prompt
stringlengths
41
511
target
stringlengths
1
25
keyword
stringclasses
697 values
full_sentence
stringlengths
48
1.25k
leveraging this dataset we fine-tune llms to generate code
review
review comments
leveraging this dataset we fine-tune llms to generate code review comments that can effectively identify security issues and provide fix suggestions with our proposed secure-aware fine-tuning strategy
combining pmi and feature modeling our approach enables
continual
continual learning
combining pmi and feature modeling our approach enables continual learning of new classes by generating pseudo-images from semantic-aware projected features achieving strong effectiveness and compatibility across multiple cl settings
using the data from the dawn jwst archive we conduct a comprehensive study on the large-scale structure stellar mass function smf quiescent members and
dark
dark matter
using the data from the dawn jwst archive we conduct a comprehensive study on the large-scale structure stellar mass function smf quiescent members and dark matter halos in the cosmic vine
the origin of radio afterglows or delayed radio flares in
tidal
tidal disruption
the origin of radio afterglows or delayed radio flares in tidal disruption events tdes is not fully understood
simulation of the time-dynamics of fermionic many-body systems has
long
classical simulation
simulation of the time-dynamics of fermionic many-body systems has long been predicted to be one of the key applications of quantum computers
wave guides for classical electromagnetic fields can realize the
quantum
quantum dot
wave guides for classical electromagnetic fields can realize the quantum evolution of the wave function for a system of qubits
stakeholders share common perceptions on biodiversity impacts despite geographic disparity but they differentiate between climate and
land
land use
stakeholders share common perceptions on biodiversity impacts despite geographic disparity but they differentiate between climate and land use impacts
a decomposition of the qlf by black hole mass reveals that this boost is primarily driven by low-mass black
holes
black hole
a decomposition of the qlf by black hole mass reveals that this boost is primarily driven by low-mass black holes radiating above the eddington limit
this paper investigates the construction of channel knowledge map ckm from sparse
channel
channel state information
this paper investigates the construction of channel knowledge map ckm from sparse channel measurements
for the optimal welfare functional defined as the average value of cate on the subpopulation with nonnegative cate we establish the sqrt n asymptotic normality of the semiparametric plug-in estimators and provide an analytical
asymptotic
asymptotic normality
for the optimal welfare functional defined as the average value of cate on the subpopulation with nonnegative cate we establish the sqrt n asymptotic normality of the semiparametric plug-in estimators and provide an analytical asymptotic variance formula
experiments are conducted with ocular images from vggface2-pose a subset of vggface2 containing in-the-wild face
images
computer vision
experiments are conducted with ocular images from vggface2-pose a subset of vggface2 containing in-the-wild face images and the ufpr-periocular database which consists of selfies captured via mobile devices with user guidance on the screen
these accretion-based m_ bullet values which in this sample span nearly three orders of magnitude are consistent with
galactic
galactic disk
these accretion-based m_ bullet values which in this sample span nearly three orders of magnitude are consistent with galactic scaling relations but are significantly more precise 68 credible interval pm 0
phylogenetic inference the task of reconstructing how related sequences evolved from common ancestors is a central task in
evolutionary
phylogenetic diversity
phylogenetic inference the task of reconstructing how related sequences evolved from common ancestors is a central task in evolutionary genomics
masiero lifting partial smoothing to solve
hjb
hjb equation
masiero lifting partial smoothing to solve hjb equations and stochastic control problems siam journal on control and optimization 63 3 2025 pp
furthermore we show that even without explaining the details of
pragmatic
pragmatic theories
furthermore we show that even without explaining the details of pragmatic theories merely mentioning their names in the prompt leads to a certain performance improvement around 1-3 in larger models compared to the baseline
experimental results showed that compared to the baseline which prompts intermediate reasoning without presenting
pragmatic
vision-language models
experimental results showed that compared to the baseline which prompts intermediate reasoning without presenting pragmatic theories 0-shot chain-of-thought our methods enabled language models to achieve up to 9
vision-language models vlms exhibit uneven performance across
languages
language agents
vision-language models vlms exhibit uneven performance across languages a problem that is often exacerbated when the model size is reduced
a unified theory for causal inference direct debiased machine
learning
causal inference
a unified theory for causal inference direct debiased machine learning via bregman-riesz regression
during inference slideagent selectively activates specialized
agents
reasoning tasks
during inference slideagent selectively activates specialized agents for multi-level reasoning and integrates their outputs into coherent context-aware answers
the simulations reveal that tensile strain modifies the dynamics of defect generation and recovery promoting stress-assisted defect mobility and enhancing
defect
strain engineering
