prompt
stringlengths
41
511
target
stringlengths
1
25
keyword
stringclasses
697 values
full_sentence
stringlengths
48
1.25k
existing vision-language navigation vln approaches are mostly designed for ground robots and struggle to generalize to aerial tasks that require full 3d
spatial
visual navigation
existing vision-language navigation vln approaches are mostly designed for ground robots and struggle to generalize to aerial tasks that require full 3d spatial reasoning
this formulation leads to a point to set type of optimization problem which relaxes the requirement on
controllability
linear control
this formulation leads to a point to set type of optimization problem which relaxes the requirement on controllability of the system compared to the classic lap framework
debiased machine learning typically requires estimation of the riesz representer and the
regression
machine learning
debiased machine learning typically requires estimation of the riesz representer and the regression function
conventional control strategies typically rely on fixed setpoints and neglect frequency deviations which can compromise system
stability
control strategy
conventional control strategies typically rely on fixed setpoints and neglect frequency deviations which can compromise system stability under rapid renewable variations
while ai-assisted participants completed several
tasks
ai assistance
while ai-assisted participants completed several tasks faster and more accurately no significant pre-post differences emerged in standardized measures of problem solving or verbal comprehension
large language models llms have demonstrated exceptional
capabilities
language models
large language models llms have demonstrated exceptional capabilities across multiple domains by leveraging massive pre-training and curated fine-tuning data
we introduce human ai collaborative uncertainty quantification a framework that formalizes how an ai model can refine a human expert s proposed prediction set with two goals avoiding counterfactual harm ensuring the ai does not degrade correct human judgments and complementarity enabling recovery of correct outcomes th...
human
debiased machine learning
we introduce human ai collaborative uncertainty quantification a framework that formalizes how an ai model can refine a human expert s proposed prediction set with two goals avoiding counterfactual harm ensuring the ai does not degrade correct human judgments and complementarity enabling recovery of correct outcomes th...
moreover we show that the intercalation of few-nm-thick flakes also leads to the emergence of a
ferrimagnetic
magnetic ground
moreover we show that the intercalation of few-nm-thick flakes also leads to the emergence of a ferrimagnetic response
we address this gap with a token-aware causal
representation
representation learning
we address this gap with a token-aware causal representation learning crl framework grounded in a sequential language-token scm
however due to the winner s curse-an issue where the policy optimization procedure exploits prediction errors rather than finding actual improvements-predicted performance improvements are often not substantiated by downstream
policy
policy evaluation
however due to the winner s curse-an issue where the policy optimization procedure exploits prediction errors rather than finding actual improvements-predicted performance improvements are often not substantiated by downstream policy optimization
to mitigate the high-dimensional challenges of humanoid
control
optimal control
to mitigate the high-dimensional challenges of humanoid control thor introduces a reinforcement learning architecture that decouples the upper body waist and lower body
we also assess the diagnostic power of several magnitude gaps between top-ranked galaxies as proxies for
dynamical
quiescent galaxies
we also assess the diagnostic power of several magnitude gaps between top-ranked galaxies as proxies for dynamical age
approximating human preferences using a multi-judge
learned
preference data
approximating human preferences using a multi-judge learned system
however significant limitations remain mllms generalize poorly across digital-physical spaces and embodiments vision-language-action models vlas produce low-level actions yet lack robust high-level embodied reasoning and most embodied large
language
vision-language-action vla
however significant limitations remain mllms generalize poorly across digital-physical spaces and embodiments vision-language-action models vlas produce low-level actions yet lack robust high-level embodied reasoning and most embodied large language models ellms are constrained to digital-space with poor generalization...
