3.69745046e-02 -1.38241192e-02 -1.76848397e-01 -1.12052786e+00 9.85307574e-01 1.62561059e-01 3.24116617e-01 -1.71120614e-01 6.94596529e-01 8.51266980e-01 2.35323757e-01 1.57973856e-01 8.87439311e-01 -7.73176968e-01 -9.31077003e-01 -1.10966909e+00 2.74081022e-01 1.22585869e+00 6.12741351e-01 8.24309647e-01 4.22798190e-03 -1.78493723e-01 2.33277485e-01 -5.90214990e-02 -1.28384739e-01 5.95720708e-01 -7.57082164e-01 1.57892480e-01 8.59742284e-01 -1.55350313e-01 -8.11843395e-01 -5.60901642e-01 -4.53315169e-01 -4.70396668e-01 2.20313951e-01 4.79227245e-01 4.24993262e-02 -5.27945817e-01 1.38466251e+00 1.19698040e-01 3.73107374e-01 -3.46514165e-01 3.02384406e-01 1.06134810e-01 1.03510582e+00 5.82987033e-02 -3.49368930e-01 1.31466687e+00 -1.01547992e+00 -3.86619002e-01 -3.11994135e-01 9.99116123e-01 -7.12089479e-01 6.29470050e-01 5.49968302e-01 -6.55095160e-01 -4.88972723e-01 -1.28214526e+00 2.38533989e-01 -1.14834689e-01 -1.20231427e-01 1.30458403e+00 6.89448655e-01 -5.13134599e-01 6.88930929e-01 -1.48552787e+00 8.72372985e-02 1.93187267e-01 4.49855417e-01 1.29829288e-01 8.86359066e-02 -2.59446174e-01 5.69844484e-01 5.81875205e-01 -4.76658404e-01 -5.67033529e-01 -8.46506715e-01 -6.44263387e-01 1.67546421e-01 3.69991332e-01 -3.63226444e-01 1.30449128e+00 -1.04582763e+00 -1.23037660e+00 5.57540476e-01 -4.39943463e-01 -4.11130309e-01 2.99631149e-01 -4.13413554e-01 -3.05731565e-01 -4.76948678e-01 -8.24835524e-02 -3.45772982e-01 6.28304064e-01 -1.02884758e+00 -8.64241838e-01 -4.43658382e-01 -1.88618734e-01 -3.16079646e-01 -1.11537434e-01 -6.72988817e-02 -3.22023600e-01 -5.46087623e-01 -1.34547427e-01 -1.04647982e+00 -3.33470792e-01 -6.78439081e-01 -1.21104248e-01 -1.26091927e-01 6.27470255e-01 -3.58689755e-01 1.71961510e+00 -2.07559371e+00 7.35930428e-02 1.55658960e-01 3.22215676e-01 -2.24420607e-01 1.22450978e-01 4.61502105e-01 1.36171971e-02 -6.54466972e-02 2.72416472e-01 4.41845834e-01 -2.15550110e-01 3.48447174e-01 -6.26632452e-01 6.13809466e-01 8.57721865e-02 6.00920916e-01 -1.07149374e+00 -3.61017972e-01 -1.70053259e-01 1.66484281e-01 -5.60491681e-01 6.08709931e-01 -2.85817206e-01 2.69614100e-01 -6.85114980e-01 5.78498423e-01 2.05252543e-01 -4.51012194e-01 3.60639274e-01 -1.10398941e-02 -4.75180417e-01 4.73746181e-01 -9.17048514e-01 1.46955371e+00 -6.94743216e-01 8.57536376e-01 -5.84408224e-01 -9.48350787e-01 7.11526036e-01 -1.09810546e-01 7.22569704e-01 -8.62558365e-01 1.83626190e-01 1.99716389e-01 3.50017160e-01 -2.32321024e-01 6.88227117e-01 2.12920755e-01 -5.27166009e-01 7.07439721e-01 -1.91248000e-01 8.29939321e-02 3.96038830e-01 -1.47850052e-01 1.73691726e+00 1.20011397e-01 8.56818616e-01 -6.64588213e-01 3.33555609e-01 3.74577492e-01 6.44168317e-01 7.75183320e-01 5.76703670e-03 -2.47146953e-02 1.22271740e+00 -8.59357595e-01 -9.62285757e-01 -8.56019258e-01 -3.00906878e-02 1.71550417e+00 -1.12701736e-01 -1.02703404e+00 -5.29938400e-01 -5.14237285e-01 -1.01803079e-01 6.51133418e-01 -1.11993504e+00 -2.44346380e-01 -1.12388718e+00 -9.21556652e-01 9.80434716e-02 7.96935558e-01 -3.03246677e-02 -6.45418167e-01 -1.33724499e+00 4.24464613e-01 1.76688865e-01 -9.25884604e-01 -3.43124837e-01 7.99548209e-01 -1.19797158e+00 -1.23477805e+00 9.17228907e-02 -3.76827270e-01 4.09664601e-01 1.81955010e-01 1.71403933e+00 2.16998771e-01 -6.59493208e-01 1.56306308e-02 -3.34890515e-01 -4.23678160e-01 -6.92484975e-01 3.07309330e-01 -3.50651234e-01 -1.72333941e-01 3.80723506e-01 -5.66799819e-01 -5.60779870e-01 3.28383356e-01 -5.55979371e-01 9.29956064e-02 5.74976802e-01 1.07040155e+00 7.79699385e-01 2.35056579e-02 -2.99464911e-01 -1.03387976e+00 5.06941319e-01 -6.46390736e-01 -9.17171180e-01 1.40190393e-01 -5.30064821e-01 6.51019394e-01 7.84927189e-01 -6.10093057e-01 -5.53268135e-01 4.61642861e-01 3.03311765e-01 -1.17036499e-01 5.74040450e-02 2.50606179e-01 1.72941998e-01 6.38858676e-02 6.22522235e-01 2.16639996e-01 -4.07206416e-01 -1.69745669e-01 -5.16842678e-02 9.89331678e-02 3.73510748e-01 -1.13996732e+00 9.83303264e-02 2.57220030e-01 5.88911355e-01 -8.43887866e-01 -7.95169055e-01 -5.40530860e-01 -5.72675943e-01 -2.67370045e-01 5.66442013e-01 -5.81114352e-01 -8.24331284e-01 -3.55004556e-02 -1.20315373e+00 -3.96087974e-01 -1.19195178e-01 2.05321357e-01 -6.75914407e-01 1.74850538e-01 -5.38468122e-01 -6.71547592e-01 -2.41697371e-01 -1.26274443e+00 9.46580410e-01 9.53949243e-02 -6.54600203e-01 -8.48888814e-01 6.59172475e-01 -6.58955574e-02 2.99721569e-01 6.55633509e-01 1.20802248e+00 -5.63058078e-01 -6.92207098e-01 -1.62507609e-01 -2.13897705e-01 -2.98107386e-01 1.16949290e-01 2.13232607e-01 -7.50670671e-01 -2.99917191e-01 2.31039435e-01 -5.26526496e-02 6.99868202e-01 2.86988437e-01 1.56493258e+00 -1.94596663e-01 -5.23943603e-01 5.99247515e-01 1.79898596e+00 1.62908167e-01 5.58656573e-01 3.76981646e-01 7.48153210e-01 4.23729628e-01 5.21153092e-01 7.35953748e-01 -3.37250024e-01 8.04762661e-01 2.19068170e-01 2.66870171e-01 1.89955205e-01 -2.19345950e-02 2.79838026e-01 1.04495978e+00 -3.48603815e-01 -4.20766622e-02 -1.45694864e+00 3.28803122e-01 -1.80915463e+00 -9.69794154e-01 -6.06494725e-01 2.24959064e+00 4.87050116e-01 4.73635614e-01 3.40966165e-01 2.56280988e-01 1.97704509e-01 1.68730840e-01 -4.46984261e-01 -1.00909555e+00 4.60900515e-01 5.40780783e-01 8.41322422e-01 8.21411312e-02 -8.59570801e-01 6.49493635e-01 7.67779589e+00 7.94527054e-01 -1.10257709e+00 -8.02255888e-03 7.33891189e-01 2.36712247e-02 2.38730922e-01 3.75825018e-02 -8.67447615e-01 6.16744459e-01 1.50495803e+00 -6.17769241e-01 2.27022231e-01 8.40007722e-01 -1.22256935e-01 -5.26071310e-01 -1.62524188e+00 1.07315421e+00 -1.07400469e-01 -1.42427588e+00 -9.77556333e-02 4.19850796e-01 8.27227354e-01 1.81054026e-01 -2.69490987e-01 3.31953257e-01 5.11109710e-01 -1.10425770e+00 4.37690526e-01 3.17994863e-01 3.03177446e-01 -8.86816859e-01 5.97536683e-01 3.28055322e-01 -1.46217251e+00 -4.03413385e-01 -7.86703639e-03 -5.05458772e-01 -1.90715313e-01 5.21430790e-01 -6.70184016e-01 6.77018985e-02 4.41877872e-01 6.17299080e-01 -6.91398203e-01 1.02763963e+00 1.20787993e-01 9.83240366e-01 -1.32058844e-01 -2.33660668e-01 2.14001790e-01 1.54548148e-02 -1.23731464e-01 1.56965625e+00 2.69015670e-01 -1.00835778e-01 3.51435840e-01 5.00786781e-01 9.63222310e-02 9.14871618e-02 -5.85867643e-01 -9.83679816e-02 -4.67641978e-03 9.24697399e-01 -1.15316999e+00 -3.56190622e-01 -7.25237966e-01 7.30006635e-01 4.61988956e-01 -2.14191511e-01 -9.08311844e-01 1.85719743e-01 9.45716560e-01 2.54274815e-01 1.91923961e-01 -5.78057408e-01 -6.27530217e-01 -9.02576625e-01 -2.52992153e-01 -7.34296858e-01 3.83879900e-01 -4.36529927e-02 -1.01785147e+00 5.37982345e-01 -1.84965998e-01 -1.28051019e+00 -4.43696141e-01 -8.20128620e-01 -6.78240001e-01 4.72842395e-01 -1.08612752e+00 -4.23903853e-01 -4.14021999e-01 3.30861121e-01 8.38591635e-01 -1.04166143e-01 8.07597220e-01 -1.48792922e-01 -5.37726641e-01 3.63011181e-01 3.05273801e-01 1.37456320e-02 2.31000274e-01 -1.33862782e+00 4.70153451e-01 5.27276695e-01 2.30002820e-01 3.06446105e-01 9.54748213e-01 -5.36583185e-01 -1.95570326e+00 -8.96232426e-01 3.87752265e-01 -7.46705532e-01 1.19608474e+00 -4.19526339e-01 -1.00344825e+00 6.17837369e-01 -1.57055017e-02 2.79868692e-01 1.14997876e+00 3.94315004e-01 -5.00505805e-01 1.26563817e-01 -2.86509335e-01 2.79262900e-01 8.57224822e-01 -4.52793270e-01 -5.39687097e-01 3.06123883e-01 2.35048398e-01 -5.14638305e-01 -8.68517160e-01 1.77212119e-01 5.63961804e-01 -9.63428795e-01 7.20752418e-01 -6.50970042e-01 7.75394022e-01 1.48649395e-01 -2.78494179e-01 -9.30212080e-01 -4.07319248e-01 -3.85772735e-01 -7.85916448e-01 5.83493948e-01 1.66573450e-01 -2.26421550e-01 1.10929728e+00 2.97849923e-01 1.22438990e-01 -9.35448349e-01 -7.87435591e-01 -1.03295052e+00 4.05568890e-02 -2.71306455e-01 4.55043942e-01 9.81013179e-01 1.13773443e-01 5.14171481e-01 -3.59665394e-01 -9.60685909e-02 4.73321378e-01 6.67077482e-01 8.41224432e-01 -1.37756193e+00 -8.34807277e-01 -7.24417925e-01 -7.61944711e-01 -8.59658957e-01 1.23228937e-01 -8.66117835e-01 -8.43741372e-02 -8.76446664e-01 4.79544431e-01 -2.07440332e-01 -3.95419627e-01 1.28989413e-01 2.28325408e-02 -2.42585152e-01 -1.96392789e-01 1.88838080e-01 -6.62706971e-01 7.58476853e-02 7.81673074e-01 -4.49460968e-02 -2.24736124e-01 1.81917474e-02 -1.54969394e-01 5.61628222e-01 9.03228283e-01 -6.60595238e-01 -3.15541178e-01 -3.83928120e-01 7.07338929e-01 3.65957648e-01 -8.39517862e-02 -1.22193730e+00 3.73026431e-02 -4.58460748e-01 2.95356721e-01 -5.12364805e-01 2.13430412e-02 -7.21855342e-01 2.43417755e-01 9.21477795e-01 -1.87250361e-01 6.77827537e-01 5.05042017e-01 7.37578809e-01 -1.91673085e-01 7.41065387e-03 6.17164552e-01 -6.26941547e-02 -9.52778041e-01 -2.41106469e-02 -5.07607818e-01 1.12861626e-01 1.13638484e+00 -1.80921227e-01 -2.21287623e-01 2.44356290e-01 -2.27956787e-01 7.74022145e-03 4.94289786e-01 5.05453110e-01 1.57977924e-01 -9.41005409e-01 -5.79290926e-01 1.37315184e-01 2.93003738e-01 -7.14245260e-01 -1.06428608e-01 5.71638882e-01 -7.50748336e-01 3.59598190e-01 -5.64304650e-01 -6.00362003e-01 -1.33251953e+00 7.56817162e-01 1.12950228e-01 -7.54846811e-01 -5.13856411e-01 3.37226152e-01 6.66182563e-02 4.26278114e-01 -2.18695328e-01 -3.19357842e-01 7.10191485e-03 2.35392768e-02 7.63718367e-01 7.73689747e-01 1.86646327e-01 -4.42145377e-01 -2.93816566e-01 4.37225997e-01 -3.17007780e-01 3.54708552e-01 1.34746253e+00 5.91003597e-01 -2.78878033e-01 8.34039688e-01 1.37974286e+00 -1.09386839e-01 -1.11832130e+00 6.89815581e-02 5.91907144e-01 -7.14920700e-01 1.97294936e-01 -5.29607177e-01 -8.85393500e-01 7.43888676e-01 6.63186014e-01 2.57105887e-01 9.26512718e-01 6.54946044e-02 6.23009861e-01 1.92202181e-01 7.56158471e-01 -1.10126650e+00 2.27194309e-01 8.00569236e-01 3.32618058e-01 -1.13172400e+00 4.12367970e-01 -4.82757658e-01 1.81198165e-01 1.37125635e+00 8.01611245e-01 -3.08154017e-01 6.44422472e-01 8.61812770e-01 -5.52692771e-01 -2.70163924e-01 -9.81266439e-01 2.38045782e-01 1.20318606e-01 1.85353205e-01 8.43575239e-01 2.97100455e-01 -5.74193358e-01 4.55494046e-01 -1.72029108e-01 3.29221152e-02 7.25273073e-01 1.51385677e+00 -5.00410438e-01 -1.20387697e+00 -3.78695607e-01 7.75188148e-01 -4.67300445e-01 1.43333167e-01 -2.67170463e-02 7.52430737e-01 -1.70488343e-01 4.47738081e-01 3.40305597e-01 -4.73704755e-01 1.19802572e-01 -3.98807153e-02 6.83688998e-01 -6.36650383e-01 -7.56262183e-01 -2.34235287e-01 -1.13681048e-01 -9.87510145e-01 7.37144351e-02 -6.05400264e-01 -1.14320195e+00 -4.84295428e-01 9.32606012e-02 1.82101965e-01 6.83168769e-01 6.25160992e-01 2.59897560e-01 9.05191660e-01 8.15970480e-01 -7.97339201e-01 -2.73140311e-01 -4.16017443e-01 -6.88046396e-01 4.42691952e-01 2.04830453e-01 -7.37349570e-01 -2.37915441e-01 -2.81460509e-02]	[7.551483631134033, 7.512197971343994]
13f1ee32-1856-45df-b516-550b30fbaa01	legendre-memory-units-continuous-time	null	null	http://papers.nips.cc/paper/9689-legendre-memory-units-continuous-time-representation-in-recurrent-neural-networks	http://papers.nips.cc/paper/9689-legendre-memory-units-continuous-time-representation-in-recurrent-neural-networks.pdf	Legendre Memory Units: Continuous-Time Representation in Recurrent Neural Networks	We propose a novel memory cell for recurrent neural networks that dynamically maintains information across long windows of time using relatively few resources. The Legendre Memory Unit~(LMU) is mathematically derived to orthogonalize its continuous-time history -- doing so by solving $d$ coupled ordinary differential equations~(ODEs), whose phase space linearly maps onto sliding windows of time via the Legendre polynomials up to degree $d - 1$. Backpropagation across LMUs outperforms equivalently-sized LSTMs on a chaotic time-series prediction task, improves memory capacity by two orders of magnitude, and significantly reduces training and inference times. LMUs can efficiently handle temporal dependencies spanning $100\text{,}000$ time-steps, converge rapidly, and use few internal state-variables to learn complex functions spanning long windows of time -- exceeding state-of-the-art performance among RNNs on permuted sequential MNIST. These results are due to the network's disposition to learn scale-invariant features independently of step size. Backpropagation through the ODE solver allows each layer to adapt its internal time-step, enabling the network to learn task-relevant time-scales. We demonstrate that LMU memory cells can be implemented using $m$ recurrently-connected Poisson spiking neurons, $\mathcal{O}( m )$ time and memory, with error scaling as $\mathcal{O}( d / \sqrt{m} )$. We discuss implementations of LMUs on analog and digital neuromorphic hardware.	