text
string
source
string
Anomaly detection in videos is a significant yet challenging problem. Previous approaches based on deep neural networks employ either reconstruction-based or prediction-based approaches. Nevertheless, existing reconstruction-based methods 1) rely on old-fashioned convolutional autoencoders and are poor at modeling temp...
http://arxiv.org/abs/2301.12048v1
Human speakers can generate descriptions of perceptual concepts, abstracted from the instance-level. Moreover, such descriptions can be used by other speakers to learn provisional representations of those concepts. Learning and using abstract perceptual concepts is under-investigated in the language-and-vision field. T...
http://arxiv.org/abs/2303.04053v3
Optical wireless communication (OWC) offers several complementary advantages to radio-frequency wireless networks such as its massive available spectrum; hence, it is widely anticipated that OWC will assume a pivotal role in the forthcoming sixth generation wireless communication networks. Although significant progress...
http://arxiv.org/abs/2305.09821v2
Physics-informed neural networks (PINNs) have been demonstrated to be efficient in solving partial differential equations (PDEs) from a variety of experimental perspectives. Some recent studies have also proposed PINN algorithms for PDEs on surfaces, including spheres. However, theoretical understanding of the numerica...
http://arxiv.org/abs/2308.09605v2
The possibility to engineer artificial Kitaev chains in arrays of quantum dots coupled via narrow superconducting regions has emerged as an attractive way to overcome the disorder issues that complicate the realization and detection of topological superconducting phases in other platforms. Although a true topological p...
http://arxiv.org/abs/2306.16289v1
In teleparallel geometries the coframe and corresponding spin-connection are the principal geometric objects and consequently the appropriate definition of a symmetry is that of an affine symmetry. The set of invariant coframes and their corresponding spin connections that respect the full six dimensional Lie algebra o...
http://arxiv.org/abs/2310.14378v1
The classification of galaxy morphology is a hot issue in astronomical research. Although significant progress has been made in the last decade in classifying galaxy morphology using deep learning technology, there are still some deficiencies in spatial feature representation and classification accuracy. In this study,...
http://arxiv.org/abs/2304.10081v1
Energy shortfall and electricity load shedding are the main problems for developing countries. The main causes are lack of management in the energy sector and the use of non-renewable energy sources. The improved energy management and use of renewable sources can be significant to resolve energy crisis. It is necessary...
http://arxiv.org/abs/2307.09080v1
One of the primary reasons behind the success of neural networks has been the emergence of an array of new, highly-successful optimizers, perhaps most importantly the Adam optimizer. It is widely used for training neural networks, yet notoriously hard to interpret. Lacking a clear physical intuition, Adam is difficult ...
http://arxiv.org/abs/2305.16901v3
The ability to actively ground task instructions from an egocentric view is crucial for AI agents to accomplish tasks or assist humans virtually. One important step towards this goal is to localize and track key active objects that undergo major state change as a consequence of human actions/interactions to the environ...
http://arxiv.org/abs/2310.15066v1
Quench dynamics of fermionic superfluids are an active topic both experimentally and theoretically. Using the BCS theory, such non-equilibrium problems can be reduced to nearly independent spin dynamics, only with a time-dependent mean-field pairing term. This results in persisting oscillations of the pairing strength ...
http://arxiv.org/abs/2303.02422v2
Background - Startup companies are becoming important suppliers of innovative and software intensive products. The failure rate among startups is high due to lack of resources, immaturity, multiple influences and dynamic technologies. However, software product engineering is the core activity in startups, therefore ina...
http://arxiv.org/abs/2308.07628v1
Early results from the JWST-MIRI guaranteed time programs on protostars (JOYS) and disks (MINDS) are presented. Thanks to the increased sensitivity, spectral and spatial resolution of the MIRI spectrometer, the chemical inventory of the planet-forming zones in disks can be investigated with unprecedented detail across ...