recent breakthroughs in artificial intelligence ai are reshaping the way we construct computational counterparts of the brain giving rise to a new class of
surrogate
surrogate brain
recent breakthroughs in artificial intelligence ai are reshaping the way we construct computational counterparts of the brain giving rise to a new class of surrogate brains
multiplexes have emerged as a key instrument for modeling large-scale complex
systems
multiplex networks
multiplexes have emerged as a key instrument for modeling large-scale complex systems due to the widespread coexistence of diverse interactions in social industrial and biological domains
2023 introduced dynamical similarity analysis dsa a method to measure the similarity of two systems based on their
recurrent
dynamical systems
2023 introduced dynamical similarity analysis dsa a method to measure the similarity of two systems based on their recurrent dynamics rather than geometry or topology
large language models llms often struggle with
complex
language agents
large language models llms often struggle with complex mathematical reasoning where prose-based generation leads to unverified and arithmetically unsound solutions
current progress in electro-optical modulation within silicon
integrated
photonic devices
current progress in electro-optical modulation within silicon integrated photonics driven by the unique capabilities of advanced functional materials has led to significant improvements in device performance
2023 which derives general non-parametric bounds on biases due to omitted variables and is fully compatible with though not limited to modern inferential tools of causal
machine
debiased machine learning
2023 which derives general non-parametric bounds on biases due to omitted variables and is fully compatible with though not limited to modern inferential tools of causal machine learning
in all three scenarios we observe an inverse correlation between
x-ray
emission line
in all three scenarios we observe an inverse correlation between x-ray and optical emissions
taken together our study revealed a novel difference of functional property in response to task performance in the
sensorimotor
visual stimuli
taken together our study revealed a novel difference of functional property in response to task performance in the sensorimotor areas versus the association areas
purpose the purpose of this study was to determine if an ensemble of multiple llm agents could be used collectively to provide a more reliable assessment of a pixel-based ai
triage
triage tool
purpose the purpose of this study was to determine if an ensemble of multiple llm agents could be used collectively to provide a more reliable assessment of a pixel-based ai triage tool than a single llm
stellar-mass compact objects cos embedded in active
galactic
star-forming region
stellar-mass compact objects cos embedded in active galactic nucleus agn discs are commonly assumed to accrete via bondi or bondi-hoyle-lyttleton bhl prescriptions neglecting gas angular momentum
a prevailing theory in cognitive neuroscience proposes that the human brain operates through hierarchical predictive processing which underlies flexible
cognition
surrogate brain
a prevailing theory in cognitive neuroscience proposes that the human brain operates through hierarchical predictive processing which underlies flexible cognition and behavior by integrating multimodal sensory signals
wigner negativity and genuine multipartite entanglement
gme
wigner negativity
wigner negativity and genuine multipartite entanglement gme are key nonclassical resources that enable computational advantages and broader quantum-information tasks
for improvements we propose view-grpo a novel
reinforcement
reinforcement learning
for improvements we propose view-grpo a novel reinforcement learning framework that effectively strengthens view-specific temporal reasoning while encouraging consistent comprehension across viewpoints
recently morell and skutella suggested an alternative conjecture stating that one can turn any fractional
flow
maximum flow
recently morell and skutella suggested an alternative conjecture stating that one can turn any fractional flow into an unsplittable one without changing the load on any arc by more than the maximum demand
these findings establish quadratic nonlinear waveguide arrays as a promising platform to explore the interplay of nonlinearity topology and disorder in
quantum
quantum technologies
these findings establish quadratic nonlinear waveguide arrays as a promising platform to explore the interplay of nonlinearity topology and disorder in quantum photonic circuits
this paper studies the point convergence of accelerated gradient methods for unconstrained convex smooth multiobjective optimization problems covering both continuous-time
gradient
accelerated gradient
this paper studies the point convergence of accelerated gradient methods for unconstrained convex smooth multiobjective optimization problems covering both continuous-time gradient flows and discrete-time algorithms