['Ivana Kajić', 'Chris Eliasmith', 'Aaron Voelker']	2019-12-01	null	null	null	neurips-2019-12	['sequential-image-classification']	['computer-vision']	[ 2.24143609e-01 -3.80594164e-01 3.36774617e-01 -4.97711040e-02 -3.18741947e-01 -5.35552025e-01 1.51207939e-01 -2.59814024e-01 -1.06824982e+00 9.95903254e-01 -4.49911922e-01 -2.98558205e-01 -1.37823313e-01 -8.18177342e-01 -9.05144334e-01 -8.69921029e-01 -6.17102265e-01 2.05347463e-01 3.30652714e-01 -1.87549815e-01 4.01741534e-01 6.57415986e-01 -1.40483892e+00 1.04100458e-01 2.82048821e-01 1.22148764e+00 4.54798311e-01 7.52120614e-01 7.07103834e-02 7.47957766e-01 -5.63792706e-01 2.66294032e-01 1.53880954e-01 -4.04016256e-01 -3.89856011e-01 -7.04622984e-01 8.73050988e-02 1.34239897e-01 -9.92843568e-01 8.15984190e-01 5.38438678e-01 3.50846708e-01 5.84613919e-01 -7.75413454e-01 -7.47711003e-01 5.85776448e-01 6.00760467e-02 8.14283550e-01 -3.03338230e-01 3.47266823e-01 5.89199662e-01 -6.94918990e-01 5.62694192e-01 7.39881337e-01 9.90075648e-01 7.40005910e-01 -1.56221604e+00 -1.07896984e+00 1.44840181e-01 -1.16664968e-01 -1.39856064e+00 -3.25881481e-01 4.73343581e-01 -1.06932789e-01 2.09682918e+00 9.50558782e-02 1.00256813e+00 7.95857012e-01 9.89226401e-01 2.53546983e-01 1.20949006e+00 -4.28102091e-02 5.45301914e-01 -5.61602294e-01 3.91849190e-01 7.26542413e-01 5.12038264e-03 8.86543915e-02 -7.25770533e-01 1.02617212e-01 1.27979457e+00 1.82812363e-01 -3.57917517e-01 3.73752683e-01 -1.11401284e+00 3.74838352e-01 5.52332699e-01 3.65153491e-01 -1.94347680e-01 1.01371121e+00 2.50223815e-01 6.95947111e-01 6.67610690e-02 7.90465176e-01 -4.82493728e-01 -4.34103489e-01 -9.74330127e-01 7.33397305e-02 8.80221605e-01 9.17105973e-01 6.75213873e-01 7.37298548e-01 1.76560283e-01 6.57080829e-01 -6.92759231e-02 8.48195136e-01 9.57688689e-01 -1.08883989e+00 1.03734583e-01 4.54053789e-01 -2.19226822e-01 -5.49636781e-01 -6.87424183e-01 -5.36893785e-01 -1.22724092e+00 2.41771668e-01 1.89679489e-01 -3.05096865e-01 -1.17934191e+00 2.01975393e+00 -6.21242404e-01 1.70139417e-01 1.96936205e-02 3.52186114e-01 1.47994429e-01 1.03506625e+00 -1.70134172e-01 -6.06248260e-01 1.12690639e+00 -3.13226789e-01 -5.06393850e-01 -4.39148158e-01 3.21901381e-01 -2.82036275e-01 7.08489001e-01 1.69867799e-01 -1.62936115e+00 -3.63290280e-01 -1.14283478e+00 -5.64799309e-02 -5.31256080e-01 -2.40099311e-01 4.85665500e-01 1.36740267e-01 -1.61071992e+00 1.13326824e+00 -1.24726224e+00 1.20386705e-01 3.68352592e-01 1.04369843e+00 2.34649852e-02 5.15524507e-01 -1.09677243e+00 8.61136615e-01 2.27006659e-01 2.18043119e-01 -7.89018333e-01 -7.47772992e-01 -4.73301649e-01 1.12512849e-01 -5.58667660e-01 -6.29043937e-01 1.05994749e+00 -4.36531544e-01 -1.70003581e+00 6.47709250e-01 -5.35373092e-01 -1.07564223e+00 -8.03407878e-02 2.74718106e-01 -5.88951409e-01 6.90736808e-03 -2.05932125e-01 7.56778777e-01 8.10282171e-01 -4.48408186e-01 -1.21767938e-01 -3.44894052e-01 -5.49536705e-01 -1.96278706e-01 -4.34288085e-01 -3.01440895e-01 -1.14691995e-01 -8.04917812e-01 5.79892516e-01 -1.20753372e+00 -3.63611251e-01 -1.33390114e-01 9.45130140e-02 1.22388020e-01 6.61041141e-01 -2.50407398e-01 1.41064596e+00 -2.10266948e+00 2.67924726e-01 1.94269270e-01 1.90215752e-01 1.81612179e-01 1.37042515e-02 3.58634084e-01 -9.20441076e-02 -7.12213963e-02 -5.11677504e-01 -1.78832471e-01 -2.48901412e-01 3.05072337e-01 -7.88449049e-01 3.40423793e-01 2.37283438e-01 1.21278167e+00 -5.40983438e-01 1.59592450e-01 -2.83384919e-01 6.49172843e-01 -3.58477741e-01 -3.10423583e-01 -3.37384731e-01 2.11282477e-01 -1.17440429e-02 3.44566166e-01 2.89901406e-01 -5.58690846e-01 8.60506073e-02 -5.01903798e-03 -5.70717633e-01 4.46294188e-01 -8.82435441e-01 1.48416770e+00 -4.03058350e-01 1.09161937e+00 -3.94477218e-01 -1.01342893e+00 1.02879596e+00 2.50878721e-01 4.18543071e-01 -1.16000605e+00 2.97132492e-01 6.19279742e-01 8.94879326e-02 2.66173221e-02 2.06540331e-01 -2.57822245e-01 -1.13744475e-01 7.02364862e-01 2.72658229e-01 -1.99296594e-01 9.18688998e-02 -2.85317600e-01 1.53687167e+00 -2.88397551e-01 -1.08595192e-01 -4.79344219e-01 -1.73593462e-02 -2.68026203e-01 4.57585841e-01 8.58503520e-01 -7.38386363e-02 2.25235030e-01 3.80509645e-01 -6.01820111e-01 -1.21143985e+00 -1.30750191e+00 -4.27067071e-01 1.01938939e+00 -1.00269109e-01 2.85999123e-02 -5.81407487e-01 5.52514911e-01 -1.04110330e-01 5.70018172e-01 -5.92517138e-01 -3.25820595e-01 -1.17627025e+00 -9.36916530e-01 1.06217158e+00 7.36065447e-01 7.19557226e-01 -1.57062125e+00 -1.24555063e+00 6.61100507e-01 4.42491025e-01 -5.89464128e-01 -3.95861745e-01 1.20255721e+00 -1.34848452e+00 -2.56525040e-01 -7.84451783e-01 -1.17646897e+00 6.86610043e-01 -2.28012353e-01 7.80104697e-01 -5.19836068e-01 -5.03282070e-01 1.42864168e-01 3.62347662e-01 -6.00018799e-02 5.13659455e-02 1.06323779e-01 4.89784300e-01 -6.66123033e-01 2.64383346e-01 -1.46753550e+00 -7.97890544e-01 1.90848067e-01 -7.32854903e-01 -1.14219397e-01 4.93789434e-01 9.14935529e-01 1.03679013e+00 -2.13276088e-01 5.79662383e-01 -4.73435283e-01 5.67168176e-01 -1.00108050e-01 -8.00322533e-01 -5.95458522e-02 -6.26129150e-01 5.94063461e-01 9.69562769e-01 -1.02852619e+00 -3.35373670e-01 -2.35936180e-01 1.63526535e-01 -5.24500787e-01 5.56598425e-01 3.58208060e-01 8.09889734e-01 -4.82433766e-01 7.95974553e-01 1.00201333e+00 4.04411368e-02 4.69569229e-02 -4.61405935e-03 -8.28190818e-02 6.03582323e-01 -4.68390286e-01 4.25654054e-01 4.12921220e-01 1.17904544e-01 -7.65783787e-01 -5.18226884e-02 1.88439161e-01 -2.74787098e-01 6.91409484e-02 5.53236961e-01 -8.13302279e-01 -1.30355632e+00 6.93645835e-01 -8.96602571e-01 -8.83258760e-01 -4.84880239e-01 2.75824606e-01 -7.69832850e-01 -4.40865457e-01 -1.26494586e+00 -8.51303041e-01 -7.01679289e-01 -8.96155596e-01 2.80072421e-01 3.08150113e-01 -3.43278289e-01 -8.64253938e-01 1.51893541e-01 -7.93496609e-01 8.80942643e-01 -1.12183057e-01 1.07369030e+00 -1.76502571e-01 -8.50233078e-01 -1.29467726e-01 -7.66432062e-02 2.04033986e-01 -3.03520143e-01 -2.81877548e-01 -9.31700706e-01 -5.07692456e-01 3.73310417e-01 -1.69864237e-01 1.23916495e+00 6.41726553e-01 1.16585970e+00 -4.87444550e-01 -4.21250403e-01 7.38396466e-01 1.35955620e+00 5.73447883e-01 7.49910772e-01 6.28654584e-02 3.17289174e-01 -1.58972532e-01 -3.74686003e-01 3.47160250e-01 1.55901089e-01 4.59297784e-02 4.66041081e-03 4.74758446e-01 5.47945872e-02 3.30951177e-02 6.89987838e-01 1.50081992e+00 -1.89927876e-01 1.80457637e-01 -8.57363462e-01 4.36840504e-01 -1.53615797e+00 -1.20904756e+00 3.72996509e-01 2.31947470e+00 1.09164751e+00 5.27664661e-01 -2.23699898e-01 6.10087216e-02 4.07606691e-01 4.08469066e-02 -1.17533529e+00 -5.57931244e-01 -3.17359209e-01 1.02321243e+00 9.00218248e-01 3.49517614e-01 -4.84745413e-01 9.94882226e-01 6.96401215e+00 6.16952062e-01 -1.76336706e+00 -1.02076389e-01 5.31650484e-01 -5.69616497e-01 -1.60753384e-01 -2.48638660e-01 -9.05414999e-01 5.97414672e-01 1.47606194e+00 -1.87652811e-01 1.04286349e+00 4.71449107e-01 -5.30107319e-02 5.24776019e-02 -1.13792896e+00 1.22710836e+00 -2.82001942e-01 -1.71803415e+00 -1.74241483e-01 9.04183164e-02 8.27857435e-01 5.83639741e-01 6.88592851e-01 4.50008303e-01 2.64740467e-01 -1.31640232e+00 5.82861483e-01 7.46332824e-01 9.66894448e-01 -5.97120404e-01 -1.83595195e-02 3.00793618e-01 -1.38283050e+00 -4.07397956e-01 -6.43119216e-01 -4.41238999e-01 -9.26955938e-02 4.63315517e-01 -3.21993440e-01 -5.72357297e-01 9.69937742e-01 5.63243091e-01 -6.45446926e-02 6.37420833e-01 4.46367294e-01 4.33813125e-01 -7.71974623e-01 -6.66648209e-01 1.57615855e-01 -1.15818769e-01 2.76631027e-01 1.14121413e+00 7.47488678e-01 5.88769615e-01 -5.35737455e-01 1.14064240e+00 -2.17532903e-01 -6.28914833e-01 -5.13391554e-01 -3.63607913e-01 8.41054142e-01 7.65548348e-01 -8.81579876e-01 -1.08343944e-01 2.42098287e-01 1.12616754e+00 5.33791661e-01 6.13322377e-01 -5.51262677e-01 -6.74455047e-01 5.95855117e-01 -5.14694664e-04 5.94782948e-01 -7.98747182e-01 -5.46338737e-01 -9.46919918e-01 1.67955942e-02 -3.33524793e-01 -6.95872530e-02 -6.49313569e-01 -8.82496417e-01 9.50722396e-01 -5.87812543e-01 -1.24214971e+00 -5.16325414e-01 -8.56727183e-01 -4.76038158e-01 9.83710706e-01 -1.04662025e+00 -3.72760624e-01 4.36989874e-01 6.81297600e-01 1.12180993e-01 -1.68271810e-01 1.20388961e+00 2.21357439e-02 -5.30683398e-01 6.87915981e-01 3.91172409e-01 -2.49080956e-02 1.60936907e-01 -9.04479206e-01 7.46145308e-01 4.52446312e-01 -2.61347592e-02 1.16606247e+00 6.64301813e-01 -4.61450487e-01 -1.70620680e+00 -9.70176637e-01 7.76508451e-01 -1.88127026e-01 9.65855777e-01 -4.16927248e-01 -1.11437428e+00 8.15670490e-01 -1.01968348e-01 2.12457627e-01 6.24810636e-01 -2.89860636e-01 -5.92660129e-01 -3.26778114e-01 -1.04491603e+00 8.63314390e-01 1.20832646e+00 -9.61730480e-01 -3.34598422e-01 -8.30282494e-02 5.88925600e-01 -5.20718515e-01 -1.06197584e+00 2.57714063e-01 8.02326918e-01 -6.75119519e-01 8.41756880e-01 -2.08963364e-01 1.75071940e-01 -5.94910197e-02 -2.19907254e-01 -1.04053783e+00 -3.59311819e-01 -8.77443671e-01 -4.28611398e-01 2.43002847e-01 5.76084673e-01 -1.14122665e+00 5.20596802e-01 5.02515018e-01 -1.92073256e-01 -9.49002683e-01 -1.40071762e+00 -6.15128398e-01 4.05703843e-01 -4.63927627e-01 -5.50503358e-02 5.87754130e-01 3.34817261e-01 3.00650299e-02 7.77902603e-02 8.55572820e-02 3.00365776e-01 5.14305532e-02 -2.17165232e-01 -8.34269345e-01 -4.34323817e-01 -8.04017603e-01 -4.93788749e-01 -1.24231851e+00 -6.18175045e-03 -8.84463787e-01 1.42908901e-01 -9.91793036e-01 -1.56585202e-01 -5.19455016e-01 -7.92172492e-01 6.04754984e-01 3.74883801e-01 3.65344465e-01 7.39123449e-02 4.07934844e-01 -3.32426637e-01 3.83493185e-01 9.18756425e-01 1.29572013e-02 -3.74131858e-01 -1.85246333e-01 -1.05174191e-01 5.01137495e-01 7.50442624e-01 -4.37293828e-01 -3.41044456e-01 -6.56597078e-01 5.45512319e-01 3.25801700e-01 4.08434689e-01 -1.67337883e+00 8.38032603e-01 1.03908405e-01 6.85560048e-01 -5.45979083e-01 7.13396847e-01 -3.04214388e-01 2.55672187e-01 8.13289344e-01 -5.22518575e-01 6.29998028e-01 7.41317689e-01 4.25279111e-01 1.11049421e-01 2.63082404e-02 8.53024602e-01 -3.10358256e-01 -4.91029203e-01 3.46977949e-01 -8.76003683e-01 1.35023192e-01 6.92135215e-01 -4.35400635e-01 -4.99695301e-01 -9.32432786e-02 -8.07437420e-01 -3.13342474e-02 2.35448286e-01 -1.32214978e-01 7.12040007e-01 -1.04194748e+00 -8.08135495e-02 4.94331032e-01 -4.68310475e-01 -7.30215311e-02 4.09233958e-01 5.57983518e-01 -5.66459298e-01 5.65186739e-01 -5.40392220e-01 -5.92253149e-01 -4.85920370e-01 1.86677903e-01 7.16930032e-01 -1.03451312e-01 -5.19462407e-01 1.30050957e+00 -3.52452755e-01 -1.24803171e-01 1.30582869e-01 -7.13447809e-01 2.93193191e-01 -1.17140606e-01 5.00371158e-01 2.07571596e-01 -7.28320181e-02 -1.17378354e-01 -2.26014718e-01 7.46142089e-01 -2.25726187e-01 -5.84883630e-01 1.44619250e+00 2.79647499e-01 -5.20619452e-01 1.11078441e+00 1.45022535e+00 -6.14290833e-01 -1.58055317e+00 -2.55672604e-01 -1.97986364e-01 5.96548140e-01 -1.69686884e-01 -6.30910754e-01 -8.11669350e-01 1.05815411e+00 8.54409993e-01 2.51453727e-01 1.00501573e+00 -4.12709385e-01 9.66523290e-01 1.02712560e+00 5.34647584e-01 -1.18274570e+00 2.75641650e-01 1.29374695e+00 6.62571847e-01 -4.17451084e-01 -4.23261702e-01 4.60649192e-01 -1.04017265e-01 1.29737568e+00 3.90369356e-01 -7.77929008e-01 9.21256244e-01 7.60344982e-01 -2.12347522e-01 1.00027826e-02 -1.03007877e+00 3.08555335e-01 -2.47035008e-02 1.71409637e-01 3.86793643e-01 9.22789797e-02 9.94132310e-02 3.52752149e-01 -5.26623368e-01 1.74062297e-01 2.13593304e-01 1.09344852e+00 -7.22643375e-01 -6.96752846e-01 8.05574507e-02 7.46397734e-01 -1.96417853e-01 -3.94094288e-01 4.28523839e-01 4.71269101e-01 -3.49829286e-01 3.42315614e-01 6.38811886e-01 -4.59451109e-01 -1.04715697e-01 3.57399464e-01 7.52785802e-01 -2.25044623e-01 -5.53412974e-01 -2.45310906e-02 -5.49905956e-01 -4.25930679e-01 1.64049208e-01 -4.91972417e-01 -2.03693891e+00 -5.76508641e-01 1.35233864e-01 -2.20087707e-01 5.84169745e-01 9.20301735e-01 5.89809358e-01 8.44573498e-01 4.38593864e-01 -1.02597976e+00 -5.01747191e-01 -6.61677182e-01 -4.12333637e-01 -2.45818600e-01 4.86771464e-01 -1.24147363e-01 -3.20057154e-01 -1.17324658e-01]	[8.109709739685059, 2.6685543060302734]
e316463c-40b9-4498-ad40-74753edc644a	action-q-transformer-visual-explanation-in	2306.13879	null	https://arxiv.org/abs/2306.13879v1	https://arxiv.org/pdf/2306.13879v1.pdf	Action Q-Transformer: Visual Explanation in Deep Reinforcement Learning with Encoder-Decoder Model using Action Query	The excellent performance of Transformer in supervised learning has led to growing interest in its potential application to deep reinforcement learning (DRL) to achieve high performance on a wide variety of problems. However, the decision making of a DRL agent is a black box, which greatly hinders the application of the agent to real-world problems. To address this problem, we propose the Action Q-Transformer (AQT), which introduces a transformer encoder-decoder structure to Q-learning based DRL methods. In AQT, the encoder calculates the state value function and the decoder calculates the advantage function to promote the acquisition of different attentions indicating the agent's decision-making. The decoder in AQT utilizes action queries, which represent the information of each action, as queries. This enables us to obtain the attentions for the state value and for each action. By acquiring and visualizing these attentions that detail the agent's decision-making, we achieve a DRL model with high interpretability. In this paper, we show that visualization of attention in Atari 2600 games enables detailed analysis of agents' decision-making in various game tasks. Further, experimental results demonstrate that our method can achieve higher performance than the baseline in some games.	