http://arxiv.org/abs/2307.11817v1
Point-to-multi-point (PtMP) optical networks become the main solutions for network-edge applications such as passive optical networks and radio access networks. Entropy-loading digital subcarrier multiplexing (DSCM) is the core technology to achieve low latency and approach high capacity for flexible PtMP optical netwo...
http://arxiv.org/abs/2310.12709v1
Kilonovae are generally believed to originate from the ejecta of binary neutron stars (NSs) or black hole and NS mergers. Free neutrons might be retained in the outermost layer of the ejecta to produce a precursor via $\beta$-decay. During the propagation of kilonovae to observers, a small percentage of them might be g...
http://arxiv.org/abs/2301.07323v1
Despite their immense success in numerous fields, machine and deep learning systems have not yet been able to firmly establish themselves in mission-critical applications in healthcare. One of the main reasons lies in the fact that when models are presented with previously unseen, Out-of-Distribution samples, their per...
http://arxiv.org/abs/2303.11338v3
Recent research has shown that hardware fuzzers can effectively detect security vulnerabilities in modern processors. However, existing hardware fuzzers do not fuzz well the hard-to-reach design spaces. Consequently, these fuzzers cannot effectively fuzz security-critical control- and data-flow logic in the processors,...
http://arxiv.org/abs/2304.02485v3
Machine learning is traditionally studied at the model level: researchers measure and improve the accuracy, robustness, bias, efficiency, and other dimensions of specific models. In practice, the societal impact of machine learning is determined by the surrounding context of machine learning deployments. To capture thi...
http://arxiv.org/abs/2307.05862v2
In this work, we propose a new transformer-based regularization to better localize objects for Weakly supervised semantic segmentation (WSSS). In image-level WSSS, Class Activation Map (CAM) is adopted to generate object localization as pseudo segmentation labels. To address the partial activation issue of the CAMs, co...
http://arxiv.org/abs/2308.04321v2
In this short note we prove an analogue of Auslander correspondence for exact dg categories whose $H^0$-category is $0$-Auslander in the sense of Gorsky--Nakaoka--Palu.
http://arxiv.org/abs/2306.15958v2
We present here uGMRT band 4 (~650MHz) polarization images of 8 BL~Lac objects belonging to the Palomar-Green (PG) `blazar' sample. A large fraction of the sources (~63%) reveal core-halo radio structures with most of the polarization detected in the inner core-jet regions. PG1101+385 and PG2254+075 exhibit a `spine-sh...
http://arxiv.org/abs/2310.08989v1
A novel hack involving Large Language Models (LLMs) has emerged, exploiting adversarial suffixes to deceive models into generating perilous responses. Such jailbreaks can trick LLMs into providing intricate instructions to a malicious user for creating explosives, orchestrating a bank heist, or facilitating the creatio...
http://arxiv.org/abs/2308.14132v3
This paper studies the endemic behavior of a multi-competitive networked susceptible-infected-susceptible (SIS) model. Specifically, the paper deals with three competing virus systems (i.e., tri-virus systems). First, we show that a tri-virus system, unlike a bi-virus system, is not a monotone dynamical system. Using t...
http://arxiv.org/abs/2303.16457v1
In this paper, we investigate a semiparametric regression model under the context of treatment effects via a localized neural network (LNN) approach. Due to a vast number of parameters involved, we reduce the number of effective parameters by (i) exploring the use of identification restrictions; and (ii) adopting a var...
http://arxiv.org/abs/2306.05593v2
We propose a new optimization-based approach for feature selection in tree ensembles, an important problem in statistics and machine learning. Popular tree ensemble toolkits e.g., Gradient Boosted Trees and Random Forests support feature selection post-training based on feature importance scores, while very popular, th...
http://arxiv.org/abs/2310.18542v2
Modeling text-based time-series to make prediction about a future event or outcome is an important task with a wide range of applications. The standard approach is to train and test the model using the same input window, but this approach neglects the data collected in longer input windows between the prediction time a...
