prompt
stringlengths
41
511
target
stringlengths
1
25
keyword
stringclasses
697 values
full_sentence
stringlengths
48
1.25k
despite explicit instructions and contextual information our results show that models
perform
vision-language models
despite explicit instructions and contextual information our results show that models perform poorly in distinguishing loanwords from native ones
towards that direction this paper studies the performance of time series
classification
support vector machines
towards that direction this paper studies the performance of time series classification methods used as model selection for anomaly detection
with the current progress of artificial intelligence ai technology and its increasingly broader applications
trust
trustworthy ai
with the current progress of artificial intelligence ai technology and its increasingly broader applications trust is seen as a required criterion for ai usage acceptance and deployment
together these findings advance understanding of how density and urbanisation shape accident
risk
crash risk
together these findings advance understanding of how density and urbanisation shape accident risk and provide evidence to support more targeted road safety interventions and policy planning
across prediction horizons several deep learning models consistently outperformed classical
approaches
deep learning
across prediction horizons several deep learning models consistently outperformed classical approaches with the best model producing informative forecasts up to 1
while large language models llms have demonstrated remarkable performance across various reasoning tasks their ability to handle normative reasoning
remains
language models
while large language models llms have demonstrated remarkable performance across various reasoning tasks their ability to handle normative reasoning remains underexplored
moreover real-world preference learning often involves users with
different
preference optimization
moreover real-world preference learning often involves users with different preferences
to build the controller we first define tracking errors between the measured and desired
quadrotor
disturbance observer
to build the controller we first define tracking errors between the measured and desired quadrotor outputs which allow the system to be rewritten in a new set of state variables
878 and used 78 percent fewer tokens than wide
context
context engineering
878 and used 78 percent fewer tokens than wide context processing
asymmetric outcomes in two-actor conflict dynamics stability bifurcations and
emergent
emergent behaviors
asymmetric outcomes in two-actor conflict dynamics stability bifurcations and emergent behaviors
reducing base drag on road vehicles using pulsed jets optimized by hybrid
genetic
genetic algorithm
reducing base drag on road vehicles using pulsed jets optimized by hybrid genetic algorithms
higher-order hypergraph learning hohl was recently introduced as a principled alternative to classical hypergraph regularization enforcing
higher-order
graph neural
higher-order hypergraph learning hohl was recently introduced as a principled alternative to classical hypergraph regularization enforcing higher-order smoothness via powers of multiscale laplacians induced by the hypergraph structure
never too rigid to reach adaptive virtual model control with llm- and lyapunov-based
reinforcement
deep reinforcement learning
never too rigid to reach adaptive virtual model control with llm- and lyapunov-based reinforcement learning
for example comparing two neural systems can shed light on the nature of emergent
computations
neural codes
for example comparing two neural systems can shed light on the nature of emergent computations in the brain and deep neural networks
a single-loop first-order algorithm for linearly constrained
bilevel
optimization problem
a single-loop first-order algorithm for linearly constrained bilevel optimization
this is the first evidence connecting the complexity of 2d
compressed
compressed indexing
this is the first evidence connecting the complexity of 2d compressed indexing to long-standing open problems in the 1d setting
current slt evaluation metrics such as bleu and rouge are only text-based and it remains unclear to what extent text-based
metrics
evaluation metrics
current slt evaluation metrics such as bleu and rouge are only text-based and it remains unclear to what extent text-based metrics can reliably capture the quality of slt outputs
diverse emission patterns from precessing super-eddington disks formed in
tidal
tidal disruption
diverse emission patterns from precessing super-eddington disks formed in tidal disruption events
evaluated on a comprehensive dataset from the canadian prairies cypress demonstrates superior performance over existing deep learning-based
yield
yield prediction
evaluated on a comprehensive dataset from the canadian prairies cypress demonstrates superior performance over existing deep learning-based yield prediction models highlighting the effectiveness of fine-tuning foundation models for specialized agricultural applications
enhancing the reachability of variational quantum
algorithms
quantum dot