['Komei Sugiura', 'Hironobu Fujiyoshi', 'Takayoshi Yamashita', 'Tsubasa Hirakawa', 'Hidenori Itaya']	2023-06-24	null	null	null	null	['q-learning', 'atari-games', 'decision-making']	['methodology', 'playing-games', 'reasoning']	[-3.32415313e-01 -1.20618485e-01 -5.47843166e-02 -2.60725051e-01 -2.85727620e-01 -3.48520637e-01 6.67998314e-01 5.89937009e-02 -5.14051199e-01 5.10010123e-01 1.65067211e-01 -4.39027995e-01 7.98338950e-02 -8.85021687e-01 -2.90923059e-01 -7.92583823e-01 -1.19884036e-01 4.09112722e-01 2.82631606e-01 -5.83950162e-01 3.72086734e-01 1.02491915e-01 -1.28555059e+00 5.10765672e-01 9.37841356e-01 9.65439558e-01 4.67754006e-01 4.61685061e-01 -2.46209294e-01 1.65506935e+00 -1.08795881e+00 -4.07946229e-01 1.40546754e-01 -6.93201065e-01 -8.42312276e-01 -2.72303373e-01 -2.39052102e-01 -8.90654087e-01 -2.78268158e-01 1.17009044e+00 5.93160689e-01 2.35116929e-01 5.39390445e-01 -1.71316862e+00 -8.78884614e-01 6.90524220e-01 -4.93803650e-01 4.21150953e-01 6.00440502e-01 7.07574606e-01 1.04127085e+00 -5.40717542e-01 1.94183469e-01 1.74549294e+00 7.42849559e-02 6.09650314e-01 -8.60916078e-01 -8.01444292e-01 3.66434664e-01 7.61129797e-01 -8.16933453e-01 4.50832918e-02 8.46789956e-01 -4.26066399e-01 1.24342084e+00 -1.45761102e-01 1.10549116e+00 8.17381620e-01 4.41215158e-01 1.38772380e+00 1.21255660e+00 -2.22319309e-02 5.65484762e-01 -2.87674785e-01 -1.24509074e-01 7.74679184e-01 -3.79345983e-01 5.52303016e-01 -7.33609021e-01 2.55767822e-01 1.13220727e+00 1.13509066e-01 8.02532658e-02 -4.15114492e-01 -1.16301072e+00 9.92430210e-01 8.60786080e-01 -6.81638345e-02 -4.72965986e-01 3.63179594e-01 5.52701056e-01 6.75274551e-01 1.11948133e-01 3.50289881e-01 -2.47638196e-01 -5.98869026e-01 -6.67518154e-02 3.30613881e-01 3.10331732e-01 6.82629943e-01 7.19941676e-01 5.02172291e-01 -6.14612460e-01 6.69742823e-01 5.70999801e-01 2.89663494e-01 7.40534425e-01 -1.19178903e+00 5.39764941e-01 9.83440638e-01 9.92899910e-02 -6.77164018e-01 -3.12634677e-01 -3.56380522e-01 -7.07898557e-01 9.46184397e-01 2.12482929e-01 -1.41270682e-01 -6.60318315e-01 1.72699690e+00 2.59558707e-01 5.09811565e-02 4.86423343e-01 1.08033276e+00 8.46915424e-01 6.61472857e-01 3.44255209e-01 -3.13049853e-02 1.41037095e+00 -1.11060441e+00 -9.01304841e-01 -3.52311492e-01 7.20267594e-01 -1.45925269e-01 1.49965417e+00 4.84804779e-01 -8.85536492e-01 -7.82132626e-01 -9.94845212e-01 -6.56628683e-02 -7.23792687e-02 -4.28233966e-02 7.22980976e-01 1.98214382e-01 -1.07629955e+00 4.11278397e-01 -8.98464799e-01 -2.58815382e-02 6.20566547e-01 2.84328908e-01 1.60959303e-01 3.56398374e-01 -1.54554701e+00 1.02211833e+00 5.03105879e-01 5.56926336e-03 -1.52802312e+00 -3.02938938e-01 -1.06499398e+00 3.62747908e-01 4.76510346e-01 -4.92862165e-01 1.89108038e+00 -9.88420784e-01 -1.98686004e+00 3.56280059e-01 2.16499388e-01 -6.63102686e-01 5.16765952e-01 -1.65221199e-01 -1.48581490e-01 2.94927508e-02 2.26650581e-01 8.44299734e-01 8.09694469e-01 -7.96445429e-01 -1.05974185e+00 -3.31398040e-01 7.71328032e-01 8.47385108e-01 -3.97052690e-02 -1.09033711e-01 -3.49174052e-01 -5.71113706e-01 -4.00789797e-01 -4.86199111e-01 -1.75705135e-01 -9.55767259e-02 3.75062972e-02 -8.78205299e-01 8.40268433e-01 -4.32318032e-01 1.20629191e+00 -2.28758693e+00 1.06836937e-01 -3.48432928e-01 5.10558963e-01 4.04258817e-01 -2.00034365e-01 4.83642519e-01 3.11146230e-02 -1.43629387e-01 -6.83731586e-02 -1.00225851e-01 2.82845855e-01 4.81466174e-01 -2.51626402e-01 9.41870883e-02 3.25033218e-01 1.19652855e+00 -1.20230711e+00 -4.04259294e-01 5.11403561e-01 2.49781877e-01 -6.67396426e-01 6.37224495e-01 -3.33407402e-01 4.07977581e-01 -7.42733479e-01 2.21249402e-01 2.14442328e-01 -1.59754127e-01 7.34057203e-02 7.18239099e-02 3.84482816e-02 6.22911692e-01 -9.89121318e-01 1.63900685e+00 -1.46837920e-01 5.65099239e-01 -2.95372754e-01 -9.67970431e-01 8.65899801e-01 1.55173391e-01 2.44799361e-01 -1.52519393e+00 8.46865922e-02 -3.93184572e-01 4.64155048e-01 -4.85741228e-01 2.59587079e-01 -5.88431954e-02 -1.02779903e-01 7.88454771e-01 -8.13277364e-02 1.67092402e-02 4.01626736e-01 5.00561893e-01 9.63341117e-01 4.87909347e-01 4.32346076e-01 -4.61153686e-02 4.59323496e-01 -8.98515135e-02 6.31594419e-01 5.68389893e-01 -4.59752917e-01 -2.72373348e-01 8.79069686e-01 -6.68206751e-01 -5.65612912e-01 -8.76413524e-01 5.51826715e-01 1.49007237e+00 2.44373992e-01 -5.34928739e-01 -7.60680616e-01 -8.32295656e-01 1.56711917e-02 8.91761422e-01 -8.02623928e-01 -6.73606575e-01 -7.00951874e-01 -4.57548618e-01 2.31708393e-01 8.17049980e-01 1.14743233e+00 -1.78293049e+00 -1.24989295e+00 2.64767885e-01 -6.55895323e-02 -5.95199287e-01 -4.12732184e-01 1.85828298e-01 -7.50537574e-01 -1.11249578e+00 -3.53723288e-01 -7.72259712e-01 4.26478148e-01 1.28754511e-01 1.19364870e+00 1.08035952e-01 1.06711328e-01 2.19604090e-01 -3.70311677e-01 -3.20176929e-01 -4.56685215e-01 -3.58679831e-01 -7.32295439e-02 -3.12993288e-01 4.50087100e-01 -2.14949772e-01 -6.65293753e-01 3.34827721e-01 -6.57191813e-01 4.88250911e-01 5.08377254e-01 7.47249544e-01 1.61741585e-01 1.47845253e-01 5.06168544e-01 -5.36147237e-01 1.29661381e+00 -2.62732357e-01 -8.21692050e-01 1.07442096e-01 -6.02415979e-01 4.90361869e-01 6.98498785e-01 -4.09255713e-01 -1.12836146e+00 -3.93676221e-01 -1.65782750e-01 -2.00793445e-01 9.54373404e-02 5.16428888e-01 -2.39172921e-01 4.55081284e-01 4.26746637e-01 4.35415626e-01 2.68929273e-01 -3.18650633e-01 3.44606191e-01 4.88224566e-01 4.42528248e-01 -6.01400971e-01 3.25764716e-01 1.08541368e-04 -4.01525915e-01 1.81715693e-02 -6.84478283e-01 5.08603305e-02 -1.58609569e-01 -4.87336904e-01 8.66232455e-01 -7.69354939e-01 -1.58729386e+00 6.00708246e-01 -9.08197701e-01 -9.31554675e-01 -5.22954941e-01 2.97685117e-01 -8.35028708e-01 2.96252854e-02 -8.66973519e-01 -7.41524279e-01 -3.55649620e-01 -1.64082360e+00 7.45480239e-01 4.72387940e-01 5.83366938e-02 -1.01842034e+00 1.40028611e-01 1.48663864e-01 2.96993077e-01 -2.78957486e-01 9.86081243e-01 -4.25210178e-01 -4.98062700e-01 5.49489439e-01 -2.06461534e-01 1.83947757e-01 2.97521859e-01 -3.82033885e-01 -7.26832092e-01 -3.96991968e-01 2.83172075e-02 -5.49995303e-01 5.76228917e-01 1.24158032e-01 1.24628592e+00 -3.19123417e-01 7.17181936e-02 3.22863460e-01 9.90609050e-01 9.32186604e-01 7.04429984e-01 5.54666579e-01 5.11918128e-01 2.93561667e-01 9.73502696e-01 5.83083928e-01 8.14371049e-01 7.72222757e-01 8.99541914e-01 -2.49325618e-01 -9.90521535e-02 -5.53230166e-01 8.66001010e-01 4.89847690e-01 -7.07719624e-02 -1.05735339e-01 -7.18804479e-01 4.51459698e-02 -2.07842588e+00 -1.05208051e+00 2.99558401e-01 1.88769066e+00 1.00669324e+00 1.42591864e-01 3.28391075e-01 7.30828866e-02 3.91957521e-01 1.68491021e-01 -1.15730655e+00 -5.55062473e-01 3.60484958e-01 -9.86671820e-02 -2.02376410e-01 6.22435570e-01 -7.86654651e-01 1.25215983e+00 6.45153475e+00 8.97855818e-01 -9.59936678e-01 -1.19668804e-01 4.99862105e-01 1.53071001e-01 -9.01131332e-02 -2.34684095e-01 -5.91510296e-01 5.22952676e-01 6.30491138e-01 -4.56428915e-01 7.20377266e-01 7.78407991e-01 3.91556650e-01 -2.85488904e-01 -1.24430561e+00 1.23792219e+00 -2.11184353e-01 -9.96948421e-01 3.47379565e-01 7.03354329e-02 1.68212622e-01 -2.33409360e-01 1.80258274e-01 9.06747997e-01 1.17830944e+00 -9.89100099e-01 7.95686781e-01 1.96130052e-01 5.28727591e-01 -9.04875100e-01 6.20998204e-01 5.78828096e-01 -1.23103583e+00 -3.80056381e-01 -3.95372003e-01 -5.68876982e-01 -1.79046303e-01 -2.61071622e-01 -1.02506781e+00 2.54859567e-01 8.06335390e-01 1.17147529e+00 -5.33805430e-01 6.46401525e-01 -9.20989156e-01 4.35324490e-01 2.08485529e-01 -3.13555896e-01 5.25231004e-01 -4.76074547e-01 9.22760963e-02 7.67104805e-01 -5.19487821e-02 2.66474903e-01 2.99874276e-01 7.90273786e-01 1.55826018e-03 -5.51619902e-02 -3.97810131e-01 -1.13302886e-01 3.02287072e-01 1.03521144e+00 -4.87491399e-01 -5.63529611e-01 -2.08381012e-01 9.52499926e-01 8.24984491e-01 4.44722712e-01 -8.50959957e-01 -1.02304637e-01 9.20859158e-01 -2.65269727e-01 1.08574212e-01 -2.16021106e-01 1.87094852e-01 -8.54843795e-01 -2.11548716e-01 -1.26810801e+00 6.62419975e-01 -1.13894081e+00 -8.88368547e-01 5.04649520e-01 -1.14482969e-01 -1.34661579e+00 -6.46539927e-01 -5.64481378e-01 -7.29794204e-01 7.56133318e-01 -1.54790592e+00 -6.99127913e-01 -1.89294457e-01 8.57346475e-01 9.60365832e-01 -5.22153080e-01 7.01555967e-01 -3.30108777e-02 -5.89243054e-01 2.59504110e-01 -3.03125083e-01 4.15159672e-01 3.10173541e-01 -1.68256903e+00 5.55732429e-01 5.93961775e-01 1.28974229e-01 2.28060544e-01 4.67154264e-01 -5.00288010e-01 -1.35895252e+00 -7.40524411e-01 2.43734702e-01 -2.96690792e-01 5.30743718e-01 -1.75677389e-01 -8.06046963e-01 6.80574656e-01 6.19638920e-01 -4.53672141e-01 6.19809330e-01 -2.40316913e-01 -5.73817343e-02 -2.30728060e-01 -9.04477596e-01 7.94566453e-01 8.81913960e-01 -5.20434797e-01 -9.56337571e-01 1.09822728e-01 7.28447795e-01 -4.06487525e-01 -4.89384443e-01 -9.11146477e-02 2.70819485e-01 -1.17936277e+00 9.50802326e-01 -8.62287760e-01 6.06584430e-01 -6.99232638e-01 2.53134847e-01 -1.77673542e+00 -5.43467879e-01 -4.07864153e-01 -1.82810634e-01 8.16616118e-01 -1.02594092e-01 -7.44882405e-01 4.86436784e-01 5.10458887e-01 -1.29323736e-01 -7.00874150e-01 -7.84572423e-01 -4.32198822e-01 -8.24953914e-02 -4.03108567e-01 9.56438184e-01 8.25640440e-01 2.74807096e-01 4.89967078e-01 -3.38533461e-01 -1.01831138e-01 4.28261191e-01 1.78934753e-01 6.93041682e-01 -9.95677352e-01 -3.35562527e-01 -6.44479334e-01 -2.25740984e-01 -1.51609135e+00 9.14265886e-02 -8.40541661e-01 -1.17338903e-01 -1.96926188e+00 1.06367208e-01 -2.91843891e-01 -5.02731562e-01 9.69589591e-01 -2.95501947e-01 -1.37046635e-01 4.97890592e-01 2.91697085e-01 -9.53534901e-01 1.01856244e+00 1.77921879e+00 -4.23782587e-01 -2.01647833e-01 -1.11778930e-01 -7.55856037e-01 7.36190021e-01 9.42480326e-01 -2.93534130e-01 -7.94152737e-01 -7.05246866e-01 3.73659611e-01 2.72744507e-01 3.29374045e-01 -9.04705584e-01 3.51907760e-01 -3.77425790e-01 4.62889642e-01 -4.86127496e-01 3.27423900e-01 -6.95429325e-01 -3.66216898e-01 9.01419163e-01 -5.52361965e-01 5.79355955e-01 1.34491965e-01 2.26251841e-01 -2.86497205e-01 1.63910817e-02 5.01390994e-01 -2.80516416e-01 -1.10650575e+00 3.38018537e-01 -6.95328355e-01 9.02999863e-02 1.06640208e+00 -1.21604674e-01 -2.81330645e-01 -6.43519163e-01 -5.53811431e-01 9.10349965e-01 9.54347327e-02 5.17617166e-01 8.87070596e-01 -1.53512490e+00 -6.33513749e-01 4.21325207e-01 -9.53106806e-02 1.03440851e-01 5.65467328e-02 3.77850473e-01 -3.53676558e-01 1.36844799e-01 -7.24564850e-01 -4.92615253e-01 -1.17106795e+00 6.63092494e-01 6.69332564e-01 -3.85495722e-01 -7.23701537e-01 5.99445760e-01 5.96538424e-01 -2.98061699e-01 3.62666667e-01 -4.25863087e-01 -9.08843279e-01 3.59542742e-02 8.42915177e-01 2.03467295e-01 -3.30541819e-01 -3.15889210e-01 -2.46303260e-01 2.32836604e-01 -2.51806051e-01 -1.71778068e-01 1.25638926e+00 7.40073621e-02 1.08404987e-01 3.59405637e-01 5.73615074e-01 -5.99600852e-01 -1.86024117e+00 -2.56681651e-01 -2.06976399e-01 -4.80385512e-01 5.11281081e-02 -1.01085389e+00 -1.22438991e+00 1.20013916e+00 6.55606329e-01 3.45204830e-01 1.21674562e+00 -1.26421615e-01 3.25183302e-01 4.82206285e-01 5.60290098e-01 -1.15735590e+00 5.76633394e-01 6.70631707e-01 1.10862410e+00 -1.20245564e+00 -2.99679726e-01 1.86540008e-01 -1.12107182e+00 1.02280140e+00 1.06840384e+00 -5.06295599e-02 1.46927476e-01 2.75855809e-01 3.69250059e-01 -3.00498247e-01 -1.15872788e+00 -2.99527854e-01 -1.54348612e-01 7.79735565e-01 1.48365512e-01 1.25896379e-01 3.00000366e-02 3.99522454e-01 -4.07039136e-01 -9.61381383e-03 2.57200062e-01 9.45966125e-01 -4.65341479e-01 -1.10610998e+00 -3.49262469e-02 2.74800301e-01 7.07449540e-02 -8.24261084e-02 -2.21974686e-01 6.33143544e-01 -2.07643151e-01 9.46966231e-01 2.67215908e-01 -3.43355179e-01 4.10076231e-01 -1.73506469e-01 2.66490310e-01 -5.55486143e-01 -9.38644588e-01 -6.51877224e-02 -2.05579877e-01 -8.67083549e-01 -3.27464134e-01 -3.76320481e-01 -1.79137063e+00 -2.98454136e-01 2.74407268e-01 3.42231512e-01 1.49905846e-01 1.03692007e+00 2.38383234e-01 1.01180005e+00 5.30021071e-01 -4.44386989e-01 -7.18933403e-01 -9.69491541e-01 -4.13042814e-01 4.94795114e-01 2.83667028e-01 -9.28978622e-01 3.30097154e-02 -2.27374449e-01]	[3.9414782524108887, 1.6065796613693237]