http://arxiv.org/abs/2301.10887v1
We demonstrate a fast adjoint-based method to optimize tokamak and stellarator equilibria against a pressure-driven instability known as the infinite-$n$ ideal ballooning mode. We present three finite-$\beta$ (the ratio of thermal to magnetic pressure) equilibria: one tokamak equilibrium and two stellarator equilibria ...
http://arxiv.org/abs/2302.07673v1
We develop a framework to assess the risk of cascading failures when a team of agents aims to rendezvous in time in the presence of exogenous noise and communication time-delay. The notion of value-at-risk (VaR) measure is used to evaluate the risk of cascading failures (i.e., waves of large fluctuations) when agents h...
http://arxiv.org/abs/2303.08669v1
Plasma technologies are a promising way of addressing a number of challenges in combustion, ranging from stabilization and flame-holding for hypersonic vehicles, to enabling low-emissions propulsion and power. Whereas most works in the literature have centered on the forward problem: the impact of the plasma on differe...
http://arxiv.org/abs/2306.04855v1
SDSS J1004+4112 is a well studied gravitational lens with a recently measured time delay between its first and fourth arriving quasar images. Using this new constraint, we present updated free-form lens reconstructions using the lens inversion method {\tt GRALE}, which only uses multiple image and time delay data as in...
http://arxiv.org/abs/2310.19884v1
Modern sciences and astrophysics in particular study objects and phenomena not visible in physical terms, that is they cannot be investigated with the eyes or analogous optical systems. Nevertheless, they make intensive use of visual representations, showing data in a figurative way, using lights, colors, and shapes fa...
http://arxiv.org/abs/2302.08226v1
In this paper, we propose a numerical method to uniformly handle the random genetic drift model for pure drift with or without natural selection and mutation. For pure drift and natural selection case, the Dirac $\delta$ singularity will develop at two boundary ends and the mass lumped at the two ends stands for the ...
http://arxiv.org/abs/2310.17117v1
This paper presents a novel modular robot system that can self-reconfigure to achieve omnidirectional movements for collaborative object transportation. Each robotic module is equipped with a steerable omni-wheel for navigation and is shaped as a regular icositetragon with a permanent magnet installed on each corner fo...
http://arxiv.org/abs/2308.03328v1
Planning for sequential robotics tasks often requires integrated symbolic and geometric reasoning. TAMP algorithms typically solve these problems by performing a tree search over high-level task sequences while checking for kinematic and dynamic feasibility. This can be inefficient because, typically, candidate task pl...
http://arxiv.org/abs/2310.01775v4
We propose a protocol to excite the Goldstone modes of a supersolid dipolar Bose-Einstein condensed gas confined in a ring geometry. By abruptly removing an applied periodic modulation proportional to cos($\varphi$), where $\varphi$ is the azimuthal angle, we explore the resulting oscillations of the gas, by solving th...
http://arxiv.org/abs/2308.05981v2
We study gravitational lensing of neutrinos in a Kaluza-Klein black hole spacetime and compare the oscillation probabilities of neutrinos with the case of lensing by black holes in General Relativity. We show that measuring neutrino oscillations in curved spacetimes may allow us to distinguish the two kinds of black ho...
http://arxiv.org/abs/2302.01564v2
Multiple intent detection and slot filling are two fundamental and crucial tasks in spoken language understanding. Motivated by the fact that the two tasks are closely related, joint models that can detect intents and extract slots simultaneously are preferred to individual models that perform each task independently. ...
http://arxiv.org/abs/2308.14654v1
Stochastic games are a well established model for multi-agent sequential decision making under uncertainty. In practical applications, though, agents often have only partial observability of their environment. Furthermore, agents increasingly perceive their environment using data-driven approaches such as neural networ...