2024 develops riesz regression for automatic
debiased
debiased machine
2024 develops riesz regression for automatic debiased machine learning which directly estimates the riesz representer or equivalently the bias-correction term by minimizing the mean squared error
in this regard embodying soft and physical
computing
physical computing
in this regard embodying soft and physical computing presents exciting potential
yet when these vlms are adapted to the action modality it remains unclear to what extent their original vl
representations
vision-language models vlms
yet when these vlms are adapted to the action modality it remains unclear to what extent their original vl representations and knowledge are preserved
here we harness the emission of a quantum
dot
quantum coherence
here we harness the emission of a quantum dot embedded in a micropillar and explore a hybrid approach where the information is encoded on a mixture of single photons and laser pulses
the empirical performance of the proposed methods is demonstrated through
numerical
numerical experiments
the empirical performance of the proposed methods is demonstrated through numerical experiments
mathematically annealed heterogeneity introduces a variance-weighted demographic noise term that penalizes across-environment fitness variance and effectively rescales the population size thereby biasing
evolution
evolutionary game
mathematically annealed heterogeneity introduces a variance-weighted demographic noise term that penalizes across-environment fitness variance and effectively rescales the population size thereby biasing evolution toward generalist solutions
using a mix of dynamic and static point correspondences we achieve competitive or superior point
tracking
point tracking
using a mix of dynamic and static point correspondences we achieve competitive or superior point tracking results on four datasets e
however the potential perceptual demands of viewing
virtual
virtual reality
however the potential perceptual demands of viewing virtual annotations while navigating a physical environment could impact user efficacy and safety and the implications of these demands are not well understood
our initial approach uses solve_bvp to approximate optimal
control
control systems
our initial approach uses solve_bvp to approximate optimal control trajectories
these results demonstrate efficiency gains without cognitive change suggesting that current narrow ai
systems
ai literacy
these results demonstrate efficiency gains without cognitive change suggesting that current narrow ai systems serve as cognitive scaffolds extending performance without transforming underlying mental capacities
braincognizer brain decoding with human visual
cognition
cognitive neuroscience
braincognizer brain decoding with human visual cognition simulation for fmri-to-image reconstruction
8 that is consistent with mrk 1018 s brightness before and after its latest
changing
surface brightness
8 that is consistent with mrk 1018 s brightness before and after its latest changing look event in the early 2010s
humans possess spatial reasoning abilities that enable them to understand spaces through
multimodal
computer vision
humans possess spatial reasoning abilities that enable them to understand spaces through multimodal observations such as vision and sound
in the era of rapid technological advancement
social
social media
in the era of rapid technological advancement social media platforms such as twitter x have emerged as indispensable tools for gathering consumer insights capturing diverse opinions and understanding public attitudes
furthermore we use our bounds to investigate the problem of data memorization raised in those works and which asserts that there are learning problem instances for which any learning algorithm that has good prediction there exist distributions under which the
algorithm
learning algorithm
furthermore we use our bounds to investigate the problem of data memorization raised in those works and which asserts that there are learning problem instances for which any learning algorithm that has good prediction there exist distributions under which the algorithm must memorize a big fraction of the training datas...
this superior performance not only sets a new benchmark for device responsivity and compactness but also opens promising avenues for future research including the incorporation of gain media for loss compensation at eps and the exploration of alternative tunable materials for next-generation ultracompact
photonic
photonic circuits
this superior performance not only sets a new benchmark for device responsivity and compactness but also opens promising avenues for future research including the incorporation of gain media for loss compensation at eps and the exploration of alternative tunable materials for next-generation ultracompact photonic devic...