the simulations reveal that tensile strain modifies the dynamics of defect generation and recovery promoting stress-assisted defect mobility and enhancing defect survival compared to the unstrained case
anomaly detection in time-series data is a critical challenge with significant
implications
time series classification
anomaly detection in time-series data is a critical challenge with significant implications for network security
simulation of the time-dynamics of fermionic many-body systems has
long
numerical simulations
simulation of the time-dynamics of fermionic many-body systems has long been predicted to be one of the key applications of quantum computers
in recent work it has been shown that this random return can be well approximated by a finite number of random variables which we call truncated
random
random return
in recent work it has been shown that this random return can be well approximated by a finite number of random variables which we call truncated random return
our results have applications across diverse contexts from behavioural ecology to bio-inspired collective
systems
evolutionary dynamics
our results have applications across diverse contexts from behavioural ecology to bio-inspired collective systems design
the approach is evaluated across different network models
network
viral replication
the approach is evaluated across different network models network sizes and fraction of observed infections
sensitivity analysis for treatment effects in
difference-in-differences
treatment effect boundaries
sensitivity analysis for treatment effects in difference-in-differences models using riesz representation
the system integrates industrial cameras to monitor equipment operation alignment and hot bar
motion
point tracking
the system integrates industrial cameras to monitor equipment operation alignment and hot bar motion in real time along the process line
the simulated galaxies reproduce the observed stellar-to-halo mass mass--metallicity and size--mass relations yielding stellar
masses
galaxy cgm
the simulated galaxies reproduce the observed stellar-to-halo mass mass--metallicity and size--mass relations yielding stellar masses of 10 6-10 8 m_ odot and metallicities consistent with those of local group dwarf galaxies
building on this insight we propose attncache a framework that accelerates the prefill stage of llm inference by retrieving and reusing similar
attention
vision-language models
building on this insight we propose attncache a framework that accelerates the prefill stage of llm inference by retrieving and reusing similar attention maps
these results demonstrate that spiral density waves can persist in fully
cosmological
galaxy cgm
these results demonstrate that spiral density waves can persist in fully cosmological disks linking internal dynamical processes to galaxy assembly and offering testable predictions for present and future surveys such as jwst and roman
biological and artificial learners are inherently exposed to a stream of data and
experience
artificial intelligence
biological and artificial learners are inherently exposed to a stream of data and experience throughout their lifetimes and must constantly adapt to learn from or selectively ignore the ongoing input
specifically braincognizer introduces two modules the cognitive integration module which incorporates prior human knowledge to extract hierarchical region semantics and the
cognitive
brain activity
specifically braincognizer introduces two modules the cognitive integration module which incorporates prior human knowledge to extract hierarchical region semantics and the cognitive correlation module which captures contextual semantic relationships across regions
2023 shows that nn matching is an instance of density-ratio estimation with their new
density-ratio
density estimation
2023 shows that nn matching is an instance of density-ratio estimation with their new density-ratio estimator
our work significantly extends the existing results on the convergence of gd and sgd by guaranteeing that they apply to practical
neural
neural network
our work significantly extends the existing results on the convergence of gd and sgd by guaranteeing that they apply to practical neural network settings and has the potential to unlock further exploration of learning dynamics
gaussian processes gps on the other hand are often preferred in
uncertainty
uncertainty quantification
gaussian processes gps on the other hand are often preferred in uncertainty quantification tasks due to their interpretability
anomaly detection is a fundamental task for time
series
time series classification
anomaly detection is a fundamental task for time series analytics with important implications for the downstream performance of many applications
in this article we provide a self-contained introduction to the limit theory of dense
multiplex
multiplex networks
in this article we provide a self-contained introduction to the limit theory of dense multiplex networks analogous to the theory of graphons limit theory of dense graphs
however these evaluations rely on llms proxy
llms
models llms
however these evaluations rely on llms proxy llms to gauge compliance with privacy norms overlooking real users perceptions
with stesso it has been experimentally proven that n 1 -bit toffoli
gates
-bit toffoli gates
with stesso it has been experimentally proven that n 1 -bit toffoli gates always have lower quantum costs than using conventional composition methods
human feedback is critical for aligning ai
systems
ai use
human feedback is critical for aligning ai systems to human values
prior research examined how llms alter user views yet little work extended beyond one-way influence to address how user input can affect
llm
llm raters