http://arxiv.org/abs/2310.11566v3
Variational quantum algorithms, which consist of optimal parameterized quantum circuits, are promising for demonstrating quantum advantages in the noisy intermediate-scale quantum (NISQ) era. Apart from classical computational resources, different kinds of quantum resources have their contributions to the process of co...
http://arxiv.org/abs/2301.07445v3
Emission line galaxies (ELGs) are now the preeminent tracers of large-scale structure at z>0.8 due to their high density and strong emission lines, which enable accurate redshift measurements. However, relatively little is known about ELG evolution and the ELG-halo connection, exposing us to potential modeling systemat...
http://arxiv.org/abs/2310.09329v1
In this work and its sister paper [5] we give a new proof of the famous Linnik theorem bounding the least prime in an arithmetic progression. Using sieve machinery in both papers, we are able to dipense with the log-free zero density bounds and the repulsion property of exceptional zeros, two deep innovations begun by ...
http://arxiv.org/abs/2303.06122v1
We present a statistical analysis of ion distributions in magnetic reconnection jets using data from the Magnetospheric Multiscale spacecraft. Compared with the quiet plasma in which the jet propagates, we often find anisotropic and non-Maxwellian ion distributions in the plasma jets. We observe magnetic field fluctuat...
http://arxiv.org/abs/2301.10139v3
Nanoparticles, such as viruses, can enter cells via endocytosis. During endocytosis, the cell surface wraps around the nanoparticle to effectively eat it. Prior focus has been on how nanoparticle size and shape impacts endocytosis. However, inspired by the noted presence of extracellular vimentin affecting viral and ba...
http://arxiv.org/abs/2301.08776v2
The 2023 Multilingual Speech Universal Performance Benchmark (ML-SUPERB) Challenge expands upon the acclaimed SUPERB framework, emphasizing self-supervised models in multilingual speech recognition and language identification. The challenge comprises a research track focused on applying ML-SUPERB to specific multilingu...
http://arxiv.org/abs/2310.05513v1
While several recent works have identified societal-scale and extinction-level risks to humanity arising from artificial intelligence, few have attempted an {\em exhaustive taxonomy} of such risks. Many exhaustive taxonomies are possible, and some are useful -- particularly if they reveal new risks or practical approac...
http://arxiv.org/abs/2306.06924v2
We propose learning a depth covariance function with applications to geometric vision tasks. Given RGB images as input, the covariance function can be flexibly used to define priors over depth functions, predictive distributions given observations, and methods for active point selection. We leverage these techniques fo...
http://arxiv.org/abs/2303.12157v2
Under mild assumptions, we investigate the geometry of the loss landscape for two-layer neural networks in the vicinity of global minima. Utilizing novel techniques, we demonstrate: (i) how global minima with zero generalization error become geometrically separated from other global minima as the sample size grows; and...
http://arxiv.org/abs/2309.00508v3
For more than two decades it has been known that any compact Stein surface (of real dimension four) admits a compatible Lefschetz fibration over a two-disk. More recently, Giroux and Pardon have generalized this result by giving a complex geometric proof for the existence of compatible Lefschetz fibrations on Stein dom...
http://arxiv.org/abs/2310.15700v3
We endow the classical stochastic fluid process with a duration-dependent Markovian arrival process (DMArP). We show that this provides a flexible model for the revenue of a solar energy generator. In particular, it allows for heavy-tailed interarrival times and for seasonality embedded into the state-space. It general...
http://arxiv.org/abs/2304.06185v1
Evaluating the factual consistency of automatically generated summaries is essential for the progress and adoption of reliable summarization systems. Despite recent advances, existing factuality evaluation models are not robust, being especially prone to entity and relation errors in new domains. We propose FactKB, a s...
http://arxiv.org/abs/2305.08281v2
Compact obscured nuclei (CONs) are relatively common in the centers of local (U)LIRGs, yet their nature remains unknown. Both AGN activity and extreme nuclear starbursts have been suggested as plausible nuclear power sources. The prevalence of outflows in these systems suggest that CONs represent a key phase in the nuc...