quantitative electron magnetic circular dichroism emcd in transmission electron microscopy tem enables the measurement of
magnetic
magnetic properties
quantitative electron magnetic circular dichroism emcd in transmission electron microscopy tem enables the measurement of magnetic moments with elemental and atomic site sensitivity but its practical application is fundamentally limited by noise
for a perfect crystal under parallel illumination with a pink beam our results show that chromatic aberration is absent whereas under condensed
illumination
optical interference
for a perfect crystal under parallel illumination with a pink beam our results show that chromatic aberration is absent whereas under condensed illumination it becomes significant
moreover employing multiple riss significantly enhances performance compared to a single ris especially in line-of-sight los -dominated
wireless
wireless systems
moreover employing multiple riss significantly enhances performance compared to a single ris especially in line-of-sight los -dominated wireless environments
tmle is a method for constructing regression function estimators such that the leading bias
term
bias-correction term
tmle is a method for constructing regression function estimators such that the leading bias term becomes zero
to our knowledge this is the first algorithm to achieve an o log n competitive
ratio
competitive ratio
to our knowledge this is the first algorithm to achieve an o log n competitive ratio for non-trivial metrics beyond the i
quantum computation can be formulated through various models each highlighting distinct structural and resource-theoretic aspects of quantum
computational
quantum batteries
quantum computation can be formulated through various models each highlighting distinct structural and resource-theoretic aspects of quantum computational power
in this paper we prove that the iterates of the accelerated nesterov
s
superlinear convergence
in this paper we prove that the iterates of the accelerated nesterov s algorithm in the critical regime do converge in the weak topology to a global minimizer of an l -smooth function in a real hilbert space hence answering positively a conjecture posed by h
on the go with ar attention to virtual and
physical
virtual reality
on the go with ar attention to virtual and physical targets while varying augmentation density
in this work we review the underlying theory of these novel optical
scatterers
optical interference
in this work we review the underlying theory of these novel optical scatterers and some demonstrated configurations with enhanced resolution
distributional evaluation of generative models via
relative
generative models
distributional evaluation of generative models via relative density ratio
conventional robots possess a limited understanding of their kinematics and are confined to preprogrammed tasks hindering their ability to leverage
tools
mobile robots
conventional robots possess a limited understanding of their kinematics and are confined to preprogrammed tasks hindering their ability to leverage tools efficiently
specifically when f x acts safely but u0 acts unsafely the gain can be decreased by up to half and when f x acts unsafely we establish that if u0 acts safely the gain can be increased arbitrarily whereas if u0
acts
sensitivity tradeoff
specifically when f x acts safely but u0 acts unsafely the gain can be decreased by up to half and when f x acts unsafely we establish that if u0 acts safely the gain can be increased arbitrarily whereas if u0 acts unsafely the control recovers the full gain margin 1 2 inf
this energy is believed to impact the star formation
activity
star clusters
this energy is believed to impact the star formation activity and contribute to the quenching of galaxies
graph-theoretical mapping of resting-state
eeg
cognitive neuroscience
graph-theoretical mapping of resting-state eeg reveals neural signatures of creativity
while spike train data is well-approximated with a binary model here we apply these ig methods to data from electroencephalography
eeg
brain-computer interface
while spike train data is well-approximated with a binary model here we apply these ig methods to data from electroencephalography eeg a continuous signal requiring appropriate discretization strategies
the goal is to compute the starting positions of all fragments of t that can be obtained from p with edits of
total
-time algorithm
the goal is to compute the starting positions of all fragments of t that can be obtained from p with edits of total cost at most k
we show that minimizing the bayes cost mean of the cross-entropy loss implies maximizing the
mutual
brain decoding
we show that minimizing the bayes cost mean of the cross-entropy loss implies maximizing the mutual information between the set of categories and the neural activities prior to the decision layer
the numerical methods to do this efficiently depend on the properties of the
loss
loss function
the numerical methods to do this efficiently depend on the properties of the loss function
these results have implications for amplified oversight -- the challenge of combining humans and ai to supervise ai
systems
artificial intelligence
these results have implications for amplified oversight -- the challenge of combining humans and ai to supervise ai systems even as they surpass human expert performance
defining the urban local with low dimensional manifolds of human
mobility
scale-free networks
defining the urban local with low dimensional manifolds of human mobility networks
these findings provide new insights into how algorithmic reinforcement of clustering in online platforms can either sustain diversity of