enhancing the reachability of variational quantum algorithms via input-state design
this paper extends this theoretical framework to the multiobjective setting focusing on the multiobjective inertial gradient system with asymptotically vanishing damping mavd with alpha 3 and the multiobjective accelerated proximal
gradient
accelerated gradient
this paper extends this theoretical framework to the multiobjective setting focusing on the multiobjective inertial gradient system with asymptotically vanishing damping mavd with alpha 3 and the multiobjective accelerated proximal gradient algorithm mapg
this flexibility enables detailed studies of orbitaland pseudo spin characteristics in
quantum
quantum materials
this flexibility enables detailed studies of orbitaland pseudo spin characteristics in quantum materials
we observe that it is attributed to the secret key mostly used in the watermark detection -- it cannot be public or the adversary may launch removal attacks provided the key nor can it be private or the
watermarking
watermarking schemes
we observe that it is attributed to the secret key mostly used in the watermark detection -- it cannot be public or the adversary may launch removal attacks provided the key nor can it be private or the watermarking detection is opaque to the public
however in data-sensitive fields such as healthcare the lack of high-quality domain-specific training corpus hinders
llms
large language models llms
however in data-sensitive fields such as healthcare the lack of high-quality domain-specific training corpus hinders llms adaptation for specialized applications
this tradeoff leaves neutral atom systems stuck between slow but accurate
readout
qubit readout
this tradeoff leaves neutral atom systems stuck between slow but accurate readout and fast but unreliable readout
however existing methods often rely on prior
knowledge
existing methods
however existing methods often rely on prior knowledge of problem parameters-such as smoothness convexity or communication network topologies-to determine appropriate stepsizes
we show that the rdr as a functional summary of the goodness-of-fit for the
generative
generative models
we show that the rdr as a functional summary of the goodness-of-fit for the generative model possesses several desirable theoretical properties
to ensure reward fidelity our automated grader calibration pipeline systematically purges noise from the llm-based
reward
reinforcement learning
to ensure reward fidelity our automated grader calibration pipeline systematically purges noise from the llm-based reward model with minimal human supervision
a closed-form approximation-free control law is derived to ensure that each agent remains within its evolving stt thereby avoiding dynamic
obstacles
optimal control
a closed-form approximation-free control law is derived to ensure that each agent remains within its evolving stt thereby avoiding dynamic obstacles while also preventing inter-agent collisions in a socially aware manner and reaching the target within a prescribed time
in this paper we explore a novel joint uplink and
downlink
uplink communication
in this paper we explore a novel joint uplink and downlink framework utilizing a pinching antenna system pass
in this paper we calculate one-dimensional thermal profiles of the third
interstellar
interstellar medium
in this paper we calculate one-dimensional thermal profiles of the third interstellar object 3i atlas throughout its trajectory in an attempt to gain insight into its bulk properties based on measurements of its volatiles
multi-dataset joint pre-training of emotional
eeg
brain-computer interface
multi-dataset joint pre-training of emotional eeg enables generalizable affective computing
state-of-the-art robotic tool usage methods focused on procedurally generating or crowd-sourcing datasets of tools for a
task
tool usage
state-of-the-art robotic tool usage methods focused on procedurally generating or crowd-sourcing datasets of tools for a task to learn how to grasp and manipulate them for that task
aci advances the detection of instantaneous causal relationships and the intermittent reversal of
causal
causal effects
aci advances the detection of instantaneous causal relationships and the intermittent reversal of causal roles over time
because observational inputs may be biased in ways unknown ex ante we develop a minimax proportional regret objective that evaluates any candidate design relative to an oracle that knows the bias and jointly chooses the
design
randomized experiments
because observational inputs may be biased in ways unknown ex ante we develop a minimax proportional regret objective that evaluates any candidate design relative to an oracle that knows the bias and jointly chooses the design and estimator
this paper proposes a structural multivariate unobserved components model with external instrument smuc-iv to investigate the effects of monetary
policy
monetary policy
this paper proposes a structural multivariate unobserved components model with external instrument smuc-iv to investigate the effects of monetary policy shocks on key u
we first establish nonparametric identification of the att under two minimal sets of
sufficient