http://arxiv.org/abs/2309.01616v1
Electron optics deals with condensed matter platforms for manipulating and guiding electron beams with high efficiency and robustness. Common devices rely on the spatial confinement of the electrons into one-dimensional channels. Recently, there is growing interest in electron optics applications in two dimensions, whi...
http://arxiv.org/abs/2307.07913v2
A brain-computer interface (BCI) is a technology that enables direct communication between the brain and an external device or computer system. It allows individuals to interact with the device using only their thoughts, and holds immense potential for a wide range of applications in medicine, rehabilitation, and human...
http://arxiv.org/abs/2305.11908v1
We study the exact linearization of configuration flat Lagrangian control systems with p degrees of freedom and p-1 inputs by quasi-static feedback of classical states. First, we present a detailed analysis of the structure of the parameterization of the system variables by the flat output. Based on that, we systematic...
http://arxiv.org/abs/2310.13371v2
The reconstruction of trajectories of charged particles is a key computational challenge for current and future collider experiments. Considering the rapid progress in quantum computing, it is crucial to explore its potential for this and other problems in high-energy physics. The problem can be formulated as a quadrat...
http://arxiv.org/abs/2303.13249v1
Distributed machine learning paradigms, such as federated learning, have been recently adopted in many privacy-critical applications for speech analysis. However, such frameworks are vulnerable to privacy leakage attacks from shared gradients. Despite extensive efforts in the image domain, the exploration of speech pri...
http://arxiv.org/abs/2302.10441v1
Vertex splitting replaces a vertex by two copies and partitions its incident edges amongst the copies. This problem has been studied as a graph editing operation to achieve desired properties with as few splits as possible, most often planarity, for which the problem is NP-hard. Here we study how to minimize the number...
http://arxiv.org/abs/2301.09440v1
Understanding the complex evolution of protoplanetary disks (PPDs) and their dispersal via energetic stellar radiation are prominent challenges in astrophysics. It has recently been established that specifically the X-ray luminosity from the central protostar can significantly heat the surface of the disk, causing powe...
http://arxiv.org/abs/2310.01985v1
Integrating turbulence into stellarator optimization is shown by targeting the onset for the ion-temperature-gradient mode, highlighting effects of parallel connection length, local magnetic shear, and flux surface expansion. The result is a compact quasihelically symmetric stellarator configuration, admitting a set of...
http://arxiv.org/abs/2301.06773v2
Brain activities are featured by spatially distributed neural clusters of coherent firings and a spontaneous switching of the clusters between the synchrony and asynchrony states. Evidences from {\it in vivo} experiments suggest that astrocytes, a type of glial cell regarded previously as providing only structural and ...
http://arxiv.org/abs/2302.00652v1
Domain-adaptive pre-training (or DA-training for short), also known as post-training, aims to train a pre-trained general-purpose language model (LM) using an unlabeled corpus of a particular domain to adapt the LM so that end-tasks in the domain can give improved performances. However, existing DA-training methods are...
http://arxiv.org/abs/2301.08986v1
Sequence-to-sequence (seq2seq) models have been widely used for natural language processing, computer vision, and other deep learning tasks. We find that seq2seq models trained with early-stopping suffer from issues at the token level. In particular, while some tokens in the vocabulary demonstrate overfitting, others u...
http://arxiv.org/abs/2305.04493v2
Real-world data is complex and often consists of objects that can be decomposed into multiple entities (e.g. images into pixels, graphs into interconnected nodes). Randomized smoothing is a powerful framework for making models provably robust against small changes to their inputs - by guaranteeing robustness of the maj...
http://arxiv.org/abs/2310.16221v4
Despite their impressive offline results, deep learning models for symbolic music generation are not widely used in live performances due to a deficit of musically meaningful control parameters and a lack of structured musical form in their outputs. To address these issues we introduce LooperGP, a method for steering a...