88123ee2-e244-4f28-a7d6-2a657dcaac62	does-generative-face-completion-help-face	1906.02858	null	https://arxiv.org/abs/1906.02858v1	https://arxiv.org/pdf/1906.02858v1.pdf	Does Generative Face Completion Help Face Recognition?	Face occlusions, covering either the majority or discriminative parts of the face, can break facial perception and produce a drastic loss of information. Biometric systems such as recent deep face recognition models are not immune to obstructions or other objects covering parts of the face. While most of the current face recognition methods are not optimized to handle occlusions, there have been a few attempts to improve robustness directly in the training stage. Unlike those, we propose to study the effect of generative face completion on the recognition. We offer a face completion encoder-decoder, based on a convolutional operator with a gating mechanism, trained with an ample set of face occlusions. To systematically evaluate the impact of realistic occlusions on recognition, we propose to play the occlusion game: we render 3D objects onto different face parts, providing precious knowledge of what the impact is of effectively removing those occlusions. Extensive experiments on the Labeled Faces in the Wild (LFW), and its more difficult variant LFW-BLUFR, testify that face completion is able to partially restore face perception in machine vision systems for improved recognition.	['Wael Abd-Almageed', 'Joe Mathai', 'Iacopo Masi']	2019-06-07	null	null	null	null	['facial-inpainting']	['computer-vision']	[ 4.62311894e-01 5.11682034e-01 2.62492508e-01 -7.41454124e-01 -3.09579343e-01 -3.00807834e-01 6.29140198e-01 -7.35633671e-01 -8.20188597e-02 3.46661448e-01 2.43034273e-01 -2.75563151e-02 7.85513520e-02 -5.17009795e-01 -1.04611969e+00 -7.71168411e-01 1.02658058e-02 2.71413803e-01 -3.05867136e-01 -1.46401450e-01 1.37741044e-01 8.81733954e-01 -2.02078295e+00 4.41356748e-01 3.18730623e-01 7.89911866e-01 -1.92816138e-01 4.03498858e-01 6.38433397e-02 3.47671777e-01 -6.98203921e-01 -7.65769422e-01 6.31074667e-01 -2.07840934e-01 -5.93062043e-01 5.21788418e-01 9.47657228e-01 -6.81181610e-01 -4.93779957e-01 8.17395568e-01 6.70761168e-01 -2.39840180e-01 5.22071421e-01 -1.19426298e+00 -6.47807002e-01 1.66105941e-01 -6.54704034e-01 -1.22667901e-01 4.86606866e-01 2.24381119e-01 5.55284619e-01 -1.13411725e+00 7.48189151e-01 1.79272735e+00 6.62111640e-01 1.13914502e+00 -1.50685310e+00 -4.31550443e-01 5.57436198e-02 -8.07173476e-02 -1.34060514e+00 -1.27122974e+00 6.35689259e-01 -2.55816668e-01 7.86788881e-01 3.72475564e-01 1.98219702e-01 1.27140653e+00 1.10766187e-01 5.05416572e-01 1.04883802e+00 -5.78603685e-01 5.90262003e-02 3.61987054e-02 -6.08320348e-02 7.01956630e-01 2.55781621e-01 2.61739135e-01 -6.61442935e-01 -1.33391649e-01 5.15584588e-01 5.37380502e-02 -4.59768891e-01 -3.55046481e-01 -3.84073347e-01 4.44871575e-01 2.12062582e-01 -3.93407270e-02 -2.34069303e-01 1.43804312e-01 1.24013647e-01 3.81521046e-01 4.66993779e-01 -1.97134316e-02 -2.74242908e-01 3.62055659e-01 -8.48462820e-01 1.36024058e-01 9.11872268e-01 6.43182993e-01 8.73127937e-01 1.04238391e-01 -3.35212469e-01 8.55354667e-01 5.20992100e-01 4.03032094e-01 1.31120369e-01 -9.02501822e-01 -4.33506891e-02 5.15303373e-01 -2.45248169e-01 -7.21624494e-01 -2.93166161e-01 -2.74991095e-01 -5.56016803e-01 6.58561110e-01 3.55276644e-01 -5.91489747e-02 -1.20869768e+00 2.01266217e+00 4.34831679e-01 2.25258231e-01 -1.09997295e-01 8.23018610e-01 8.50522578e-01 1.34948924e-01 -5.28260618e-02 -7.02389106e-02 1.44684672e+00 -5.28941393e-01 -7.22146034e-01 -2.74047494e-01 3.07136357e-01 -9.75451291e-01 8.91969204e-01 2.55253196e-01 -1.03032482e+00 -6.60635412e-01 -1.04191136e+00 -2.36193389e-01 -1.23106591e-01 2.19710127e-01 6.15971088e-01 1.33703339e+00 -1.30886221e+00 6.85692966e-01 -7.42233336e-01 -3.75981063e-01 9.82986093e-01 8.61976743e-01 -9.22157407e-01 -5.60358644e-01 -6.82989836e-01 9.85889554e-01 -3.53824526e-01 5.25819480e-01 -1.26888800e+00 -6.53985739e-01 -9.27267849e-01 3.55768174e-01 2.04350784e-01 -5.98040521e-01 8.39173734e-01 -1.18483710e+00 -1.44598353e+00 1.15317869e+00 -5.89700818e-01 -1.72317699e-01 4.20635581e-01 -2.65761822e-01 -1.92807987e-01 -1.03706531e-01 -3.35414708e-01 9.75549817e-01 1.32213080e+00 -1.25384557e+00 1.80346176e-01 -8.26618373e-01 7.18295947e-02 -1.08276129e-01 -3.75943422e-01 -6.64252648e-03 -4.18783545e-01 -3.60258222e-01 3.93165611e-02 -8.74797285e-01 1.16167337e-01 2.82006294e-01 -3.93654376e-01 5.34227826e-02 9.94708836e-01 -8.03693414e-01 5.61247587e-01 -2.40546489e+00 -7.55958557e-02 1.10000089e-01 -3.95568926e-03 6.16543174e-01 -7.20245004e-01 3.88307571e-01 -3.85378748e-01 5.94690666e-02 -1.28826603e-01 -1.03973353e+00 -1.27648581e-02 6.43655479e-01 -4.43294257e-01 7.17294574e-01 6.75228536e-01 8.05040061e-01 -1.25335515e-01 1.08301237e-01 2.98706368e-02 1.14569509e+00 -1.00149071e+00 5.48953451e-02 -6.09446838e-02 4.56623405e-01 -1.20913357e-01 6.94452941e-01 1.17679954e+00 3.06653738e-01 1.96495295e-01 -8.84340331e-02 2.92087227e-01 1.10500023e-01 -9.60731447e-01 1.72928953e+00 -2.22489953e-01 6.26923740e-01 3.26674551e-01 -8.82546246e-01 8.41689169e-01 4.44029301e-01 1.78415492e-01 -7.28549600e-01 1.88113287e-01 1.18361011e-01 1.77592933e-01 -6.08083844e-01 -6.85510365e-03 -1.95099011e-01 5.58471501e-01 2.25396901e-01 3.07893574e-01 3.15559208e-01 -2.88668405e-02 -6.16928190e-02 1.30563939e+00 2.86531746e-01 -2.55878150e-01 -2.78533041e-01 5.83588362e-01 -6.54086709e-01 4.87372398e-01 5.12241542e-01 -1.54087812e-01 9.91068840e-01 6.88913465e-01 -5.82287431e-01 -8.90588939e-01 -8.46189082e-01 -3.42125326e-01 9.24271226e-01 -2.28946254e-01 -3.31925899e-01 -1.19820845e+00 -5.58359861e-01 6.96549341e-02 3.56068611e-01 -7.30249763e-01 -5.22723138e-01 -5.33623993e-01 -9.37129736e-01 6.57832921e-01 2.33439595e-01 4.03604686e-01 -1.04725337e+00 -4.64111418e-01 -2.51793295e-01 1.11781999e-01 -1.11653244e+00 -2.79132009e-01 -1.07644439e-01 -6.55046284e-01 -1.06404436e+00 -4.21287566e-01 -5.79568863e-01 1.11877644e+00 2.07667217e-01 9.70177829e-01 4.75424469e-01 -8.30258906e-01 5.76512873e-01 -5.16992025e-02 -3.06883246e-01 -4.03716028e-01 -2.44938359e-01 2.40356307e-02 4.84394014e-01 3.97742987e-01 -6.90955877e-01 -5.02859652e-01 2.71473765e-01 -9.85681474e-01 -2.35239714e-01 5.11696875e-01 9.81019855e-01 8.69652405e-02 -2.32621014e-01 5.32810032e-01 -1.07017756e+00 2.85753727e-01 4.87422459e-02 -3.60654265e-01 1.53841317e-01 -3.08939517e-01 1.26338705e-01 3.17461461e-01 -2.71085262e-01 -1.23157012e+00 4.05859977e-01 -5.77328444e-01 -6.35207355e-01 -3.65894914e-01 -1.63080722e-01 -8.42518687e-01 -3.37387592e-01 6.13492668e-01 -4.93263006e-02 2.57592320e-01 -6.71076596e-01 3.53725910e-01 3.75284374e-01 4.15835202e-01 -4.65399414e-01 6.78129256e-01 8.56809616e-01 2.59080648e-01 -9.10314202e-01 -5.43184578e-01 -6.36691302e-02 -6.63548231e-01 -1.22184269e-01 5.34273803e-01 -7.72558153e-01 -9.50660944e-01 2.47823507e-01 -1.33731759e+00 -1.07805036e-01 -3.49934548e-01 1.02065854e-01 -4.51152027e-01 2.01354891e-01 -3.46120209e-01 -8.34158778e-01 -4.29410078e-02 -1.26291478e+00 1.31487012e+00 1.84408769e-01 1.62856042e-01 -5.86235762e-01 -1.56069249e-01 3.22707444e-01 4.30776358e-01 6.88069388e-02 7.31105447e-01 -4.09816146e-01 -6.59852982e-01 -2.83694476e-01 -1.42138198e-01 5.19257009e-01 1.72337115e-01 -2.84438226e-02 -1.87483239e+00 -5.31933725e-01 1.76053852e-01 -2.38297343e-01 1.23187375e+00 1.93970099e-01 1.12026823e+00 -4.36841875e-01 -3.05721253e-01 7.91530848e-01 1.22625387e+00 -7.48367980e-02 1.34712267e+00 -5.05201638e-01 4.09310788e-01 8.77479315e-01 -8.78851116e-02 2.74829328e-01 -1.86149567e-01 6.22838080e-01 6.89735949e-01 -2.16160804e-01 -6.76412225e-01 -1.46888226e-01 5.38181067e-01 -3.82178719e-03 -7.11115450e-02 -2.85638809e-01 -4.74553436e-01 1.18192405e-01 -1.28534448e+00 -8.25842738e-01 2.81921238e-01 2.20288920e+00 4.68930632e-01 -1.00350603e-01 -4.36910719e-01 1.73915029e-01 6.79547548e-01 5.75035252e-02 -2.72236764e-01 -4.08408195e-01 -1.39082208e-01 6.22813940e-01 1.72262996e-01 6.97395623e-01 -9.04635966e-01 8.68824542e-01 6.81900883e+00 5.90277970e-01 -1.23776734e+00 1.42626554e-01 7.56351411e-01 -2.80618161e-01 -2.60009170e-01 1.57191381e-01 -9.71825898e-01 1.32236063e-01 7.09704757e-01 6.50465667e-01 7.19635367e-01 6.59946620e-01 5.40153775e-03 7.79858753e-02 -1.45523310e+00 8.96698415e-01 5.07624209e-01 -9.89789307e-01 2.07068503e-01 3.17885101e-01 4.42061633e-01 -2.55271316e-01 3.45966578e-01 2.35640720e-01 -1.46397173e-01 -1.37970960e+00 4.75176096e-01 7.53139138e-01 8.29332709e-01 -4.91201699e-01 4.81454343e-01 1.94220096e-01 -5.04237890e-01 -9.68388841e-02 -6.53737605e-01 -6.69360235e-02 -1.23895548e-01 6.85083091e-01 -9.77851331e-01 2.18764409e-01 3.57323766e-01 2.47815058e-01 -7.53918946e-01 6.50208473e-01 -2.23148227e-01 4.95633185e-01 -3.53961110e-01 6.51960373e-01 -4.06498224e-01 1.24501020e-01 4.81797993e-01 8.43606830e-01 1.93356171e-01 1.42281409e-02 -2.23469988e-01 9.89026666e-01 -5.38967729e-01 -9.48556736e-02 -7.56852210e-01 1.24178201e-01 7.71194920e-02 1.24478698e+00 -5.23613989e-01 4.96860780e-02 -5.46318829e-01 1.04883516e+00 3.90843153e-01 4.76458073e-01 -5.54443002e-01 1.37584671e-01 9.87431169e-01 3.78999591e-01 3.10614705e-01 7.33617023e-02 -2.06732422e-01 -1.04553413e+00 3.07832807e-01 -7.65332162e-01 -3.49303856e-02 -5.82088232e-01 -9.64215875e-01 5.87147236e-01 -4.15146977e-01 -5.56494117e-01 1.05181169e-02 -7.12327957e-01 -5.32869875e-01 9.40514743e-01 -1.57033455e+00 -1.22315240e+00 -2.53147691e-01 4.94372755e-01 4.55840677e-01 -1.56278670e-01 9.36994970e-01 6.88625634e-01 -7.42321610e-01 8.87189448e-01 -2.54232794e-01 7.98873752e-02 7.62884915e-01 -5.77387691e-01 4.93333071e-01 8.85832489e-01 3.45745176e-01 8.26828361e-01 7.10001826e-01 -3.75444680e-01 -1.65284085e+00 -9.21711922e-01 8.45099390e-01 -4.92896587e-01 -4.59997326e-01 -8.54888797e-01 -9.86613750e-01 8.21304679e-01 1.53183177e-01 3.74663442e-01 5.02295077e-01 1.26590177e-01 -7.47559786e-01 -3.35452139e-01 -1.33627236e+00 5.06986856e-01 1.33729184e+00 -5.13034105e-01 -3.09728146e-01 2.06589311e-01 3.06250811e-01 -6.04627877e-02 -3.97081554e-01 6.52700961e-01 8.01568747e-01 -1.28035688e+00 1.12138915e+00 -7.15688825e-01 2.88280934e-01 -1.32479578e-01 -2.04313785e-01 -9.74563181e-01 -3.18434462e-02 -7.13821948e-01 4.03778255e-03 1.41280329e+00 2.54536490e-03 -8.01304221e-01 9.25563693e-01 7.58438289e-01 -2.49500081e-01 -5.24123788e-01 -1.25110531e+00 -5.78511417e-01 -1.33273169e-01 -1.67766660e-01 5.91286361e-01 5.51335156e-01 -4.70898241e-01 1.93867192e-01 -4.07897592e-01 3.05634677e-01 5.02056539e-01 -1.37908474e-01 9.03576672e-01 -1.25079679e+00 -1.74689561e-01 -2.88198680e-01 -4.98892695e-01 -8.09398890e-01 4.27486360e-01 -8.28992009e-01 -2.10850984e-02 -1.10075390e+00 2.55053878e-01 -2.33742863e-01 1.77154139e-01 8.03660452e-01 3.33800539e-02 6.83855653e-01 9.72747654e-02 2.41838600e-02 -7.50458837e-02 6.02461457e-01 1.29785597e+00 -4.48383987e-02 2.42701903e-01 -1.92550763e-01 -6.97281063e-01 7.54364729e-01 3.24052423e-01 -4.42706704e-01 -3.76866877e-01 -6.35361493e-01 -9.31994170e-02 -2.00872377e-01 6.15055978e-01 -8.89290690e-01 6.70620054e-02 3.92980397e-01 7.61520088e-01 -5.67546934e-02 7.85780489e-01 -7.55919158e-01 2.73124874e-01 4.36640471e-01 -1.79028660e-01 -4.21452075e-01 3.29643279e-01 3.02729666e-01 -1.98295061e-02 -3.17195833e-01 1.00139129e+00 -1.73078980e-02 -2.94000387e-01 3.66693139e-01 -1.88269496e-01 -2.46112466e-01 7.22667336e-01 -1.82739004e-01 -3.08499366e-01 -1.47091031e-01 -1.04100680e+00 -3.24538380e-01 5.19945621e-01 4.52777445e-01 6.11746371e-01 -1.03830123e+00 -8.64739895e-01 1.12884223e+00 -1.84548825e-01 -3.24506044e-01 3.90995204e-01 6.22407377e-01 -3.27934980e-01 3.68134588e-01 -3.46272320e-01 -4.95756388e-01 -1.43769062e+00 7.12710083e-01 5.13269424e-01 1.35769576e-01 -4.57207859e-01 9.53458011e-01 7.39140689e-01 -1.47846475e-01 3.99034321e-01 1.82276033e-02 4.79232408e-02 -5.87709360e-02 6.59919858e-01 1.66246239e-02 4.97937709e-01 -6.33028746e-01 -3.07708889e-01 4.67881203e-01 -1.97353840e-01 7.63292685e-02 1.26593983e+00 -6.37295246e-02 -1.24708943e-01 -3.56296629e-01 1.22747731e+00 -7.53320195e-03 -1.41906846e+00 -6.45032153e-02 -2.64622599e-01 -8.76736164e-01 -1.38620540e-01 -6.34044647e-01 -1.33131289e+00 1.25318575e+00 8.25964630e-01 -1.05012491e-01 1.28684986e+00 -4.64198962e-02 3.13700259e-01 1.53840393e-01 3.35867256e-01 -4.84724581e-01 7.57706314e-02 2.53295571e-01 1.19919622e+00 -1.01506472e+00 -2.50492599e-02 -7.48065948e-01 -1.60618156e-01 8.94611001e-01 4.91284400e-01 -1.57357752e-01 7.41165936e-01 4.70695674e-01 -4.69442420e-02 -2.57689834e-01 -8.62441659e-01 -1.95802063e-01 1.75036520e-01 6.85108840e-01 3.99919987e-01 -3.87273610e-01 -8.61891657e-02 3.13813835e-01 -5.99143878e-02 -1.43239945e-01 3.59977752e-01 6.51678562e-01 -1.46815240e-01 -1.31182957e+00 -5.09561419e-01 2.59583980e-01 -5.63972652e-01 2.16694139e-02 -3.85261685e-01 6.20842040e-01 5.04462361e-01 7.28243113e-01 2.03129038e-01 -5.28383814e-02 3.19768399e-01 4.25737381e-01 9.01770771e-01 -7.51352847e-01 -4.43052828e-01 9.59372744e-02 -9.78246257e-02 -7.28520930e-01 -3.46229762e-01 -7.89624095e-01 -8.82395566e-01 -3.93677831e-01 -3.93963546e-01 -3.68693769e-01 8.31464112e-01 9.23770249e-01 5.53497076e-01 5.51770985e-01 5.57397842e-01 -1.04420435e+00 -5.53415835e-01 -9.99858856e-01 -5.99088788e-01 5.82693100e-01 4.68700349e-01 -8.17081094e-01 -3.80465120e-01 2.00687975e-01]	[13.070182800292969, 0.3230769634246826]