nonparametric identification
we first establish nonparametric identification of the att under two minimal sets of sufficient conditions
we then train a neural network a multilayer perceptron to
predict
neural network
we then train a neural network a multilayer perceptron to predict local curvature from baseline sensor outputs recorded under no applied load achieving an r2 score of 0
predicted decline in common bird and butterfly
species
butterfly species
predicted decline in common bird and butterfly species despite conservation policies in europe
however the fundamental question of emph which problems with a deterministic complexity of
omega
omega log
however the fundamental question of emph which problems with a deterministic complexity of omega log n can be solved exponentially faster using randomization still remains wide open
certification and classification of linear quantum
error
quantum key distribution
certification and classification of linear quantum error mitigation methods
with the increasing number of flexible energy devices in distribution grids coordination between transmission system operators tsos and
distribution
power systems
with the increasing number of flexible energy devices in distribution grids coordination between transmission system operators tsos and distribution system operators dsos becomes critical for optimal system operation
large language models llms have demonstrated exceptional
capabilities
language agents
large language models llms have demonstrated exceptional capabilities across multiple domains by leveraging massive pre-training and curated fine-tuning data
self-improvement has emerged as a mainstream paradigm for advancing the reasoning capabilities of large
vision-language
language models
self-improvement has emerged as a mainstream paradigm for advancing the reasoning capabilities of large vision-language models lvlms where models explore and learn from successful trajectories iteratively
we evaluate csi2q on one synthetic csi dataset involving 85 devices and two real
csi
channel state information csi
we evaluate csi2q on one synthetic csi dataset involving 85 devices and two real csi datasets including 10 and 25 wifi routers respectively
moreover gromov-wasserstein exhibits an almost perfect correlation
rho
gromov-wasserstein distance
moreover gromov-wasserstein exhibits an almost perfect correlation rho 0
conventional robots possess a limited understanding of their kinematics and are confined to preprogrammed tasks hindering their ability to leverage
tools
robotic systems
conventional robots possess a limited understanding of their kinematics and are confined to preprogrammed tasks hindering their ability to leverage tools efficiently
saber symbolic regression-based angle of arrival and
beam
beam pattern
saber symbolic regression-based angle of arrival and beam pattern estimator
a minimal model of self-organized clusters with phase transitions in
ecological
ecological communities
a minimal model of self-organized clusters with phase transitions in ecological communities
combined with the distribution of serial intervals or generation times the rate gives
basic
disease transmission
combined with the distribution of serial intervals or generation times the rate gives basic and instantaneous values of the reproduction number that govern development and ultimate outcome of the epidemic
optimal learning makes this neural fisher information follow a category-specific
fisher
fisher information
optimal learning makes this neural fisher information follow a category-specific fisher information measuring the sensitivity of the category membership
this dataset establishes a foundation for advancing research in streaming video understanding complex
temporal
action recognition
this dataset establishes a foundation for advancing research in streaming video understanding complex temporal reasoning and multimodal inference
first we integrate a realistic network formation model where homophily and triadic closure co-evolve with a mean-field model of
opinion
opinion dynamics
first we integrate a realistic network formation model where homophily and triadic closure co-evolve with a mean-field model of opinion expression
the emergence of extremely large-scale antenna arrays elaa in millimeter-wave mmwave communications particularly in high-mobility scenarios highlights the importance of near-field
beam
near-field beam
the emergence of extremely large-scale antenna arrays elaa in millimeter-wave mmwave communications particularly in high-mobility scenarios highlights the importance of near-field beam prediction
our method represents observed motions using graph-based hierarchies explicitly decomposing global absolute motions into parent-inherited patterns and local
motion
optical flow
our method represents observed motions using graph-based hierarchies explicitly decomposing global absolute motions into parent-inherited patterns and local motion residuals
we then provide a taxonomy of linear mitigation methods characterising them by their
features
mitigation methods
we then provide a taxonomy of linear mitigation methods characterising them by their features and requirements
beam shaping techniques for pulsed laser ablation in