http://arxiv.org/abs/2303.01665v1
Given an algebra in a monoidal 2-category, one can construct a 2-category of right modules. Given a braided algebra in a braided monoidal 2-category, it is possible to refine the notion of right module to that of a local module. Under mild assumptions, we prove that the 2-category of local modules admits a braided mono...
http://arxiv.org/abs/2307.02843v1
Current fake audio detection algorithms have achieved promising performances on most datasets. However, their performance may be significantly degraded when dealing with audio of a different dataset. The orthogonal weight modification to overcome catastrophic forgetting does not consider the similarity of genuine audio...
http://arxiv.org/abs/2308.03300v1
Most Neural Radiance Fields (NeRFs) exhibit limited generalization capabilities, which restrict their applicability in representing multiple scenes using a single model. To address this problem, existing generalizable NeRF methods simply condition the model on image features. These methods still struggle to learn preci...
http://arxiv.org/abs/2304.04962v2
In this note, we prove: \medskip \noindent {\bf Theorem A:} \emph{ There is a fixed constant $C$ such that for any positive integer $n$ and prime $p$, every finite subgroup $G$ of order coprime to $p$ of ${\rm GL}(n,\mathbb{C})$ has an Abelian normal subgroup $A$ with $$[G:A] \leq (Cp)^{n-1}.$$}
http://arxiv.org/abs/2301.10111v1
We investigate the role of the magnetohydrodynamic (MHD) turbulence measured by Voyager in the very local interstellar medium (VLISM) in modeling the Interstellar Boundary Explorer (IBEX) ribbon. We demonstrate that the mirroring by compressible modes of MHD turbulence dominates over that by the mean magnetic field. Ba...
http://arxiv.org/abs/2310.06032v1
We propose the concept of pair-wise coherence to study the relation between the l1 norm of coherence and the quantum interference in Young's multi-photon multi-path experiments, where the input photons may be entangled each other. We find that only the local coherence of each single photon can make quantum interference...
http://arxiv.org/abs/2303.09239v1
Intuitively, the totality of physical reality -- the Cosmos -- has a beginning only if (i) all parts of the Cosmos agree on the direction of time (the Direction Condition) and (ii) there is a boundary to the past of all non-initial spacetime points such that there are no spacetime points to the past of the boundary (th...
http://arxiv.org/abs/2308.13113v1
The mathematical objects employed in physical theories do not always behave well. Einstein's theory of space and time allows for spacetime singularities and Van Hove singularities arise in condensed matter physics, while intensity, phase and polarization singularities pervade wave physics. Within dissipative systems go...
http://arxiv.org/abs/2307.03585v1
We consider a chance-constrained multi-robot motion planning problem in the presence of Gaussian motion and sensor noise. Our proposed algorithm, CC-K-CBS, leverages the scalability of kinodynamic conflict-based search (K-CBS) in conjunction with the efficiency of the Gaussian belief trees used in the Belief-A framewor...
http://arxiv.org/abs/2303.11476v4
In economic and financial applications, there is often the need for analysing multivariate time series, comprising of time series for a range of quantities. In some applications such complex systems can be associated with some underlying network describing pairwise relationships among the quantities. Accounting for the...
http://arxiv.org/abs/2305.16097v3
We present fundamental scaling relationships between properties of the optical/UV light curves of tidal disruption events (TDEs) and the mass of the black hole that disrupted the star. We have uncovered these relations from the late-time emission of TDEs. Using a sample of 63 optically-selected TDEs, the latest catalog...
http://arxiv.org/abs/2308.08255v2
We establish a sharp lower-bound for the number of conjugacy classes $k(A_n)$ in the alternating group $A_n$, for $n \geq 3$. Namely, we show that $k\left(A_n\right) \geq \frac{k\left(A_7\right)}{\log_2\left|A_7\right|} \cdot \log_2\left|A_n\right|$ with equality if, and only if, $n = 7$. The observations leading towar...