5d3352e5-1a8d-4a7b-a847-9951c7334663	discrete-optimization-for-unsupervised	2005.01791	null	https://arxiv.org/abs/2005.01791v1	https://arxiv.org/pdf/2005.01791v1.pdf	Discrete Optimization for Unsupervised Sentence Summarization with Word-Level Extraction	Automatic sentence summarization produces a shorter version of a sentence, while preserving its most important information. A good summary is characterized by language fluency and high information overlap with the source sentence. We model these two aspects in an unsupervised objective function, consisting of language modeling and semantic similarity metrics. We search for a high-scoring summary by discrete optimization. Our proposed method achieves a new state-of-the art for unsupervised sentence summarization according to ROUGE scores. Additionally, we demonstrate that the commonly reported ROUGE F1 metric is sensitive to summary length. Since this is unwillingly exploited in recent work, we emphasize that future evaluation should explicitly group summarization systems by output length brackets.	['Yao Lu', 'Lili Mou', 'Olga Vechtomova', 'Katja Markert', 'Raphael Schumann']	2020-05-04	discrete-optimization-for-unsupervised-1	https://aclanthology.org/2020.acl-main.452	https://aclanthology.org/2020.acl-main.452.pdf	acl-2020-6	['abstractive-sentence-summarization', 'unsupervised-sentence-summarization']	['natural-language-processing', 'natural-language-processing']	[ 2.88009375e-01 4.86111552e-01 -4.47153866e-01 -3.50365907e-01 -1.13535464e+00 -4.93096888e-01 4.60020632e-01 9.64682579e-01 -3.96541178e-01 1.08706462e+00 1.13244951e+00 3.46514806e-02 -6.28189594e-02 -4.87583399e-01 -3.08196843e-01 -2.29656816e-01 8.32571536e-02 3.38606715e-01 -2.53708009e-02 -3.74488384e-01 8.69306505e-01 9.47121903e-02 -1.28662956e+00 3.93517941e-01 1.68948710e+00 2.96741635e-01 4.86034244e-01 8.95213127e-01 -2.47990340e-01 8.17017078e-01 -1.16020000e+00 -5.05658984e-01 -3.81295979e-01 -9.97235298e-01 -1.11164320e+00 1.48501666e-02 5.53196251e-01 2.71886587e-02 5.02407961e-02 1.14037335e+00 4.63384062e-01 3.55661929e-01 9.05219555e-01 -6.68751121e-01 -6.38821900e-01 9.93145525e-01 -2.72992820e-01 3.27354819e-01 8.30931604e-01 -1.45382538e-01 1.44587529e+00 -4.76734519e-01 8.62776577e-01 1.14706242e+00 4.05843109e-01 3.58009517e-01 -1.21854830e+00 -4.28098552e-02 1.06787756e-01 -4.79180217e-02 -8.65125954e-01 -5.29101789e-01 8.53730857e-01 -3.24707508e-01 1.27885723e+00 5.98342061e-01 6.08492196e-01 7.42001057e-01 5.91242909e-01 9.13063169e-01 6.61752641e-01 -5.10683119e-01 2.73177296e-01 1.12375610e-01 6.93679273e-01 6.38329923e-01 4.91903961e-01 -7.64208138e-01 -6.73966944e-01 -1.72235340e-01 -6.32097293e-03 -4.53825265e-01 -2.42317691e-01 6.76958039e-02 -1.22929990e+00 9.83152747e-01 3.22687849e-02 5.17282307e-01 -5.82662106e-01 -1.66614830e-01 6.42209053e-01 4.15456295e-01 8.80323648e-01 1.06247187e+00 -1.99090719e-01 -4.47223514e-01 -1.54308116e+00 2.88498700e-01 1.11344564e+00 8.54779601e-01 6.65319860e-01 -2.84458268e-02 -5.39906085e-01 9.87342894e-01 -2.33437512e-02 3.11486334e-01 7.09725440e-01 -1.11885524e+00 5.20817518e-01 7.24654496e-01 -8.84400383e-02 -9.80234921e-01 -4.25296158e-01 -6.27259433e-01 -9.22854006e-01 -5.63916564e-01 -1.77931473e-01 -7.32110739e-02 -2.72034138e-01 1.62385964e+00 -3.19935203e-01 -4.59171116e-01 3.44435781e-01 4.57955778e-01 1.18740189e+00 6.42026603e-01 -4.57262397e-02 -9.79552448e-01 1.11766875e+00 -1.27750170e+00 -1.12744689e+00 -2.32935041e-01 8.58289123e-01 -8.42495799e-01 1.11202514e+00 1.59492478e-01 -1.55579603e+00 -4.76441652e-01 -1.12167382e+00 -2.68677533e-01 3.21980491e-02 4.64542866e-01 4.84291673e-01 6.60809219e-01 -1.27318025e+00 8.92386675e-01 -5.56921124e-01 -8.72446358e-01 6.55650496e-02 1.58905745e-01 -2.68037111e-01 3.62204999e-01 -1.06236517e+00 9.50095654e-01 6.79915309e-01 -5.47933042e-01 -1.52388766e-01 -4.72956985e-01 -9.49063182e-01 1.71354547e-01 1.80175990e-01 -1.20177531e+00 1.31220365e+00 -7.22526371e-01 -1.64786267e+00 8.69924605e-01 -5.24669826e-01 -7.28149474e-01 2.89625943e-01 -2.66907454e-01 -2.95655787e-01 5.52166760e-01 4.91155982e-01 4.58955169e-01 5.62251329e-01 -9.47444320e-01 -4.13257211e-01 -2.72865504e-01 -1.45754233e-01 4.34560597e-01 -5.33676684e-01 2.59428024e-01 4.06615529e-03 -7.53279686e-01 -1.66334659e-01 -3.05162221e-01 -2.73401141e-01 -9.50619459e-01 -6.43331289e-01 -3.85289311e-01 2.51944184e-01 -9.05428112e-01 1.84248233e+00 -1.51552296e+00 3.41125309e-01 -2.49558151e-01 2.54615009e-01 1.63750082e-01 -1.99264631e-01 9.70706403e-01 2.09799051e-01 3.51591796e-01 -4.07625526e-01 -5.43064117e-01 1.00455610e-02 -2.67148346e-01 -2.98978955e-01 2.39968941e-01 1.50779814e-01 9.62182581e-01 -1.20243979e+00 -7.93964446e-01 -1.24338098e-01 -2.17325121e-01 -6.67073667e-01 2.83944935e-01 -2.35905677e-01 1.41090259e-01 -4.16421741e-01 2.95147061e-01 2.98542529e-01 -1.14692442e-01 1.04987398e-01 -7.95503631e-02 -2.99904883e-01 5.56680381e-01 -5.03243506e-01 2.07030439e+00 -2.95765996e-01 5.99451840e-01 -2.95605808e-01 -9.83921826e-01 1.03008127e+00 9.99952257e-02 2.90792316e-01 -2.80496508e-01 2.42151432e-02 2.02136680e-01 -1.38468236e-01 -5.87396443e-01 1.50261915e+00 -6.33802041e-02 -3.79309744e-01 5.70535660e-01 2.73348957e-01 -5.05690515e-01 7.70393729e-01 7.24857509e-01 1.07063138e+00 -2.90215403e-01 8.34601700e-01 -6.21571958e-01 5.79745591e-01 9.49615166e-02 2.54598796e-01 1.00428402e+00 -1.07305765e-01 6.48593724e-01 8.55182469e-01 3.00422937e-01 -1.09547806e+00 -9.53944266e-01 1.82861000e-01 9.75293517e-01 -9.24544707e-02 -9.75380421e-01 -1.10304356e+00 -7.58752942e-01 -2.62348324e-01 1.40211451e+00 -3.07794243e-01 -3.26920271e-01 -4.64492530e-01 -4.20830101e-01 5.70863366e-01 2.44434386e-01 2.34046057e-01 -1.09066224e+00 -3.84986818e-01 2.72198021e-01 -6.29345477e-01 -8.29396009e-01 -8.36886883e-01 -1.39860302e-01 -1.11938858e+00 -6.84155703e-01 -8.49706173e-01 -8.21032524e-01 5.33174038e-01 2.60661811e-01 1.27068257e+00 1.13090337e-03 2.42508233e-01 4.28180784e-01 -6.40800536e-01 -3.02350074e-01 -9.31842387e-01 7.52093613e-01 1.32881269e-01 -3.62457991e-01 1.85769930e-01 -3.88789505e-01 -3.73108894e-01 -3.98112357e-01 -9.56854641e-01 1.36673003e-01 4.78962809e-01 8.01075578e-01 3.44624072e-01 -2.07820952e-01 1.07990336e+00 -8.37219715e-01 1.52042651e+00 -2.81844437e-01 2.47930348e-01 4.32006896e-01 -5.95092356e-01 2.48373240e-01 8.22805285e-01 -3.80696845e-03 -1.13698018e+00 -4.75475460e-01 -1.91953659e-01 3.39035451e-01 -9.03768092e-02 7.55779445e-01 3.46699394e-02 5.40176749e-01 6.70124054e-01 5.34145951e-01 8.51469338e-02 -4.19646919e-01 3.83035958e-01 9.38807189e-01 5.78670204e-01 -3.61331969e-01 1.90186575e-01 1.34950459e-01 -2.99188524e-01 -1.31736052e+00 -1.13036203e+00 -7.79516935e-01 -7.44100034e-01 -1.82274520e-01 7.06608832e-01 -6.42864108e-01 -3.47491324e-01 5.42715266e-02 -1.52052498e+00 2.02716127e-01 -6.14108384e-01 4.20613050e-01 -7.93688953e-01 1.07404268e+00 -3.58896196e-01 -8.33610356e-01 -9.04008150e-01 -7.74614871e-01 1.08542693e+00 2.41502255e-01 -9.11219299e-01 -1.03340721e+00 3.41760278e-01 5.01385152e-01 1.84802711e-01 -2.28667688e-02 8.82114053e-01 -1.05025804e+00 8.37187618e-02 -2.22408697e-01 4.36491659e-03 4.49794590e-01 1.78579971e-01 -9.85091403e-02 -4.35835272e-01 -2.51987696e-01 2.66127735e-01 -2.82640845e-01 1.38523960e+00 5.40727437e-01 7.19303906e-01 -6.21794939e-01 8.38646013e-03 5.17458729e-02 1.12318718e+00 -1.78443030e-01 4.76308703e-01 2.53987312e-01 3.87361437e-01 7.53905118e-01 7.01518059e-01 4.76861268e-01 4.80847538e-01 2.70043194e-01 -2.08559647e-01 2.49148443e-01 -1.48724541e-01 -4.90922779e-01 5.77789307e-01 1.58130050e+00 1.84676632e-01 -4.29382712e-01 -5.44340193e-01 6.44249260e-01 -2.02612567e+00 -1.22777355e+00 -2.22039804e-01 2.11034083e+00 9.32299495e-01 2.09341943e-01 3.58402252e-01 -2.28310619e-02 6.96674168e-01 5.26955009e-01 -1.44169524e-01 -9.56864595e-01 -4.00692075e-01 -2.77998298e-02 1.77787259e-01 7.58195698e-01 -7.76687980e-01 1.20013642e+00 6.78426123e+00 9.78842676e-01 -6.12374485e-01 -1.19827595e-03 4.35627490e-01 -8.30067322e-02 -6.51887953e-01 5.69254048e-02 -6.11714780e-01 4.62071359e-01 1.03100944e+00 -9.40100789e-01 5.98799139e-02 4.77636993e-01 6.87515557e-01 -4.53314483e-01 -8.75043809e-01 6.15132749e-01 5.94925463e-01 -1.29360425e+00 4.32597280e-01 -2.81457692e-01 9.21330929e-01 -3.30710530e-01 -3.35916996e-01 2.72860467e-01 -1.43201472e-02 -7.93204069e-01 5.65743148e-01 7.28311360e-01 5.81608593e-01 -7.96617270e-01 7.86991000e-01 5.53775549e-01 -8.18394601e-01 2.03652501e-01 -4.30952787e-01 -2.00814113e-01 4.05730069e-01 7.65119612e-01 -7.26079464e-01 9.55347955e-01 -1.62008375e-01 1.00541520e+00 -6.81857646e-01 9.47240949e-01 -2.93733776e-01 6.88206673e-01 1.54396266e-01 -5.00330567e-01 1.82245225e-01 -4.72876191e-01 1.05886078e+00 1.64709973e+00 2.55870461e-01 -7.42689986e-03 2.15425968e-01 6.84229195e-01 -1.57824129e-01 7.35643685e-01 -6.81009650e-01 -3.73377293e-01 2.40304351e-01 9.45671260e-01 -7.12483227e-01 -6.51846588e-01 3.98343056e-02 1.17485666e+00 4.25385177e-01 1.45469174e-01 -3.45877349e-01 -6.52842879e-01 2.63536513e-01 -1.11877568e-01 -3.63193229e-02 -1.72511861e-01 -6.30625010e-01 -1.43848014e+00 6.79026246e-02 -6.68362200e-01 2.46573031e-01 -5.21422029e-01 -1.05960584e+00 6.92364752e-01 -2.48790309e-02 -1.02271545e+00 -3.93547118e-01 1.95083871e-01 -8.03155184e-01 6.56589746e-01 -1.27992666e+00 -7.62429714e-01 6.98197335e-02 -1.48045823e-01 1.10128474e+00 -3.86346132e-01 8.21736813e-01 -2.22201347e-01 -6.12545669e-01 5.69804609e-01 2.83434093e-01 -4.55491453e-01 8.21762979e-01 -1.45421660e+00 4.17606443e-01 8.06268215e-01 -2.14133840e-02 6.45830393e-01 1.29617047e+00 -9.76588845e-01 -9.39596832e-01 -9.51428413e-01 1.72655845e+00 -1.58423230e-01 6.27760112e-01 1.14846013e-01 -7.49959946e-01 3.41620535e-01 7.40284264e-01 -1.23662329e+00 8.41291487e-01 2.25489557e-01 4.52838466e-02 1.59175783e-01 -9.91609633e-01 6.85374975e-01 1.09717834e+00 -3.55905384e-01 -1.32856405e+00 5.44428825e-01 1.00979376e+00 -1.01233572e-02 -7.58986115e-01 1.15341134e-01 2.63739824e-01 -9.96749640e-01 5.77154338e-01 -6.65399969e-01 9.28981662e-01 1.79532409e-01 1.36957131e-02 -1.79617059e+00 -4.09919798e-01 -8.76465380e-01 -2.02386022e-01 1.41460299e+00 3.80454093e-01 -4.57750112e-01 6.56313777e-01 1.59581006e-01 -3.42294604e-01 -6.42494798e-01 -5.69294333e-01 -8.84841561e-01 2.24391043e-01 -1.19061386e-02 3.38653684e-01 6.46494627e-01 7.25983441e-01 8.95548701e-01 -2.75172174e-01 -4.39532816e-01 5.88983238e-01 1.47344932e-01 6.76005304e-01 -1.14417160e+00 -5.75133367e-04 -9.54982817e-01 -1.98712677e-01 -1.13477087e+00 6.46088481e-01 -1.12953067e+00 -1.07824020e-01 -2.06790090e+00 7.56639898e-01 4.58793819e-01 3.54240797e-02 -1.88372552e-01 -4.42786217e-01 -6.42209053e-02 2.50784844e-01 2.36355186e-01 -1.16310811e+00 8.39654803e-01 1.23426223e+00 -2.34316602e-01 -4.99075145e-01 8.31967220e-02 -1.19329512e+00 7.14092016e-01 9.54145372e-01 -5.25329053e-01 -2.52902538e-01 -1.82925567e-01 1.43406853e-01 3.41756910e-01 -2.04919145e-01 -7.84833133e-01 3.35145712e-01 -1.80732638e-01 -2.17834502e-01 -8.73391807e-01 6.25531524e-02 -4.54756338e-03 -3.42392117e-01 4.46069360e-01 -9.09968734e-01 8.38255882e-02 -8.82539004e-02 4.91191298e-01 -4.59876150e-01 -7.94309556e-01 4.39588636e-01 -1.63592026e-01 -2.28797063e-01 -3.26161921e-01 -6.37658060e-01 3.85807514e-01 5.45737028e-01 -3.51835459e-01 -2.02674165e-01 -6.89415157e-01 -3.77579123e-01 2.66929299e-01 5.72061777e-01 3.06037217e-01 5.69182813e-01 -8.93991470e-01 -1.26315379e+00 -3.53135705e-01 1.07201912e-01 -4.80021477e-01 2.04905331e-01 7.95198023e-01 -4.44386959e-01 7.43740261e-01 2.34698709e-02 -2.80923754e-01 -1.50862551e+00 2.44502336e-01 -2.21531570e-01 -7.41934180e-01 -6.03864133e-01 4.05360758e-01 -9.15214717e-02 -2.25818679e-01 8.91263187e-02 -3.16239953e-01 -5.33116639e-01 4.67575252e-01 5.67122936e-01 7.38547802e-01 1.37286201e-01 -6.12223864e-01 -1.45734385e-01 2.58949548e-01 -2.63003677e-01 -2.27965176e-01 1.24203730e+00 -3.18944812e-01 -5.17956316e-01 6.04590178e-01 1.30660832e+00 3.99343908e-01 -4.01160598e-01 -1.87531784e-02 2.29061097e-01 -1.85904741e-01 -5.10597341e-02 -7.46556699e-01 -2.96037316e-01 4.29356098e-01 -2.36390635e-01 5.36947370e-01 1.03480279e+00 9.77806300e-02 8.82604718e-01 5.91478169e-01 5.76902963e-02 -1.26187992e+00 1.67444646e-01 8.67744803e-01 1.09444702e+00 -1.09059262e+00 3.36311847e-01 -3.80340904e-01 -8.89164209e-01 9.89037514e-01 2.38651469e-01 -1.31039962e-01 5.83814867e-02 -1.56805843e-01 -2.84796804e-01 -1.41918704e-01 -7.75331616e-01 -1.23217687e-01 4.32035804e-01 2.99945921e-01 7.98341632e-01 2.13434860e-01 -1.42033446e+00 9.71340597e-01 -7.53395677e-01 -2.58706450e-01 8.77510488e-01 5.83253026e-01 -9.53309417e-01 -9.89027083e-01 -1.14628643e-01 8.66812348e-01 -6.14642739e-01 -2.70784229e-01 -8.50519955e-01 2.04929784e-01 -6.78189695e-01 1.33834803e+00 -4.88451123e-02 -1.29634202e-01 3.87097806e-01 4.84019741e-02 4.89658713e-01 -1.03825736e+00 -9.43292618e-01 -2.35546497e-03 5.52493751e-01 -5.83524778e-02 -4.18446541e-01 -8.00772011e-01 -1.11659133e+00 -2.58625031e-01 -3.43292832e-01 6.58040524e-01 4.88351613e-01 9.22426522e-01 4.09089088e-01 6.55459583e-01 6.77446365e-01 -6.39367163e-01 -8.51149440e-01 -1.30155480e+00 -7.08428204e-01 3.64435107e-01 1.85670570e-01 -2.61279318e-04 -3.38459402e-01 -7.60170817e-02]	[12.467926979064941, 9.500500679016113]