prior research examined how llms alter user views yet little work extended beyond one-way influence to address how user input can affect llm responses and how such bi-directional influence manifests throughout the multi-turn conversations
we present combined jwst nirspec and miri mrs integral field spectroscopy data of the nuclear and circumnuclear regions of the highly dust obscured seyfert 2 galaxy ngc 7582 which is part of the
sample
dwarf galaxies
we present combined jwst nirspec and miri mrs integral field spectroscopy data of the nuclear and circumnuclear regions of the highly dust obscured seyfert 2 galaxy ngc 7582 which is part of the sample of agn in the galaxy activity torus and outflow survey gatos
the generic population concept - agent-based model henceforth short gepoc abm is one of the models within
gepoc
population size
the generic population concept - agent-based model henceforth short gepoc abm is one of the models within gepoc a generic concept to model a country s population and its dynamics using causal modelling approaches
neon is an inhibitor of co hydrogenation in pre-stellar
core
star-forming region
neon is an inhibitor of co hydrogenation in pre-stellar core conditions
we find that simply imposing the m_ rm bh - sigma_ star condition is sufficient to generate a
fraction
star clusters
we find that simply imposing the m_ rm bh - sigma_ star condition is sufficient to generate a fraction of quenched galaxies consistent with current data including the newest ones from euclid
these findings demonstrate the potential of ai
agents
ai assistance
these findings demonstrate the potential of ai agents to move beyond process partners toward colleagues that share intent strengthen group dynamics and collaborate with humans to advance ideas
building on this inpainting core we design a spatiotemporal autoregressive inference pipeline that traverses virtual-camera splines and extends videos with overlapping windows enabling coherent
generation
video generation
building on this inpainting core we design a spatiotemporal autoregressive inference pipeline that traverses virtual-camera splines and extends videos with overlapping windows enabling coherent generation at bounded per-step complexity
in this work we present a fully unsupervised machine
learning
machine learning ml
in this work we present a fully unsupervised machine learning ml workflow that detects and classifies these defects directly from molecular dynamics data
augmented reality ar technologies are redefining how we perceive and interact with the world by seamlessly integrating digital elements into our
physical
augmented reality
augmented reality ar technologies are redefining how we perceive and interact with the world by seamlessly integrating digital elements into our physical surroundings
dynamic spatial treatment effects as continuous functionals theory and
evidence
spatial treatment
dynamic spatial treatment effects as continuous functionals theory and evidence from healthcare access
to construct the what system we first show that a large family of tasks can be systematically described by a
probabilistic
recurrent neural networks
to construct the what system we first show that a large family of tasks can be systematically described by a probabilistic generative model where compositionality stems from a shared underlying vocabulary of discrete task epochs
furthermore the probabilistic generative capability of the
diffusion
diffusion models
furthermore the probabilistic generative capability of the diffusion model is integrated with physical laws forming a time-varying physical guidance mechanism that adaptively regulates the generation process at different stages thereby enhancing the ability of the model to perceive the intrinsic structure of data and reducing dependence on data quality
expanding on the graph theoretic ideas of k-component order connectivity and distance-l domination we present a quadratic-complexity algorithm that finds a
tree
spanning trees
expanding on the graph theoretic ideas of k-component order connectivity and distance-l domination we present a quadratic-complexity algorithm that finds a tree s minimum failure-set cardinality i
this notion of differentiability benefits from calculus rules and is sufficient to prove superlinear
convergence
superlinear convergence
this notion of differentiability benefits from calculus rules and is sufficient to prove superlinear convergence of a newton-type method
machine learning approaches for image classification have
led
image classification
machine learning approaches for image classification have led to impressive advances in that field
we successfully generate a three-photon genuinely entangled state from two bi-separable
states
entanglement entropy
we successfully generate a three-photon genuinely entangled state from two bi-separable states via local operations and classical communication demonstrating superactivation of genuine multipartite entanglement
our comprehensive evaluation of open-weight and proprietary reasoning and non-reasoning vlms
reveals
open-source models
our comprehensive evaluation of open-weight and proprietary reasoning and non-reasoning vlms reveals that most models perform near chance and even the best lag far behind human accuracy on physically irreversible processes e
quantum error correction thermalization and quantum
chaos
quantum coherence
quantum error correction thermalization and quantum chaos are fundamental aspects of quantum many-body physics that have each developed largely independently despite their deep conceptual overlap
while deep learning has recently advanced the state-of-the-art in the time series forecasting literature its application to neural activity
forecasting
deep learning