http://arxiv.org/abs/2310.12047v1
The Debye-Falkenhagen differential equation is commonly used as a mean-field macroscopic model for describing electrochemical ionic drift and diffusion in dilute binary electrolytes when subjected to a suddenly applied potential smaller than the thermal voltage. However, the ionic transport in most electrochemical syst...
http://arxiv.org/abs/2308.00609v1
Intermediate stages between pores and sunspots are a rare phenomenon and can manifest with the formation of transient photospheric penumbral-like filaments. Although the magnetic field changes rapidly during the evolution of such filaments, they have not been shown to be connected to magnetic reconnection events yet. W...
http://arxiv.org/abs/2303.03072v1
DNA is a promising storage medium, but its stability and occurrence of Indel errors pose a significant challenge. The relative occurrence of Guanine(G) and Cytosine(C) in DNA is crucial for its longevity, and reverse complementary base pairs should be avoided to prevent the formation of a secondary structure in DNA str...
http://arxiv.org/abs/2302.14627v4
Given a compact doubling metric measure space $X$ that supports a $2$-Poincar\'e inequality, we construct a Dirichlet form on $N^{1,2}(X)$ that is comparable to the upper gradient energy form on $N^{1,2}(X)$. Our approach is based on the approximation of $X$ by a family of graphs that is doubling and supports a $2$-Poi...
http://arxiv.org/abs/2310.14436v1
We propose a novel device concept using spin-orbit-torques to realize a magnetic field sensor, where we eliminate the sensor offset using a differential measurement concept. We derive a simple analytical formulation for the sensor signal and demonstrate its validity with numerical investigations using macrospin simulat...
http://arxiv.org/abs/2303.13261v1
X-ray time-resolved tomography is one of the most popular X-ray techniques to probe dynamics in three dimensions (3D). Recent developments in time-resolved tomography opened the possibility of recording kilohertz-rate 3D movies. However, tomography requires rotating the sample with respect to the X-ray beam, which prev...
http://arxiv.org/abs/2305.11920v1
Finding multiple temporal relationships among locations can benefit a bunch of urban applications, such as dynamic offline advertising and smart public transport planning. While some efforts have been made on finding static relationships among locations, little attention is focused on studying time-aware location relat...
http://arxiv.org/abs/2306.08921v1
We present a single attention head in GPT-2 Small that has one main role across the entire training distribution. If components in earlier layers predict a certain token, and this token appears earlier in the context, the head suppresses it: we call this copy suppression. Attention Head 10.7 (L10H7) suppresses naive co...
http://arxiv.org/abs/2310.04625v1
Some young supernova remnants exhibit thin filaments of X-ray synchrotron radiation coinciding with the forward shock due to accelerated electrons interacting with the local magnetic field. The two main models accounting for the radial brightness evolution of these filaments differ in their prediction of the narrowing ...
http://arxiv.org/abs/2302.04352v1
While large-scale image-text pretrained models such as CLIP have been used for multiple video-level tasks on trimmed videos, their use for temporal localization in untrimmed videos is still a relatively unexplored task. We design a new approach for this called UnLoc, which uses pretrained image and text towers, and fee...
http://arxiv.org/abs/2308.11062v1
Out-of-distribution (OOD) detection aims to identify test examples that do not belong to the training distribution and are thus unlikely to be predicted reliably. Despite a plethora of existing works, most of them focused only on the scenario where OOD examples come from semantic shift (e.g., unseen categories), ignori...
http://arxiv.org/abs/2304.06813v2
Processes involving the manufacture of fine/bulk chemicals, pharmaceuticals, biofuels, and waste treatment require plug flow characteristics to minimise their energy consumption and costs, and maximise product quality. One such versatile flow chemistry platform is the coiled tube reactor subjected to oscillatory motion...