opinion
opinion dynamics
these findings provide new insights into how algorithmic reinforcement of clustering in online platforms can either sustain diversity of opinion or accelerate its suppression
these findings support a clear takeaway improving
representation
representation learning
these findings support a clear takeaway improving representation learning is a direct and useful path to robust world models delivering reliable long-horizon predictions without enlarging the dynamics module
we hope the geometric view of parametric memory encourages revisiting the default intuitions that guide researchers in areas like
knowledge
reasoning capabilities
we hope the geometric view of parametric memory encourages revisiting the default intuitions that guide researchers in areas like knowledge acquisition capacity discovery and unlearning
results from perturbation theory along with lyapunov stability and eigen-spectrum analysis are used to prove the
convergence
convergence guarantees
results from perturbation theory along with lyapunov stability and eigen-spectrum analysis are used to prove the convergence towards the optimal case
genesis reproduces hallmark behavioral findings including generalization in semantic memory recognition serial recall effects and gist-based distortions in episodic memory and constructive
episodic
human cognition
genesis reproduces hallmark behavioral findings including generalization in semantic memory recognition serial recall effects and gist-based distortions in episodic memory and constructive episodic simulation while capturing their dynamic interactions
exciton and biexciton preparation via coherent swing-up excitation in a gaas
quantum
quantum dot
exciton and biexciton preparation via coherent swing-up excitation in a gaas quantum dot embedded in micropillar cavity
our system achieves accuracy increases of at least 16 on the synthetic csi dataset 20 on the in-lab csi
dataset
high accuracy
our system achieves accuracy increases of at least 16 on the synthetic csi dataset 20 on the in-lab csi dataset and 17 on the in-the-wild csi dataset
this paper discusses the different contemporaneous
causal
causal effects
this paper discusses the different contemporaneous causal interpretations of panel vector autoregressions pvar
we develop an efficient algorithm to solve this
bilevel
optimization problem
we develop an efficient algorithm to solve this bilevel optimization problem which computes parameter gradients without backpropagating through the solver
leveraging ideas from control-as-inference we show that minimizing the kullback-leibler divergence between a policy-driven true distribution and a reward-driven model distribution for a suitably defined action-driven process is equivalent to maximum entropy
reinforcement
reinforcement learning
leveraging ideas from control-as-inference we show that minimizing the kullback-leibler divergence between a policy-driven true distribution and a reward-driven model distribution for a suitably defined action-driven process is equivalent to maximum entropy reinforcement learning
this principle underpins neuroscience and
predictive
predictive processing
this principle underpins neuroscience and predictive brain theories like the free-energy principle or kolmogorov algorithmic agent theory
crucially we uncover an emergent capability for instruction-based decoding control the model learns to interpret natural
language
natural language
crucially we uncover an emergent capability for instruction-based decoding control the model learns to interpret natural language commands e
the recent advancement of multimodal large language models mllms is transforming human-computer
interaction
human-ai interaction
the recent advancement of multimodal large language models mllms is transforming human-computer interaction hci from surface-level exchanges into more nuanced and emotionally intelligent communication
we find considerable dependence of the streamer morphology on the environment which may act as a utility to constrain the physical conditions of the gas surrounding planet-forming disk and therefore the
conditions
dense gas
we find considerable dependence of the streamer morphology on the environment which may act as a utility to constrain the physical conditions of the gas surrounding planet-forming disk and therefore the conditions under which planets form
the reasoning capabilities of large language
models
language models
the reasoning capabilities of large language models llms have led to their increasing employment in several critical applications particularly education where they support problem-solving tutoring and personalized study
experiments show improved success rate and sample efficiency over single-algorithm baselines dqn or td3 alone and rule-based planners with better generalization to unseen obstacle configurations and reduced abrupt
control
optimal control
experiments show improved success rate and sample efficiency over single-algorithm baselines dqn or td3 alone and rule-based planners with better generalization to unseen obstacle configurations and reduced abrupt control changes
these solutions are surprisingly simple and have some interesting implications including a necessary and sufficient condition for mutation
selection
evolutionary game
these solutions are surprisingly simple and have some interesting implications including a necessary and sufficient condition for mutation selection balance a very simple formula for mean fitness and the fact that the shape of the equilibrium fitness distribution is determined solely by mutation whereas the scale is de...