29b86639-641d-458b-84a1-95a3fe9d87f9	contrastive-learning-and-self-training-for	2105.02001	null	https://arxiv.org/abs/2105.02001v1	https://arxiv.org/pdf/2105.02001v1.pdf	Contrastive Learning and Self-Training for Unsupervised Domain Adaptation in Semantic Segmentation	Deep convolutional neural networks have considerably improved state-of-the-art results for semantic segmentation. Nevertheless, even modern architectures lack the ability to generalize well to a test dataset that originates from a different domain. To avoid the costly annotation of training data for unseen domains, unsupervised domain adaptation (UDA) attempts to provide efficient knowledge transfer from a labeled source domain to an unlabeled target domain. Previous work has mainly focused on minimizing the discrepancy between the two domains by using adversarial training or self-training. While adversarial training may fail to align the correct semantic categories as it minimizes the discrepancy between the global distributions, self-training raises the question of how to provide reliable pseudo-labels. To align the correct semantic categories across domains, we propose a contrastive learning approach that adapts category-wise centroids across domains. Furthermore, we extend our method with self-training, where we use a memory-efficient temporal ensemble to generate consistent and reliable pseudo-labels. Although both contrastive learning and self-training (CLST) through temporal ensembling enable knowledge transfer between two domains, it is their combination that leads to a symbiotic structure. We validate our approach on two domain adaptation benchmarks: GTA5 $\rightarrow$ Cityscapes and SYNTHIA $\rightarrow$ Cityscapes. Our method achieves better or comparable results than the state-of-the-art. We will make the code publicly available.	['Bin Yang', 'Mario Döbler', 'Alexander Bartler', 'Robert A. Marsden']	2021-05-05	null	null	null	null	['synthetic-to-real-translation']	['computer-vision']	[ 1.94318622e-01 8.96110460e-02 -5.53152561e-02 -5.62780261e-01 -7.22073197e-01 -8.92340064e-01 6.61578834e-01 -8.26165918e-03 -5.47497928e-01 8.51954579e-01 -3.75186265e-01 -1.41710237e-01 -4.63169366e-02 -9.19745088e-01 -8.06874096e-01 -6.05046034e-01 1.86089456e-01 7.78747976e-01 5.24703443e-01 -2.11468115e-01 -1.43208385e-01 2.37365097e-01 -1.29953945e+00 1.63059175e-01 1.15329003e+00 9.10036206e-01 1.28827512e-01 2.29707971e-01 -4.20732617e-01 3.33555222e-01 -6.43940032e-01 -4.70246613e-01 3.91332597e-01 -7.42985249e-01 -1.14963412e+00 -8.26194359e-04 4.06926662e-01 8.41529295e-02 -5.42496564e-03 1.21273148e+00 2.97743618e-01 2.48223796e-01 9.36230898e-01 -1.19591737e+00 -7.49931335e-01 2.15597898e-01 -3.75758201e-01 -2.57438868e-02 -4.64514717e-02 6.70716912e-02 5.77552140e-01 -4.67670053e-01 8.98466468e-01 8.76193285e-01 8.97289634e-01 8.31484735e-01 -1.44647241e+00 -6.94433630e-01 2.10308284e-01 1.00308955e-01 -1.32574296e+00 -2.08361536e-01 7.51067877e-01 -4.81235534e-01 6.53187513e-01 -1.17731370e-01 4.19693202e-01 1.42062557e+00 -2.73846954e-01 5.06899595e-01 1.22044408e+00 -4.34051245e-01 5.98502636e-01 2.62371182e-01 -9.90315992e-03 3.37991804e-01 -2.56456975e-02 1.33345157e-01 -3.13804001e-01 1.01547047e-01 7.46705592e-01 -2.10554868e-01 5.21387234e-02 -7.11765289e-01 -1.18819499e+00 8.64684403e-01 5.40675819e-01 5.53912580e-01 -1.75452143e-01 -5.82410730e-02 4.87544447e-01 3.43124628e-01 5.94760418e-01 6.51133597e-01 -7.25371480e-01 6.84372569e-03 -9.29256797e-01 1.57432392e-01 7.12841332e-01 9.29374516e-01 8.57466936e-01 9.24133807e-02 7.98234046e-02 1.10685003e+00 -4.32489105e-02 3.39875430e-01 7.29599535e-01 -1.02865076e+00 1.90921053e-01 5.35546124e-01 7.75575936e-02 -6.49976075e-01 -2.65761763e-01 -4.29240882e-01 -7.46331811e-01 3.88900280e-01 1.01542354e+00 -2.19063208e-01 -1.29628718e+00 2.10023856e+00 3.75986338e-01 2.78108686e-01 3.76005620e-01 8.17323029e-01 5.91784120e-01 4.31435108e-01 3.77963662e-01 2.48044923e-01 1.07031512e+00 -8.94432962e-01 -3.81908864e-01 -5.88411331e-01 5.44218183e-01 -5.24799109e-01 1.06480455e+00 1.98634490e-01 -7.36231565e-01 -7.17248857e-01 -9.72173214e-01 4.24369760e-02 -7.90894926e-01 -1.14942290e-01 4.40901756e-01 5.24137080e-01 -8.41788054e-01 7.94490159e-01 -8.58293414e-01 -5.95474839e-01 6.49575531e-01 3.14299524e-01 -4.95900005e-01 3.87638435e-02 -1.34896910e+00 9.37849700e-01 6.89483345e-01 -5.01188636e-01 -8.74537408e-01 -7.12338269e-01 -8.97120535e-01 -2.98018306e-01 2.42204309e-01 -7.47561514e-01 1.23818719e+00 -1.66605389e+00 -1.57321906e+00 1.24389076e+00 7.48795792e-02 -6.68285728e-01 5.69019139e-01 -8.78361166e-02 -3.92298728e-01 2.00788900e-02 3.61984938e-01 1.17547941e+00 8.40069175e-01 -1.39845276e+00 -5.84715128e-01 -2.82973766e-01 -9.70900431e-02 6.68851882e-02 -3.87920767e-01 -3.67450893e-01 -4.38161701e-01 -8.72233689e-01 2.55756497e-01 -1.00396204e+00 -2.13300377e-01 -6.91606551e-02 -2.49971539e-01 -1.79708630e-01 8.44856679e-01 -5.33123493e-01 7.14575946e-01 -2.21436715e+00 1.43423960e-01 -1.57574620e-02 -1.11634381e-01 4.04657304e-01 -1.99487820e-01 3.29528116e-02 -2.74374902e-01 3.11100436e-03 -7.99481034e-01 -2.89076984e-01 -3.80539261e-02 4.76747602e-01 -3.11487645e-01 3.60717207e-01 3.98206085e-01 8.36501837e-01 -1.11114728e+00 -4.03695554e-01 5.24845682e-02 3.87121886e-01 -4.22549188e-01 1.67547062e-01 -5.45996249e-01 8.28545630e-01 -3.02015126e-01 3.55743736e-01 7.26500213e-01 -2.31965587e-01 2.61794955e-01 1.10897578e-01 3.11732292e-01 1.80437788e-01 -9.90112305e-01 2.15181756e+00 -4.31910932e-01 5.21766186e-01 -2.35435113e-01 -1.42236555e+00 1.15023029e+00 1.88114583e-01 5.31359553e-01 -8.48573387e-01 3.63378264e-02 4.14335936e-01 -1.60825178e-01 -2.19065011e-01 2.93082148e-01 -4.67173189e-01 -2.88811088e-01 2.01832816e-01 4.07625198e-01 -3.13840926e-01 1.45548016e-01 -1.22103050e-01 1.00074100e+00 6.55266702e-01 1.52230620e-01 -2.45437622e-01 3.86422902e-01 3.97422969e-01 7.74209917e-01 4.90499049e-01 -3.95476431e-01 9.42451715e-01 3.94229949e-01 -4.86763149e-01 -1.22265542e+00 -1.37382400e+00 -2.28778079e-01 1.04535067e+00 3.59122187e-01 1.37090608e-01 -9.61351931e-01 -1.22603524e+00 -4.56639333e-03 8.35950673e-01 -8.03227544e-01 -2.31217489e-01 -4.98610705e-01 -6.37190700e-01 7.07639217e-01 6.53655469e-01 8.25389147e-01 -1.19153953e+00 -3.98417622e-01 2.79807299e-01 -2.38851398e-01 -1.20890367e+00 -2.71346688e-01 4.81891930e-01 -9.22053039e-01 -9.75853860e-01 -1.03818357e+00 -9.90423560e-01 7.85749257e-01 -2.60841608e-01 1.38248038e+00 -3.92186761e-01 7.60366768e-02 2.32790396e-01 -3.89971554e-01 -3.13990682e-01 -5.55994391e-01 2.48133451e-01 4.33987975e-02 -6.26731068e-02 6.21032715e-01 -7.10520387e-01 -4.92176503e-01 4.80256289e-01 -9.42914665e-01 -8.30064043e-02 2.09545746e-01 9.74758148e-01 8.04657400e-01 -1.01540266e-02 8.62726331e-01 -1.16068363e+00 2.51753092e-01 -6.59404337e-01 -5.77993989e-01 1.05378002e-01 -4.83125985e-01 2.29342073e-01 9.15287435e-01 -6.78896487e-01 -1.18309557e+00 3.34991187e-01 -2.46036842e-01 -4.90825146e-01 -6.95230842e-01 8.56869668e-02 -1.35496572e-01 1.64601043e-01 1.03526783e+00 1.69364274e-01 -7.71420673e-02 -4.13861036e-01 5.58203280e-01 3.98915201e-01 8.57907414e-01 -8.15420985e-01 7.82009661e-01 6.19718432e-01 -2.70902812e-01 -4.52176839e-01 -7.89590657e-01 -3.01624417e-01 -1.00437355e+00 1.13381691e-01 1.07330370e+00 -7.80135870e-01 -3.76765504e-02 5.78379035e-01 -1.00626707e+00 -7.54667163e-01 -6.55383706e-01 1.59009054e-01 -7.55203068e-01 2.25700960e-01 -2.84203053e-01 -3.75812411e-01 -1.16455443e-01 -1.01908743e+00 8.81217718e-01 3.05185884e-01 -3.69504482e-01 -1.27346265e+00 1.66615486e-01 1.66622832e-01 3.39604646e-01 5.65740347e-01 6.47176087e-01 -1.03602469e+00 -2.74387389e-01 -2.16980763e-02 -1.85396343e-01 6.10916913e-01 2.22793266e-01 -3.73967558e-01 -9.99834776e-01 -1.66088119e-01 -2.31833652e-01 -4.17529255e-01 8.69806826e-01 2.86596298e-01 1.02534497e+00 3.45342457e-02 -4.75027889e-01 7.00386107e-01 1.45851719e+00 3.79592896e-01 7.37923265e-01 7.88762033e-01 4.64165747e-01 7.58652389e-01 7.01304734e-01 1.36135593e-02 2.95033604e-01 7.12925375e-01 2.87536114e-01 -1.51162848e-01 -3.47372204e-01 -2.66856432e-01 1.09168373e-01 4.37422484e-01 1.60697654e-01 -1.63374931e-01 -1.16944218e+00 9.64222789e-01 -1.72854149e+00 -7.03038394e-01 1.04082689e-01 2.24002981e+00 1.11555886e+00 2.69310594e-01 1.67812854e-01 -2.64427308e-02 6.63611770e-01 -1.73410565e-01 -7.90537894e-01 -2.59693444e-01 -1.62872821e-01 4.39398885e-01 4.89792407e-01 2.31938139e-01 -1.44163966e+00 1.16032505e+00 5.78872824e+00 1.01764798e+00 -1.19424033e+00 3.99068683e-01 7.02926874e-01 2.61229992e-01 -1.58987939e-01 -2.03711614e-02 -4.76110309e-01 6.43989086e-01 9.44777071e-01 3.04477960e-02 2.32600376e-01 1.06191123e+00 -2.84026474e-01 -5.90414107e-02 -1.17414105e+00 7.15817392e-01 -1.88758016e-01 -1.20614719e+00 -2.87150621e-01 -1.22900441e-01 1.05315101e+00 9.09154266e-02 9.43477377e-02 4.30327594e-01 8.33403587e-01 -8.40150118e-01 7.16675878e-01 2.61410564e-01 1.07117796e+00 -6.79874122e-01 7.36944556e-01 3.38157326e-01 -9.85106051e-01 1.47338316e-01 -2.92550236e-01 2.19583228e-01 1.17277578e-02 4.57868755e-01 -8.79282534e-01 6.59778774e-01 8.79439592e-01 7.11512268e-01 -5.52845418e-01 7.55174637e-01 -3.14729869e-01 4.70734805e-01 -2.86939025e-01 3.37980181e-01 2.70216525e-01 -3.03779274e-01 4.01170462e-01 1.16389036e+00 2.60508925e-01 -1.93683967e-01 3.46652456e-02 1.02467632e+00 -1.16856039e-01 -1.00202337e-01 -7.64111698e-01 5.15230559e-02 4.36432809e-01 8.70197594e-01 -1.02285266e+00 -4.59869057e-01 -2.74059415e-01 1.30011988e+00 4.05081749e-01 4.31878388e-01 -8.14889550e-01 -4.30023968e-01 8.21510136e-01 3.42798829e-02 2.98610657e-01 -2.03216448e-01 -7.22466052e-01 -1.05384147e+00 -1.23414598e-01 -7.55016327e-01 5.00419736e-01 -5.92968285e-01 -1.58314741e+00 7.53138125e-01 6.23665266e-02 -1.33107913e+00 -3.78844470e-01 -4.24591154e-01 -4.03018177e-01 8.57202232e-01 -1.50265479e+00 -1.09508860e+00 -2.68509865e-01 6.18927121e-01 5.67379534e-01 -3.04264456e-01 9.79565859e-01 3.23733121e-01 -2.91501790e-01 6.49655223e-01 3.50785464e-01 3.40525001e-01 1.06853247e+00 -1.53443706e+00 5.27343154e-01 8.83855999e-01 1.74567565e-01 3.39225411e-01 6.29253626e-01 -5.68031430e-01 -4.62502301e-01 -1.30292165e+00 6.05331242e-01 -5.80381036e-01 5.51548481e-01 -2.52106965e-01 -1.26598155e+00 6.90520823e-01 2.55167633e-01 2.15492308e-01 5.33695459e-01 1.86488125e-02 -6.28450692e-01 -1.93033233e-01 -1.31843090e+00 4.13665295e-01 1.09529352e+00 -3.88830900e-01 -8.05585027e-01 2.27790043e-01 6.46280229e-01 -3.60506594e-01 -8.45025659e-01 4.54524070e-01 2.41144240e-01 -9.31482971e-01 9.93934155e-01 -5.45771658e-01 5.58609664e-01 -3.59465182e-01 2.12897565e-02 -1.57769465e+00 -7.59015009e-02 -1.54035136e-01 3.02971989e-01 1.44884646e+00 3.61449867e-01 -8.66278410e-01 9.60636854e-01 4.00806069e-01 -1.56454042e-01 -3.98073614e-01 -9.43322361e-01 -1.16722155e+00 6.86149776e-01 -4.02793556e-01 5.67688584e-01 1.35005319e+00 -2.63514072e-01 2.18203425e-01 1.21742509e-01 8.40312019e-02 5.44777691e-01 1.20657220e-01 7.14914620e-01 -1.34109926e+00 -1.86044052e-01 -4.94092047e-01 -2.91518986e-01 -8.10737431e-01 4.88098979e-01 -1.03333521e+00 2.11995825e-01 -1.44078720e+00 -1.16170079e-01 -8.59207273e-01 -2.04710469e-01 8.07660043e-01 5.69539815e-02 4.54084843e-01 -2.35817116e-02 1.51691779e-01 -6.54328942e-01 4.59512383e-01 1.21463728e+00 -1.91220075e-01 -3.36639225e-01 -1.86554849e-01 -5.54803550e-01 6.82063043e-01 9.04874802e-01 -7.17328191e-01 -4.35204118e-01 -4.79054958e-01 -1.33761466e-01 -3.05805504e-01 4.13135320e-01 -1.17003930e+00 1.22629024e-01 -7.13256523e-02 3.91491383e-01 -1.67190418e-01 1.85960770e-01 -8.25914502e-01 2.30304956e-01 2.36622930e-01 -1.39558688e-01 -3.08242202e-01 4.86030430e-01 5.41173398e-01 -4.20055509e-01 -3.38529021e-01 1.22023880e+00 -2.73525417e-01 -1.07305825e+00 8.49005952e-02 -1.41614839e-01 3.26214552e-01 1.16985834e+00 -3.89380246e-01 -1.73053399e-01 -1.17691912e-01 -1.06723797e+00 5.36077842e-02 7.78856218e-01 4.51702446e-01 2.37677261e-01 -1.28183115e+00 -5.28686583e-01 1.97686881e-01 1.44281685e-01 3.31657290e-01 1.91352293e-01 4.07129049e-01 -5.09311438e-01 1.67429507e-01 -4.50961292e-01 -7.46096611e-01 -7.72936165e-01 5.70770681e-01 6.19386554e-01 -2.75733739e-01 -3.66866082e-01 9.16052461e-01 2.85155028e-01 -8.96557748e-01 7.69974589e-02 7.04456717e-02 -9.77987573e-02 -3.69285233e-03 1.01468921e-01 1.39606312e-01 9.70209315e-02 -5.34585774e-01 -3.79463822e-01 4.88389432e-01 9.21441317e-02 -9.55402479e-02 1.31406248e+00 -3.30513157e-02 1.51710689e-01 2.75631040e-01 1.17006207e+00 -2.88519591e-01 -1.62156308e+00 -3.05843174e-01 2.10417777e-01 -2.11411193e-01 -3.34703326e-01 -1.04044163e+00 -1.06817687e+00 8.25412810e-01 7.61780620e-01 2.10836053e-01 1.32076359e+00 1.69824585e-01 8.29982340e-01 2.63778977e-02 4.48190302e-01 -1.22691607e+00 2.51008868e-01 6.46460116e-01 5.59958756e-01 -1.42785418e+00 -3.75364304e-01 -2.95140415e-01 -8.14442813e-01 9.33157146e-01 8.42387557e-01 -2.46032611e-01 4.96741205e-01 3.82471532e-02 3.20772797e-01 4.39414568e-02 -2.44320050e-01 -5.43079674e-01 1.78877234e-01 1.07728076e+00 2.40523562e-01 -1.67307146e-02 -1.69911772e-01 6.62775815e-01 -1.80666745e-01 6.17914870e-02 1.54574350e-01 8.44327092e-01 -6.89479560e-02 -1.49141514e+00 -3.89096439e-01 1.01308838e-01 -4.47193027e-01 1.12606116e-01 -2.67057180e-01 1.00192261e+00 5.65587282e-01 7.72127390e-01 2.39106908e-01 -2.45114401e-01 3.25372994e-01 4.30196404e-01 3.68854344e-01 -6.99297905e-01 -4.06281561e-01 -1.19920336e-01 -9.33200940e-02 -3.24610800e-01 -4.94245291e-01 -6.25221848e-01 -1.42590177e+00 -1.58413891e-02 6.69057621e-03 1.25217885e-01 6.16621256e-01 1.01769209e+00 3.75230998e-01 5.91229260e-01 2.93292910e-01 -6.15267515e-01 -3.36747319e-01 -9.33776617e-01 -5.07876456e-01 8.40892136e-01 1.41426042e-01 -8.57261896e-01 -1.69193253e-01 4.34099048e-01]	[9.717194557189941, 1.4654121398925781]