http://arxiv.org/abs/2305.16929v1
We present Flow-Guided Density Ratio Learning (FDRL), a simple and scalable approach to generative modeling which builds on the stale (time-independent) approximation of the gradient flow of entropy-regularized f-divergences introduced in recent work. Specifically, the intractable time-dependent density ratio is approx...
http://arxiv.org/abs/2303.03714v3
The visual system evolved to process natural scenes, yet most of our understanding of the topology and function of visual cortex derives from studies using artificial stimuli. To gain deeper insights into visual processing of natural scenes, we utilized widefield calcium-imaging of primate V4 in response to many natura...
http://arxiv.org/abs/2307.00932v2
We present a method for pretraining a recurrent mixture density network (RMDN). We also propose a slight modification to the architecture of the RMDN-GARCH proposed by Nikolaev et al. [2012]. The pretraining method helps the RMDN avoid bad local minima during training and improves its robustness to the persistent NaN p...
http://arxiv.org/abs/2302.14141v1
Human mobility patterns have shown significant applications in policy-decision scenarios and economic behavior researches. The human mobility simulation task aims to generate human mobility trajectories given a small set of trajectory data, which have aroused much concern due to the scarcity and sparsity of human mobil...
http://arxiv.org/abs/2306.09381v3
We compute explicitly the Bergman kernels of all two dimensional monomial polyhedra, a class of domains including the Hartogs triangle and some of its generalizations. The kernel is computed from the representation of such domains as quotient of the product of a disc and punctured disc.
http://arxiv.org/abs/2303.14268v1
The Erd\"os Renyi graph is a popular choice to model network data as it is parsimoniously parametrized, straightforward to interprete and easy to estimate. However, it has limited suitability in practice, since it often fails to capture crucial characteristics of real-world networks. To check the adequacy of this model...
http://arxiv.org/abs/2305.10380v2
3D multi-object tracking (MOT) is essential for an autonomous mobile agent to safely navigate a scene. In order to maximize the perception capabilities of the autonomous agent, we aim to develop a 3D MOT framework that fuses camera and LiDAR sensor information. Building on our prior LiDAR-only work, ShaSTA, which model...
http://arxiv.org/abs/2310.02532v1
Recent studies have shown that 6.7 GHz methanol maser flares can be a powerful tool for verifying the mechanisms of maser production and even the specific signatures of accretion rate changes in the early stages of high-mass star formation. We characterize the spatial structure and evolution of methanol and water maser...
http://arxiv.org/abs/2302.02723v1
Generating proper embedding of sentences through an unsupervised way is beneficial to semantic matching and retrieval problems in real-world scenarios. This paper presents Representation ALchemy (RepAL), an extremely simple post-processing method that enhances sentence representations. The basic idea in RepAL is to de-...
http://arxiv.org/abs/2305.07824v1
In this paper, we investigate in detail the structures of the variational characterization $A_{N,t}$ of the spherical $t$-design, its gradient $\nabla A_{N,t}$, and its Hessian $\mathcal{H}(A_{N,t})$ in terms of fast spherical harmonic transforms. Moreover, we propose solving the minimization problem of $A_{N,t}$ using...
http://arxiv.org/abs/2303.05365v1
Generative models are now capable of producing natural language text that is, in some cases, comparable in quality to the text produced by people. In the computing education context, these models are being used to generate code, code explanations, and programming exercises. The rapid adoption of these models has prompt...
http://arxiv.org/abs/2308.04309v1
We are concerned with nonlinear stability and existence of two-dimensional current-vortex sheets in ideal compressible magnetohydrodynamics. This is a nonlinear hyperbolic initial-boundary value problem with characteristic free boundary. It is well-known that current-vortex sheets may be at most weakly (neutrally) stab...
http://arxiv.org/abs/2305.02784v1