while deep learning has recently advanced the state-of-the-art in the time series forecasting literature its application to neural activity forecasting remains limited
we show that a single population of spiking
neurons
artificial neural
we show that a single population of spiking neurons can learn the full prediction object through a biologically grounded three factor hebbian rule
experiments demonstrate the effectiveness of the proposed framework under mobility-induced dynamics and offer useful insights for the practical deployment of fl over
wireless
wireless networks
experiments demonstrate the effectiveness of the proposed framework under mobility-induced dynamics and offer useful insights for the practical deployment of fl over wireless channels
social influence on complex networks as a perturbation to
individual
social interactions
social influence on complex networks as a perturbation to individual behavior
distributed stochastic momentum tracking with local
updates
data-driven stabilization
distributed stochastic momentum tracking with local updates achieving optimal communication and iteration complexities
stellar-mass compact objects cos embedded in active
galactic
stellar mass
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
specifically we propose an additive instrumental
variable
instrumental variable
specifically we propose an additive instrumental variable framework to identify mean potential outcomes and the average treatment effect with a weighting function
reinforcement learning rl can elicit strong
reasoning
reinforcement learning
reinforcement learning rl can elicit strong reasoning in large language models llms yet most open efforts focus on math and code
these flexible spatial-kinematic absorption models skams can be directly applied to and or easily modified expanded for studying individual or ensembles of observed absorption line systems for exploring various competing theoretical scenarios of the baryon cycle as studied through quasar
absorption
absorption line
these flexible spatial-kinematic absorption models skams can be directly applied to and or easily modified expanded for studying individual or ensembles of observed absorption line systems for exploring various competing theoretical scenarios of the baryon cycle as studied through quasar absorption line systems and or serving as pedagogical tools for developing physical intuition
detecting the use of generative ai in crowdsourced surveys implications for
data
trustworthy ai
detecting the use of generative ai in crowdsourced surveys implications for data integrity
a long-standing question remained on its capacity to tackle
deep
deep learning
a long-standing question remained on its capacity to tackle deep learning models capturing rich feature learning effects thus going beyond the narrow networks or kernel methods analysed until now
in this chapter we relate the general definition of accountability to ai we illustrate what it means for ai to be accountable and unaccountable and we explore approaches that can
improve
ai agents
in this chapter we relate the general definition of accountability to ai we illustrate what it means for ai to be accountable and unaccountable and we explore approaches that can improve our chances of living in a world where all ai is accountable to those who are affected by it
large language models llms are increasingly used as
raters
models llms
large language models llms are increasingly used as raters for evaluation tasks
emerging information technologies like social media search engines and ai can have a broad impact on public health political institutions
social
social media
emerging information technologies like social media search engines and ai can have a broad impact on public health political institutions social dynamics and the natural world
to construct the what system we first show that a large family of tasks can be systematically described by a probabilistic generative model where compositionality stems from a shared underlying
vocabulary
large language models
to construct the what system we first show that a large family of tasks can be systematically described by a probabilistic generative model where compositionality stems from a shared underlying vocabulary of discrete task epochs
finally we outline several potential applications of nonlinear sis in
wireless
wireless networks
finally we outline several potential applications of nonlinear sis in wireless communication scenarios
however the lack of dispersion between textual features can hurt calibration performance which raises concerns about
vlms
models vlms
however the lack of dispersion between textual features can hurt calibration performance which raises concerns about vlms reliability trustworthiness and safety
we present a lateral ventricular brain-computer interface lv-bci that deploys an expandable flexible
electrode
brain-computer interface
we present a lateral ventricular brain-computer interface lv-bci that deploys an expandable flexible electrode into the lateral ventricle through a minimally invasive external ventricular drainage pathway
there are two major approaches in policy learning the empirical
welfare
policy learning
there are two major approaches in policy learning the empirical welfare maximization ewm approach and the plug-in approach
by judiciously setting spatial parameters the system can be tuned to exhibit both weak and strong
coupling
coupling regimes
by judiciously setting spatial parameters the system can be tuned to exhibit both weak and strong coupling regimes between the plasmonic and dielectric modes leading to the controlled formation of ep degeneracies
to tackle this challenge we propose braincognizer a novel brain decoding model inspired by human visual
cognition
cognitive science