c513074a-f181-4d02-9a3a-9451c4abfa2c	enhancing-embedding-representations-of	2303.13566	null	https://arxiv.org/abs/2303.13566v1	https://arxiv.org/pdf/2303.13566v1.pdf	Enhancing Embedding Representations of Biomedical Data using Logic Knowledge	Knowledge Graph Embeddings (KGE) have become a quite popular class of models specifically devised to deal with ontologies and graph structure data, as they can implicitly encode statistical dependencies between entities and relations in a latent space. KGE techniques are particularly effective for the biomedical domain, where it is quite common to deal with large knowledge graphs underlying complex interactions between biological and chemical objects. Recently in the literature, the PharmKG dataset has been proposed as one of the most challenging knowledge graph biomedical benchmark, with hundreds of thousands of relational facts between genes, diseases and chemicals. Despite KGEs can scale to very large relational domains, they generally fail at representing more complex relational dependencies between facts, like logic rules, which may be fundamental in complex experimental settings. In this paper, we exploit logic rules to enhance the embedding representations of KGEs on the PharmKG dataset. To this end, we adopt Relational Reasoning Network (R2N), a recently proposed neural-symbolic approach showing promising results on knowledge graph completion tasks. An R2N uses the available logic rules to build a neural architecture that reasons over KGE latent representations. In the experiments, we show that our approach is able to significantly improve the current state-of-the-art on the PharmKG dataset. Finally, we provide an ablation study to experimentally compare the effect of alternative sets of rules according to different selection criteria and varying the number of considered rules.	['Giuseppe Marra', 'Moreno Falaschi', 'Caterina Graziani', 'Stefano Fioravanti', 'Francesco Giannini', 'Michelangelo Diligenti']	2023-03-23	null	null	null	null	['knowledge-graph-embeddings', 'knowledge-graph-completion', 'knowledge-graph-embeddings', 'relational-reasoning']	['graphs', 'knowledge-base', 'methodology', 'natural-language-processing']	[ 5.21756113e-02 5.84720671e-01 -2.73869723e-01 -2.27942288e-01 2.18800418e-02 -2.16260433e-01 7.29319513e-01 8.25918972e-01 -2.89540082e-01 7.50390708e-01 2.62761146e-01 -2.83151150e-01 -8.01607907e-01 -1.20993543e+00 -9.14126754e-01 -4.63183194e-01 -4.25638139e-01 7.82364249e-01 1.42801091e-01 -3.05748314e-01 -3.23196143e-01 6.83166862e-01 -1.40845633e+00 3.81551862e-01 6.58945084e-01 6.36829317e-01 -3.09629679e-01 1.72516540e-01 -1.27021179e-01 9.36881840e-01 -2.66022980e-01 -7.82015979e-01 -1.60987198e-01 -6.88377395e-02 -8.17889571e-01 -4.48634684e-01 2.93961495e-01 2.65366763e-01 -6.61577344e-01 1.06967115e+00 3.04461658e-01 1.53621152e-01 7.85044789e-01 -1.24843407e+00 -9.10279632e-01 1.06793404e+00 1.25098303e-01 1.04436465e-01 3.12705338e-01 -4.51721810e-02 1.31370962e+00 -7.53060400e-01 9.18534517e-01 1.47385871e+00 5.72105527e-01 5.13421774e-01 -1.51391339e+00 -3.66694897e-01 6.69748932e-02 6.58041656e-01 -1.47147560e+00 -2.14529201e-01 6.48293078e-01 -3.57032984e-01 1.29429579e+00 7.78274536e-02 6.09326422e-01 1.27708721e+00 2.61188537e-01 3.68527681e-01 7.87624061e-01 -3.96000117e-01 3.39151293e-01 1.25174090e-01 4.17961627e-01 9.27367866e-01 7.92785466e-01 -4.69478630e-02 -5.32692790e-01 -2.95773745e-01 5.37041843e-01 1.02815554e-02 -4.35866117e-01 -7.77389646e-01 -1.18177247e+00 1.04155886e+00 8.14180732e-01 5.85241675e-01 -3.89893204e-01 3.11119288e-01 5.38787425e-01 2.03027651e-01 4.46851909e-01 5.99102080e-01 -4.82585132e-01 3.43251109e-01 -1.88883305e-01 2.12497354e-01 1.00463331e+00 7.38402009e-01 5.26919127e-01 -6.11444078e-02 -6.19085990e-02 7.24288166e-01 2.69454509e-01 -7.03686401e-02 3.09574485e-01 -3.06886286e-01 4.03437555e-01 1.03691900e+00 -3.77210617e-01 -1.38908744e+00 -6.16714776e-01 -3.62940729e-01 -1.01942325e+00 -1.89388052e-01 1.65853068e-01 3.75553936e-01 -7.30217576e-01 1.75495970e+00 2.59283245e-01 3.74604970e-01 4.36908096e-01 4.76187855e-01 1.03176844e+00 3.41524273e-01 2.88377523e-01 -4.83924337e-02 1.49982452e+00 -5.36827147e-01 -8.91540647e-01 2.44422480e-02 7.11173356e-01 4.81179319e-02 6.91798091e-01 4.06088114e-01 -6.69392705e-01 -2.44785503e-01 -1.03559494e+00 -1.36511296e-01 -1.16461122e+00 6.50844444e-03 1.16278374e+00 3.96730423e-01 -9.24069762e-01 9.29647982e-01 -9.17946696e-01 -5.60954094e-01 5.77263713e-01 5.66275418e-01 -8.00446868e-01 -3.26881796e-01 -1.74810362e+00 1.10502970e+00 9.45686817e-01 3.27625036e-01 -5.52682877e-01 -7.80961156e-01 -1.17228270e+00 4.06991124e-01 6.93278730e-01 -7.58338213e-01 3.20465237e-01 -4.34988365e-02 -1.24485695e+00 7.87491500e-01 3.65854681e-01 -7.30075896e-01 1.46561131e-01 -7.23273829e-02 -7.54057646e-01 8.99137259e-02 -2.66133368e-01 6.05329275e-01 3.54726911e-01 -8.54599059e-01 8.63133520e-02 -4.55778778e-01 3.79008055e-01 -2.18330145e-01 -5.71571469e-01 -4.56342876e-01 -4.01244372e-01 -4.46767449e-01 -1.52136818e-01 -8.51904333e-01 -2.42191285e-01 -2.27507487e-01 -5.54944575e-01 -5.07892132e-01 4.01532233e-01 -3.76229733e-01 1.09942710e+00 -2.03731632e+00 7.14878500e-01 3.52518290e-01 5.65117776e-01 3.19356382e-01 -7.84019828e-02 5.53019404e-01 -5.42393506e-01 1.83060616e-01 -2.69792378e-01 8.96826312e-02 1.96638748e-01 5.77604234e-01 -2.00992852e-01 2.80352503e-01 4.45699990e-01 1.25091875e+00 -9.95783150e-01 -2.57267445e-01 1.21779099e-01 7.54534841e-01 -5.76916873e-01 -1.50307104e-01 -5.94365716e-01 -1.24684408e-01 -4.65911329e-01 3.29281449e-01 2.80188441e-01 -5.02077341e-01 7.41785467e-01 -5.32178819e-01 4.42390651e-01 1.37717932e-01 -1.04634953e+00 1.65157199e+00 -2.74151683e-01 2.84630388e-01 -5.63295066e-01 -1.22187293e+00 8.62406671e-01 3.57145637e-01 3.78839761e-01 -2.82402635e-01 1.35511711e-01 3.95529605e-02 1.74919590e-01 -4.81538057e-01 1.53136969e-01 -1.98810220e-01 -7.80167477e-03 -1.02517895e-01 4.06900764e-01 2.75260508e-02 4.31909561e-01 4.46887106e-01 1.34144688e+00 -5.18890694e-02 6.49390519e-01 -1.76726505e-01 6.69010818e-01 -1.91335648e-01 3.49213958e-01 2.90603101e-01 2.73713082e-01 -1.84176102e-01 1.10843456e+00 -6.13327742e-01 -5.68538606e-01 -1.00485539e+00 -1.51878044e-01 4.98225152e-01 -1.25766411e-01 -6.71441674e-01 -3.88679564e-01 -8.28492522e-01 2.92510360e-01 6.96202576e-01 -9.97254729e-01 -6.13528728e-01 -2.37173468e-01 -8.09960127e-01 7.33065903e-01 3.98707628e-01 1.40971392e-02 -1.12756789e+00 -1.24493636e-01 2.17441127e-01 1.81172580e-01 -1.52390873e+00 3.07749271e-01 5.17513931e-01 -8.42324793e-01 -1.53394699e+00 -1.26989618e-01 -4.50876713e-01 6.75550699e-01 -4.05190289e-01 1.11287808e+00 3.27232853e-02 -5.25581539e-01 4.34998274e-01 -4.08724427e-01 -3.23606074e-01 -6.42742753e-01 9.74629298e-02 2.80487210e-01 5.90712763e-02 5.00441551e-01 -7.55944788e-01 -1.83471143e-01 -1.71238557e-02 -1.49920177e+00 -1.39622167e-01 5.85975945e-01 8.92783642e-01 5.22676170e-01 4.39067483e-01 6.75503016e-01 -1.45514655e+00 7.33588040e-01 -5.43498397e-01 -6.66155457e-01 5.37632585e-01 -7.19559789e-01 5.78311920e-01 8.64198625e-01 -2.32342139e-01 -5.32852769e-01 -1.68580621e-01 -3.52219716e-02 -6.03200912e-01 -5.93883246e-02 1.15903163e+00 -4.11780566e-01 -7.94425383e-02 7.82826245e-01 -1.79312229e-01 -1.70688882e-01 -3.47385436e-01 6.54772401e-01 1.47019327e-01 2.44008198e-01 -5.46267331e-01 7.10565031e-01 2.85229146e-01 6.37499869e-01 -6.64468765e-01 -7.90846229e-01 -1.68561578e-01 -6.67905211e-01 3.61226022e-01 1.00746727e+00 -6.76202297e-01 -9.74920809e-01 -1.04574837e-01 -1.06651092e+00 -1.28971070e-01 -3.07239503e-01 5.83242297e-01 -4.41143215e-01 4.00841504e-01 -6.29177511e-01 -2.74773568e-01 -8.43686461e-02 -9.66089129e-01 7.22355485e-01 -1.84736744e-01 -1.01005204e-01 -1.38183093e+00 1.10102378e-01 8.05946998e-03 1.48087174e-01 6.27010167e-01 1.75842214e+00 -9.51944768e-01 -5.21961391e-01 -1.75599322e-01 -1.87020957e-01 1.49817288e-01 3.42492908e-01 -2.04273701e-01 -7.84868956e-01 -1.15895249e-01 -5.30748487e-01 -3.25065702e-01 1.07470274e+00 -7.35165104e-02 1.04996765e+00 -2.60064363e-01 -5.37209809e-01 3.76609802e-01 1.53218555e+00 -1.85115755e-01 6.82368159e-01 -4.02344614e-02 9.63331521e-01 6.85184836e-01 1.47293225e-01 1.70989752e-01 2.62112767e-01 8.50755990e-01 5.01265824e-01 1.73809677e-01 -2.52320450e-02 -2.88854390e-01 2.99261808e-01 9.81691837e-01 -1.82726964e-01 -2.04195067e-01 -9.79816616e-01 2.71851301e-01 -1.85504019e+00 -7.03250825e-01 -7.39274696e-02 1.93743420e+00 9.60616112e-01 2.91660279e-01 -3.28365147e-01 2.18030289e-01 4.54569399e-01 3.51626500e-02 -4.43439841e-01 -3.04508954e-01 -1.44061431e-01 5.80388308e-01 3.66893023e-01 4.18336481e-01 -9.60582972e-01 1.06811118e+00 5.32609606e+00 7.09421039e-01 -8.94588828e-01 -2.64729448e-02 1.08799808e-01 2.82612890e-01 -4.51550692e-01 -8.36132392e-02 -6.91007257e-01 7.05283657e-02 1.17392814e+00 -6.71828049e-04 4.91981894e-01 6.14139318e-01 -2.74431258e-01 2.83931613e-01 -1.67927611e+00 9.78483140e-01 1.17238916e-01 -1.57989454e+00 5.50539136e-01 9.53853577e-02 3.51698428e-01 -1.77990839e-01 -1.81426495e-01 5.68971097e-01 4.69002753e-01 -1.30636883e+00 1.17001936e-01 7.06985176e-01 7.41151869e-01 -6.45789266e-01 7.91412652e-01 9.47995484e-02 -1.01763892e+00 -6.09674267e-02 -6.01525605e-01 2.54695952e-01 -6.73897415e-02 8.83657813e-01 -1.04440129e+00 1.16297150e+00 4.40218925e-01 9.26516354e-01 -8.34849000e-01 4.96292830e-01 -5.83394885e-01 2.16676161e-01 -1.95094168e-01 -7.33026266e-02 7.23252892e-02 -1.19868979e-01 2.63404191e-01 1.07050383e+00 5.10523617e-02 7.28375511e-04 -1.60780966e-01 1.04408467e+00 -2.83162147e-01 1.21207148e-01 -1.06586778e+00 -5.87359726e-01 1.80860776e-02 1.26866746e+00 -7.80398130e-01 -3.64727050e-01 -2.48512164e-01 7.44302332e-01 7.38673687e-01 4.48506504e-01 -8.62872303e-01 -2.77945995e-01 5.96899688e-01 -1.24519959e-01 3.17749590e-01 -7.13672116e-02 5.13779640e-01 -1.27143717e+00 -1.04989864e-01 -7.78006017e-01 5.97570479e-01 -6.38726771e-01 -1.41665983e+00 5.97434759e-01 2.76832759e-01 -7.99954295e-01 -7.83704743e-02 -1.20497978e+00 1.05305977e-01 4.43820715e-01 -1.62661493e+00 -1.10612893e+00 -5.63296266e-02 6.70236111e-01 -4.18458506e-02 -1.91264018e-01 1.35646665e+00 4.78674263e-01 -6.04526937e-01 4.12599564e-01 -7.57814990e-03 1.31554514e-01 4.16388154e-01 -1.32784820e+00 1.46097139e-01 3.69417727e-01 5.54846823e-01 1.10986483e+00 5.13247788e-01 -7.38142312e-01 -1.66579187e+00 -1.20877552e+00 8.29489946e-01 -5.93358874e-01 9.10912454e-01 -6.46919489e-01 -1.06827605e+00 8.55109155e-01 -2.03633994e-01 5.48835039e-01 8.44162047e-01 3.98382038e-01 -6.51469111e-01 -4.55909222e-02 -8.32994044e-01 7.94271290e-01 1.16533840e+00 -7.08619595e-01 -6.11855924e-01 4.49006587e-01 9.42913651e-01 -1.84062138e-01 -1.38784683e+00 6.59113050e-01 2.59359241e-01 -5.93841434e-01 1.06821251e+00 -1.02390623e+00 4.39316064e-01 -3.19782287e-01 -1.08704045e-01 -1.43661904e+00 -3.21834028e-01 -2.29596317e-01 -5.27302563e-01 9.62007761e-01 4.92369860e-01 -7.48917401e-01 6.51049793e-01 3.58581483e-01 6.79908600e-03 -9.61511910e-01 -9.49789405e-01 -7.83030987e-01 -2.18494639e-01 -2.92637616e-01 6.33211613e-01 1.36043227e+00 2.65634537e-01 5.63428938e-01 -1.43498436e-01 2.32527167e-01 2.80144483e-01 -9.36212167e-02 5.32160580e-01 -1.65119970e+00 -4.47363108e-01 -3.88645768e-01 -1.06138253e+00 -2.13829577e-01 5.69726408e-01 -1.47715235e+00 -3.83979529e-01 -1.72580254e+00 -2.68103778e-02 -2.93247581e-01 -6.12012446e-01 8.26071799e-01 -1.17133826e-01 -6.69896379e-02 -5.87942377e-02 -3.41367185e-01 -4.46529925e-01 6.80698454e-01 1.01666081e+00 -5.69869876e-01 -7.11925477e-02 -4.39361125e-01 -6.67072892e-01 5.59029341e-01 4.84800071e-01 -4.93786544e-01 -7.62745380e-01 -1.40981600e-01 9.91929591e-01 -8.61479715e-02 4.63169277e-01 -8.32419455e-01 2.67268240e-01 2.05655366e-01 1.52870312e-01 -9.70399231e-02 3.19390327e-01 -1.09991395e+00 5.06937683e-01 4.82241690e-01 -3.28107864e-01 -2.11897552e-01 3.77329230e-01 9.83773828e-01 -3.22628438e-01 1.75586436e-02 3.03454310e-01 8.42525139e-02 -6.85802400e-01 4.32750016e-01 -1.25319539e-02 -1.14048652e-01 9.55042422e-01 8.37972760e-02 -3.78128320e-01 5.30944467e-02 -1.17130089e+00 1.34647742e-01 2.72262767e-02 6.02359295e-01 7.60048509e-01 -1.41187668e+00 -4.07714695e-01 2.31128447e-02 6.00740969e-01 -8.31695646e-02 2.48222619e-01 8.41638446e-01 -4.49970037e-01 7.20339358e-01 -2.12352604e-01 -3.17534477e-01 -1.04607332e+00 1.13641548e+00 1.58483297e-01 -7.79469371e-01 -7.07490921e-01 6.88789845e-01 1.73915997e-01 -5.65672457e-01 2.19323292e-01 -8.20453227e-01 -6.89066589e-01 2.37972856e-01 6.28185347e-02 1.78902537e-01 2.49283209e-01 -3.42047811e-01 -5.24995208e-01 3.72717887e-01 -1.60027400e-03 3.80563557e-01 1.62820637e+00 6.21867895e-01 -4.39749539e-01 5.05477607e-01 1.05964386e+00 1.52888568e-02 -4.74507660e-01 -3.12659591e-01 3.84547174e-01 4.70341742e-02 -1.72999412e-01 -5.84918261e-01 -9.40327764e-01 8.06547582e-01 2.56844908e-01 1.90767288e-01 7.65299559e-01 2.04116195e-01 2.65443891e-01 8.55753601e-01 4.79119062e-01 -5.98284602e-01 -9.51175392e-02 4.51324522e-01 9.21402216e-01 -9.18813407e-01 4.02980477e-01 -7.51070380e-01 -3.25312495e-01 1.19906187e+00 3.46400499e-01 -2.27213069e-03 8.03235352e-01 -8.03260803e-02 -4.60602134e-01 -8.34447145e-01 -8.11611772e-01 -2.31493205e-01 4.06321049e-01 4.57015336e-01 3.51497471e-01 2.62955815e-01 -2.90889442e-01 6.54810429e-01 9.08639431e-02 1.71942055e-01 3.47694606e-01 5.62842250e-01 1.89256921e-01 -1.49793458e+00 6.30579963e-02 5.41088879e-01 -3.73400629e-01 -2.75428683e-01 -5.30794203e-01 9.70271111e-01 1.95001885e-01 4.82557356e-01 -4.85617131e-01 -3.25239778e-01 5.47832966e-01 3.03830296e-01 6.07230842e-01 -8.50760341e-01 -4.10420030e-01 -5.18227994e-01 1.66511565e-01 -6.90078020e-01 -6.56607568e-01 -1.62161961e-01 -1.38059580e+00 1.85700003e-02 -3.00626874e-01 2.72244781e-01 3.18220526e-01 9.15636837e-01 4.94415879e-01 9.65935767e-01 -9.56412479e-02 -2.99246013e-01 -3.29182386e-01 -8.04049671e-01 -7.78175294e-01 5.57887316e-01 -5.79032749e-02 -1.10119188e+00 -6.92331195e-02 -3.62483561e-02]	[8.646493911743164, 7.738701820373535]