we find evidence that the dispersion in the bulge mz r is influenced by both stellar accretion from satellites and migration from the disk such that at a fixed bulge mass bulges with higher fraction of accreted and migrated
stars
massive stars
we find evidence that the dispersion in the bulge mz r is influenced by both stellar accretion from satellites and migration from the disk such that at a fixed bulge mass bulges with higher fraction of accreted and migrated stars tend to be less metal-rich
we use 44 central galaxies from the cielo
cosmological
galaxy cgm
we use 44 central galaxies from the cielo cosmological simulations
prior work establishes theoretical guarantees for langevin monte
carlo
monte carlo
prior work establishes theoretical guarantees for langevin monte carlo algorithm based on overdamped and underdamped langevin dynamics and more recently some third-order variants
where a bayesian treatment is usually associated with high-quality
predictions
debiased machine learning
where a bayesian treatment is usually associated with high-quality predictions and uncertainties the practical reality has been the opposite with unstable training poor predictive power and subpar calibration
in the case of ktao3 first principles calculations have suggested that strain can drive a ferroelectric
phase
phase transition
in the case of ktao3 first principles calculations have suggested that strain can drive a ferroelectric phase transition
we propose a mathematically principled pde
gradient
gradient flow
we propose a mathematically principled pde gradient flow framework for distributionally robust optimization dro
current text-to-image generative models are trained on large uncurated datasets to enable diverse
generation
image generation
current text-to-image generative models are trained on large uncurated datasets to enable diverse generation capabilities
in this paper we combine adaptation and control barrier functions into a real-time control architecture that guarantees
stability
control strategy
in this paper we combine adaptation and control barrier functions into a real-time control architecture that guarantees stability ensures control performance and remains safe even with the parametric uncertainties
we find that despite surfacing errors different language models learn interchangeable
representations
large language models llms
we find that despite surfacing errors different language models learn interchangeable representations of numbers that are systematic highly accurate and universal across their hidden states and the types of input contexts
this paper provides bode-like sensitivity integrals for the quasiperiodic disturbance observer in both continuous-time and discrete-time representations and clarifies the avoided
sensitivity
sensitivity tradeoff
this paper provides bode-like sensitivity integrals for the quasiperiodic disturbance observer in both continuous-time and discrete-time representations and clarifies the avoided sensitivity tradeoff with time delays
we further designed an action-memory t-maze task and developed a microstate
sequence
recurrent neural
we further designed an action-memory t-maze task and developed a microstate sequence classifier mssc to predict rats turn decisions
think outside the policy in-context steered
policy
preference optimization
think outside the policy in-context steered policy optimization
we propose a mathematically principled pde gradient
flow
optimal transport
we propose a mathematically principled pde gradient flow framework for distributionally robust optimization dro
this study investigates the effect of knowledge distillation on the transferability of debiasing
capabilities
vision-language models
this study investigates the effect of knowledge distillation on the transferability of debiasing capabilities from teacher models to student models on natural language inference nli and image classification tasks
however brain signals infused with prior knowledge and associations exhibit a significant information asymmetry when compared to raw visual features still posing challenges for decoding
fmri
brain decoding
however brain signals infused with prior knowledge and associations exhibit a significant information asymmetry when compared to raw visual features still posing challenges for decoding fmri representations under the supervision of images
data-driven projection generation for efficiently solving heterogeneous quadratic
programming
efficiently solving
data-driven projection generation for efficiently solving heterogeneous quadratic programming problems
09 and red spectral slopes provides direct evidence for
galactic
active galactic
09 and red spectral slopes provides direct evidence for galactic cosmic ray gcr processing of the outer layers of the interstellar comet nucleus