to tackle this challenge we propose braincognizer a novel brain decoding model inspired by human visual cognition which explores multi-level semantics and correlations without fine-tuning of generative models
rather than discrete treatment effect estimators the framework characterizes treatment intensity as a continuous function tau mathbf x t over space-time enabling rigorous analysis of propagation dynamics
boundary
effect boundaries
rather than discrete treatment effect estimators the framework characterizes treatment intensity as a continuous function tau mathbf x t over space-time enabling rigorous analysis of propagation dynamics boundary evolution and cumulative exposure patterns
simulation studies further demonstrate the accuracy and practical value of the
proposed
numerical experiments
simulation studies further demonstrate the accuracy and practical value of the proposed approach
we explore the fate of different strategies under sustained environmental
change
climate change
we explore the fate of different strategies under sustained environmental change by considering a mathematical model for a large population of asexual organisms
we present necessary and or sufficient conditions for identification estimation and inference in large samples both
pointwise
density-ratio estimation
we present necessary and or sufficient conditions for identification estimation and inference in large samples both pointwise and uniformly along the boundary
recent findings reveal that even when the performance remains stable the underlying neural representations can change gradually over
time
continual learning
recent findings reveal that even when the performance remains stable the underlying neural representations can change gradually over time a phenomenon known as representational drift
we present sia social insight agents an llm
agent
large language models llms
we present sia social insight agents an llm agent system that links heterogeneous multi-modal data -- including raw inputs e
recent works in continuous-time estimation techniques for
crs
state estimation
recent works in continuous-time estimation techniques for crs show a principled approach to addressing this runtime constraint but are currently restricted to offline operation
non-monotone traveling waves of the weak competition
lotka-volterra
traveling waves
non-monotone traveling waves of the weak competition lotka-volterra system
we present conceptscope a scalable and automated framework for analyzing visual datasets by discovering and quantifying human-interpretable concepts using
sparse
sparse autoencoders
we present conceptscope a scalable and automated framework for analyzing visual datasets by discovering and quantifying human-interpretable concepts using sparse autoencoders trained on representations from vision foundation models
our study addresses the challenges inherent in calculating mixed fermion-boson systems and explores the potential of quantum
computing
quantum computing
our study addresses the challenges inherent in calculating mixed fermion-boson systems and explores the potential of quantum computing to advance their analysis
the model naturally accepts interleaved vision-language inputs and generates
interleaved
vision transformers
the model naturally accepts interleaved vision-language inputs and generates interleaved vision-language outputs
that is surprising as this framework is naturally suited for hierarchical control applications in general and
autonomous
collision avoidance
that is surprising as this framework is naturally suited for hierarchical control applications in general and autonomous driving tasks in specific
star quasiconvexity an unified approach for linear
convergence
linear convergence
star quasiconvexity an unified approach for linear convergence of first-order methods beyond convexity
controlling specific behaviors in large language models while preserving their general
capabilities
large language
controlling specific behaviors in large language models while preserving their general capabilities is a central challenge for safe and reliable artificial intelligence deployment
the majority of research in both training artificial neural networks anns and modeling learning in
biological
artificial neural
the majority of research in both training artificial neural networks anns and modeling learning in biological brains focuses on synaptic plasticity where learning equates to changing the strength of existing connections
inference-cost-aware dynamic tree construction for efficient
inference
vision-language models
inference-cost-aware dynamic tree construction for efficient inference in large language models
then a central question emerges how can machines better understand our
situations
normative reasoning
then a central question emerges how can machines better understand our situations and purposes
recent advances advocate easily obtainable
channel
channel state
recent advances advocate easily obtainable channel state information csi by commercial wifi devices for lightweight rf fingerprinting while falling short in addressing the challenges of coarse granularity of csi measurements in an open-world setting
reliable uncertainty quantification is crucial for
reinforcement
reinforcement learning
reliable uncertainty quantification is crucial for reinforcement learning rl in high-stakes settings
while prior studies aim to explain this effect most
theoretical
theoretical findings
while prior studies aim to explain this effect most theoretical insights are limited to abstract frameworks or linear random feature models
automating the co-design of a robot s morphology and
control
robotic systems
automating the co-design of a robot s morphology and control is a long-standing challenge due to the vast design space and the tight coupling between body and behavior