8f303948-13bb-4773-9b20-d0a260937d0e	dosed-a-deep-learning-approach-to-detect	1812.04079	null	http://arxiv.org/abs/1812.04079v1	http://arxiv.org/pdf/1812.04079v1.pdf	DOSED: a deep learning approach to detect multiple sleep micro-events in EEG signal	Background: Electroencephalography (EEG) monitors brain activity during sleep and is used to identify sleep disorders. In sleep medicine, clinicians interpret raw EEG signals in so-called sleep stages, which are assigned by experts to every 30s window of signal. For diagnosis, they also rely on shorter prototypical micro-architecture events which exhibit variable durations and shapes, such as spindles, K-complexes or arousals. Annotating such events is traditionally performed by a trained sleep expert, making the process time consuming, tedious and subject to inter-scorer variability. To automate this procedure, various methods have been developed, yet these are event-specific and rely on the extraction of hand-crafted features. New method: We propose a novel deep learning architecure called Dreem One Shot Event Detector (DOSED). DOSED jointly predicts locations, durations and types of events in EEG time series. The proposed approach, applied here on sleep related micro-architecture events, is inspired by object detectors developed for computer vision such as YOLO and SSD. It relies on a convolutional neural network that builds a feature representation from raw EEG signals, as well as two modules performing localization and classification respectively. Results and comparison with other methods: The proposed approach is tested on 4 datasets and 3 types of events (spindles, K-complexes, arousals) and compared to the current state-of-the-art detection algorithms. Conclusions: Results demonstrate the versatility of this new approach and improved performance compared to the current state-of-the-art detection methods.	['Alexandre Gramfort', 'Stanislas Chambon', 'Pierrick J. Arnal', 'Emmanuel Mignot', 'Valentin Thorey']	2018-12-07	null	null	null	null	['sleep-quality-prediction', 'k-complex-detection', 'spindle-detection', 'sleep-apnea-detection', 'sleep-micro-event-detection', 'sleep-arousal-detection']	['medical', 'medical', 'medical', 'medical', 'medical', 'medical']	[ 1.26377076e-01 -3.39388669e-01 3.12543899e-01 -3.02703649e-01 -3.39933068e-01 -4.73264784e-01 3.99449021e-01 5.43445051e-01 -6.60300374e-01 8.67808163e-01 4.69663329e-02 -2.43333150e-02 -3.77173364e-01 -3.89523536e-01 -1.89003617e-01 -6.74327314e-01 -4.50164318e-01 2.50970513e-01 4.16273683e-01 -8.61304551e-02 3.70654792e-01 3.95264626e-01 -1.78302121e+00 2.19589427e-01 7.25065589e-01 1.27913392e+00 3.63167584e-01 5.43261349e-01 -5.57360798e-02 3.90033633e-01 -1.12299383e+00 6.28084913e-02 -2.19129369e-01 -5.89033902e-01 -2.85567373e-01 3.94989997e-02 -3.69164407e-01 3.51370990e-01 3.88758816e-02 1.01340079e+00 6.89927280e-01 4.52369973e-02 7.10789979e-01 -1.07605875e+00 -1.46656364e-01 2.70026058e-01 -3.90271276e-01 1.15500617e+00 4.77971107e-01 -1.84649423e-01 5.73476017e-01 -7.28708863e-01 1.33819595e-01 3.84317726e-01 8.11486900e-01 3.84456158e-01 -1.16312921e+00 -8.19088161e-01 -4.00667995e-01 8.81424665e-01 -1.50812638e+00 -4.67934310e-01 7.97863781e-01 -5.38555861e-01 1.25238836e+00 2.49678701e-01 1.05418003e+00 1.21384835e+00 9.34403956e-01 4.64728177e-01 1.19889748e+00 -3.29291105e-01 8.05706441e-01 1.26627102e-01 3.76025945e-01 3.13748360e-01 5.23949973e-02 -1.82851553e-01 -8.57077599e-01 2.97758207e-02 3.48417789e-01 2.62457132e-01 -4.05205041e-01 1.82240710e-01 -9.20925617e-01 5.69825947e-01 1.16872832e-01 7.48933494e-01 -8.14359963e-01 -1.91138744e-01 4.63601559e-01 1.85314134e-01 5.03623009e-01 3.75267118e-01 -3.58263463e-01 -5.07139564e-01 -1.52289236e+00 1.96579490e-02 6.80724502e-01 6.89726412e-01 5.37148237e-01 -1.60362825e-01 -3.69822472e-01 5.30828118e-01 9.50262621e-02 8.37061182e-03 1.05631781e+00 -7.23927096e-02 5.54077849e-02 1.08416128e+00 9.59512666e-02 -6.31942391e-01 -1.20647359e+00 -5.71365833e-01 -8.92478645e-01 1.13928981e-01 -4.60711308e-02 -1.15018934e-02 -8.82255554e-01 1.34821284e+00 -1.11111462e-01 4.90264118e-01 -2.15450183e-01 7.96336234e-01 8.56213987e-01 4.07612532e-01 2.09808238e-02 -5.29598594e-01 1.76768219e+00 -4.18768376e-01 -9.95849431e-01 -1.67112678e-01 -2.16576960e-02 -5.74380696e-01 8.76523852e-01 7.13044941e-01 -9.75486636e-01 -5.05841553e-01 -1.32795870e+00 3.62715095e-01 -7.31605768e-01 3.64885598e-01 2.08779812e-01 8.10236990e-01 -1.08075941e+00 4.79732782e-01 -1.32674968e+00 -6.47065938e-01 4.91698563e-01 7.51241684e-01 -2.17807204e-01 8.48704934e-01 -9.65491891e-01 1.13519847e+00 3.61945689e-01 1.84524924e-01 -9.99100626e-01 -4.56747115e-01 -4.18781251e-01 3.77287656e-01 3.28536853e-02 -4.30606991e-01 9.91999924e-01 -8.43094289e-01 -1.43107283e+00 9.15226877e-01 -3.34445804e-01 -7.59839714e-01 -5.20286076e-02 -1.47939414e-01 -8.72390330e-01 1.85832486e-01 -5.13936542e-02 -2.90856138e-02 9.82464373e-01 -5.00296712e-01 -5.73355854e-01 -5.88702381e-01 -7.94143528e-02 -1.46022186e-01 -3.64169359e-01 4.21123207e-01 -4.59571145e-02 -5.56471825e-01 -1.86870635e-01 -6.01029932e-01 1.15915470e-01 -6.05465353e-01 -3.13920438e-01 -5.93046367e-01 5.48153102e-01 -5.39690554e-01 1.61276889e+00 -2.08318186e+00 1.39134392e-01 2.99950987e-02 5.14087021e-01 2.21573934e-01 3.70099396e-01 5.63249648e-01 -3.20528835e-01 -3.82715404e-01 -3.18616807e-01 -4.58822131e-01 3.18358615e-02 -6.39561117e-02 -1.44503236e-01 6.90330744e-01 1.58444926e-01 4.93592948e-01 -7.32922971e-01 -2.64136881e-01 4.44493026e-01 4.79002386e-01 6.31177649e-02 4.02818829e-01 2.64491975e-01 4.19404954e-01 -1.32391891e-02 5.84292829e-01 3.76663566e-01 -1.47150025e-01 -3.03144217e-01 -2.51161098e-01 -4.93317038e-01 5.53399622e-01 -1.01375186e+00 1.80530095e+00 -3.81173700e-01 7.58535147e-01 -3.49795818e-01 -1.24051905e+00 9.18091953e-01 5.87623656e-01 2.89152831e-01 -5.42976379e-01 5.09019613e-01 1.62521288e-01 5.29219396e-02 -7.33654737e-01 1.37448730e-02 -1.62127480e-01 8.47797915e-02 3.05828512e-01 4.80307639e-01 1.60916746e-01 4.54844981e-01 -1.07326686e-01 1.60985982e+00 -1.60850048e-01 8.22636068e-01 -4.86648738e-01 6.39694333e-01 -4.50173080e-01 4.73888338e-01 5.01776814e-01 -1.93716705e-01 4.88069296e-01 5.30143738e-01 -6.43731534e-01 -5.81396580e-01 -1.09451628e+00 -3.38567644e-01 8.15108538e-01 9.50883850e-02 -6.92287982e-01 -9.70480323e-01 -5.46231747e-01 -4.84453291e-01 6.40792787e-01 -8.89688551e-01 -3.36857945e-01 -2.47039467e-01 -1.17987525e+00 3.95691335e-01 5.29353499e-01 2.62548715e-01 -1.44400632e+00 -1.59406817e+00 5.26555300e-01 1.26437083e-01 -9.62423861e-01 -5.79156615e-02 9.33138967e-01 -7.34723330e-01 -1.07348585e+00 -4.97228920e-01 -5.01485288e-01 4.32467759e-01 -2.18874007e-01 1.17826450e+00 -2.49506488e-01 -6.90028191e-01 1.25647709e-01 -6.15326881e-01 -8.09203506e-01 1.28135949e-01 1.54567212e-02 3.86681378e-01 3.68644983e-01 9.78882611e-01 -1.29210854e+00 -9.19787645e-01 1.12974852e-01 -8.38967264e-01 -3.59265924e-01 8.40318739e-01 4.90635514e-01 4.95975167e-01 1.79263175e-01 8.36392879e-01 -4.01821792e-01 8.55754435e-01 -5.97429514e-01 -5.72360694e-01 1.06145181e-01 -6.64394259e-01 -1.60209507e-01 9.55748200e-01 -2.85993606e-01 -5.20789623e-01 1.67770945e-02 -1.58578858e-01 -3.63008618e-01 -6.07406378e-01 2.26309121e-01 -6.16150871e-02 4.45345081e-02 7.09508479e-01 7.44656384e-01 -4.24270719e-01 -4.93313402e-01 -4.13636982e-01 9.94379580e-01 5.66963673e-01 1.05523638e-01 2.39983544e-01 3.27533007e-01 -1.08460933e-01 -1.01188529e+00 -5.68610132e-01 -7.73773789e-01 -6.87464535e-01 -1.38018966e-01 1.09159899e+00 -8.40789139e-01 -6.69207335e-01 4.26612437e-01 -1.04553628e+00 1.16219493e-02 -1.16142631e-01 5.97226262e-01 -4.67171550e-01 5.28902039e-02 -2.07979381e-01 -8.88566375e-01 -7.21584618e-01 -9.32132423e-01 1.11716306e+00 5.28818667e-01 -4.98810977e-01 -7.18038499e-01 4.02848542e-01 -3.13240916e-01 4.77527499e-01 1.59853339e-01 5.85292518e-01 -8.97682130e-01 7.74853304e-03 -2.06045941e-01 1.85438618e-01 1.33710191e-01 2.86226720e-01 -4.25518870e-01 -1.07612848e+00 -1.31486475e-01 5.13423324e-01 5.92425652e-02 6.06548727e-01 4.39675272e-01 1.09605646e+00 1.44580886e-01 -3.69046152e-01 6.32224858e-01 1.32510555e+00 6.18297219e-01 6.60458922e-01 5.28689444e-01 -1.44622132e-01 1.49108663e-01 3.67474705e-02 7.94077933e-01 3.28838557e-01 5.95712304e-01 4.31627035e-01 1.50729284e-01 1.63477913e-01 5.54188848e-01 4.09801841e-01 6.06997967e-01 -3.45910579e-01 -1.22044362e-01 -7.03476191e-01 7.20222116e-01 -1.62182999e+00 -8.56696367e-01 -3.72800715e-02 2.08008957e+00 7.20095456e-01 3.85540187e-01 2.82498598e-01 5.76649189e-01 4.95603710e-01 -1.21301971e-01 -5.84586322e-01 -4.88470376e-01 4.16386984e-02 1.07685125e+00 1.51757509e-01 -3.21639478e-01 -9.63813603e-01 4.54797685e-01 5.51856470e+00 6.23519182e-01 -1.24949396e+00 5.96123457e-01 -4.56796624e-02 -5.51073432e-01 5.49161732e-01 -4.98526365e-01 -6.44182086e-01 1.02647817e+00 1.49010479e+00 -2.35779330e-01 5.59333086e-01 5.11654377e-01 5.95222592e-01 -3.93860400e-01 -1.18233156e+00 1.60066843e+00 3.03849250e-01 -1.12402320e+00 -6.50327682e-01 -3.39182705e-01 2.57517815e-01 7.65610859e-02 -3.38387251e-01 3.23405057e-01 -5.13769627e-01 -9.72140610e-01 7.61285424e-01 7.45152891e-01 6.37470245e-01 -7.56610751e-01 1.07795513e+00 1.57647327e-01 -1.37285745e+00 -2.24959850e-01 -2.80451447e-01 -1.53642938e-01 6.36694655e-02 7.72099137e-01 -7.18082428e-01 3.75089079e-01 1.11103904e+00 7.67571926e-01 -8.61329675e-01 1.53256166e+00 -4.03934479e-01 8.20852935e-01 -3.16603005e-01 -1.86743617e-01 -3.80735402e-03 5.15115559e-02 5.60327590e-01 1.39816403e+00 6.77140594e-01 1.35125875e-01 -3.97921056e-01 1.12314320e+00 2.08818257e-01 -1.51738584e-01 -2.59105086e-01 4.94782254e-02 4.27173823e-01 1.69659257e+00 -1.24241078e+00 -2.35088378e-01 -3.13146025e-01 1.08259892e+00 8.74638334e-02 -9.84372199e-03 -7.88561940e-01 -7.63640881e-01 4.21752751e-01 -5.29312976e-02 2.01022759e-01 4.78811786e-02 -4.06726837e-01 -1.11541748e+00 1.64204255e-01 -5.44545352e-01 2.25188822e-01 -6.82525218e-01 -1.07088864e+00 1.06025422e+00 -1.04202464e-01 -1.31953490e+00 -2.43701994e-01 -4.93447751e-01 -1.09502959e+00 6.64639771e-01 -1.28021145e+00 -6.72296286e-01 -4.90567863e-01 1.00929070e+00 8.05029333e-01 -2.43561313e-01 9.64490950e-01 5.25117517e-01 -7.70320892e-01 3.71372610e-01 -2.87734661e-02 -2.87093937e-01 5.90330720e-01 -1.54119849e+00 9.37282145e-02 8.01350713e-01 3.16673160e-01 5.55130064e-01 7.41561890e-01 -2.32177854e-01 -1.19652367e+00 -8.73590946e-01 8.71967196e-01 -4.01383638e-01 6.55396104e-01 -5.70320427e-01 -6.51964426e-01 4.22594458e-01 3.56616616e-01 -1.45199955e-01 8.28788519e-01 -7.04742149e-02 4.56729889e-01 -4.57704633e-01 -1.00558007e+00 3.46260071e-02 6.87152147e-01 -4.03402001e-01 -1.22984076e+00 1.98193118e-01 -1.38869002e-01 -2.60346085e-01 -5.99605918e-01 8.29440132e-02 5.15151978e-01 -1.32956922e+00 4.75961715e-01 -1.13153964e-01 -1.36448555e-02 -4.02794003e-01 2.76436388e-01 -1.45863307e+00 -2.33127370e-01 -8.21750581e-01 -3.91292512e-01 8.43879521e-01 1.02268837e-01 -4.76722538e-01 5.08992016e-01 8.93710479e-02 -6.09845221e-01 -7.59567201e-01 -1.18169808e+00 -7.45758057e-01 -7.96259105e-01 -4.67556953e-01 5.77533424e-01 4.16504323e-01 3.11139911e-01 5.77981710e-01 1.63341686e-02 3.67012113e-01 3.58841836e-01 -1.68522634e-02 1.76630884e-01 -1.48739886e+00 -1.66062582e-02 -5.14052570e-01 -9.61773813e-01 -1.87873393e-01 -3.80329341e-02 -6.76599979e-01 6.77113831e-02 -1.41118145e+00 4.36066613e-02 1.10645518e-01 -9.02374923e-01 5.05269945e-01 3.67224105e-02 5.13527215e-01 -2.68605143e-01 -3.70375104e-02 -7.10372448e-01 4.07975674e-01 2.88366228e-01 2.05361992e-01 -5.20131171e-01 1.24007076e-01 -3.61834675e-01 9.33019340e-01 7.32264459e-01 -5.93914926e-01 -4.23373133e-01 7.56665021e-02 3.45646381e-01 -5.95555454e-02 4.58797216e-01 -1.74164140e+00 5.08056164e-01 5.37936628e-01 6.62590981e-01 -8.30359101e-01 3.44436616e-01 -7.36629546e-01 2.50378013e-01 3.38102192e-01 1.50079295e-01 4.11843717e-01 2.27913052e-01 4.90272552e-01 -2.18422577e-01 -2.15114728e-01 6.82067096e-01 -1.74457610e-01 -6.46365106e-01 8.83974251e-04 -7.45905697e-01 -2.11968094e-01 1.19365752e+00 -4.04165953e-01 -6.72383532e-02 8.55218619e-02 -1.07021880e+00 -1.91277131e-01 -6.41102269e-02 3.25558811e-01 7.75488794e-01 -9.05350327e-01 -3.92417580e-01 4.85785067e-01 8.94875005e-02 -3.89496088e-01 1.66629881e-01 1.41573560e+00 -2.56704301e-01 3.87943268e-01 -7.00321138e-01 -5.94674110e-01 -1.10312581e+00 4.62838203e-01 2.12580577e-01 -1.63495347e-01 -7.78035402e-01 8.93142641e-01 -7.95483403e-03 4.48949933e-01 2.46437669e-01 -8.02353442e-01 -8.38407815e-01 4.62885559e-01 9.06343758e-01 4.77770567e-01 7.74590731e-01 -2.82764614e-01 -8.26483548e-01 4.65687662e-01 1.58559173e-01 1.79952800e-01 1.67845452e+00 -8.41598064e-02 -3.37589949e-01 7.65932918e-01 6.84352994e-01 -2.07416192e-01 -8.54087710e-01 4.04339433e-01 2.13011637e-01 1.54178157e-01 1.47002846e-01 -9.93077099e-01 -8.48807335e-01 9.86555457e-01 1.23319733e+00 6.10241950e-01 1.67633307e+00 -1.15020253e-01 7.93030858e-01 2.25832760e-01 4.58064437e-01 -1.03014350e+00 -1.39613211e-01 7.77119398e-02 8.01512539e-01 -8.50075245e-01 -2.14856908e-01 2.85391718e-01 -2.01026484e-01 1.52270722e+00 3.56538147e-01 -4.10433978e-01 7.73664832e-01 3.15194458e-01 -3.88783097e-01 -4.65566218e-01 -7.48272955e-01 -3.50401163e-01 3.51351410e-01 2.95457363e-01 3.66339684e-01 1.09579965e-01 -7.74985552e-01 1.33803880e+00 -4.57094580e-01 5.05733073e-01 5.43569982e-01 7.75214493e-01 -4.48999256e-01 -8.03012311e-01 -4.14239854e-01 7.39188373e-01 -7.68027842e-01 -1.72514573e-01 -1.54665992e-01 6.24630213e-01 7.16371357e-01 1.16614807e+00 2.31379300e-01 -5.89637280e-01 3.81344795e-01 3.42056304e-01 5.15773833e-01 -7.31676519e-01 -8.94285202e-01 -4.72306460e-02 -2.20293477e-01 -6.19350016e-01 -6.32777750e-01 -5.67678630e-01 -1.21351850e+00 4.11837250e-01 -2.92784601e-01 2.68504739e-01 7.61604607e-01 1.27419889e+00 5.05991876e-01 9.02619481e-01 4.68535990e-01 -1.05349040e+00 -8.93166736e-02 -1.37658226e+00 -8.53791714e-01 2.06784651e-01 6.40398026e-01 -1.07852113e+00 -4.81402695e-01 3.10488194e-01]	[13.454808235168457, 3.4717748165130615]

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.