liquids
pulsed laser
beam shaping techniques for pulsed laser ablation in liquids unlocking tunable control of nanoparticle synthesis in liquids
focusing on scientific topics at the secondary education level we explore the potential of large
language
large language models llms
focusing on scientific topics at the secondary education level we explore the potential of large language models to generate chains of hints that scaffold learners without revealing answers
our approach bypasses the omega log n barrier introduced by probabilistic metric embeddings instead of analyzing the
embedding
tree embedding
our approach bypasses the omega log n barrier introduced by probabilistic metric embeddings instead of analyzing the embedding distortion and the algorithm separately we directly bound the cost of the algorithm on the target metric of a simple deterministic embedding
while cave automatic virtual environment cave systems have long enabled room-scale
virtual
physical virtual
while cave automatic virtual environment cave systems have long enabled room-scale virtual reality and various kinds of interactivity their content has largely remained predetermined
while humans intuitively perform this translation based on common sense and embodied understanding whether large
language
large language models llms
while humans intuitively perform this translation based on common sense and embodied understanding whether large language models llms can effectively replicate this ability remains underexplored
plugging regression functions estimated by machine
learning
machine learning
plugging regression functions estimated by machine learning methods into the identifying equations can yield poor performance because of first-stage bias
to address these challenges we propose unifiedfl a dynamic federated learning framework that represents heterogeneous local networks as nodes and edges in a directed model graph optimized by a shared
graph
graph neural
to address these challenges we propose unifiedfl a dynamic federated learning framework that represents heterogeneous local networks as nodes and edges in a directed model graph optimized by a shared graph neural network gnn
in many real-world scenarios multiple adapters are loaded simultaneously to enable
llm
models llms
in many real-world scenarios multiple adapters are loaded simultaneously to enable llm customization for personalized user experiences or to support a diverse range of tasks
choosing what to learn experimental design when combining experimental with
observational
randomized experiments
choosing what to learn experimental design when combining experimental with observational evidence
the accretion is viscosity-limited when q alpha xi 1 mathcal m 2 3 3 1 2 h 3 where q is the mass ratio between the co and the
supermassive
black hole
the accretion is viscosity-limited when q alpha xi 1 mathcal m 2 3 3 1 2 h 3 where q is the mass ratio between the co and the supermassive black hole alpha the viscosity parameter mathcal m the mach number of the bulk relative motion and h the aspect ratio of the agn disc
riesz regression covariate balancing dre and the matching estimator are methods for estimating the balancing weights where
riesz
riesz regression
riesz regression covariate balancing dre and the matching estimator are methods for estimating the balancing weights where riesz regression is essentially equivalent to dre in the ate context the matching estimator is a special case of dre and dre is in a dual relationship with covariate balancing
under exclusion restrictions analogous to those in the lgr model we establish nonparametric point
identification
nonparametric identification
under exclusion restrictions analogous to those in the lgr model we establish nonparametric point identification of the latent categorical distribution
we find that despite surfacing errors different
language
large language
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
point convergence of nesterov s accelerated
gradient
gradient flow
point convergence of nesterov s accelerated gradient method an ai-assisted proof
to render the theory practically applicable we further develop an emph online variance estimator for the asymptotic variance appearing in the clt and
establish
central limit theorem
to render the theory practically applicable we further develop an emph online variance estimator for the asymptotic variance appearing in the clt and establish emph high-probability deviation bounds for this estimator
we find that more massive smbhs have higher surface densities of non-rotating
molecular
dense gas
we find that more massive smbhs have higher surface densities of non-rotating molecular gas within their sphere of influence
spin systems have emerged as powerful tools for understanding collective phenomena in
complex
complex systems
spin systems have emerged as powerful tools for understanding collective phenomena in complex systems
in the absence of disturbances we find that standard
inverse
inverse optimal
in the absence of disturbances we find that standard inverse optimal safe controllers have a certain degree of gain margin
we analyse 19 spiral disks from the cosmological rtnscrimhd azahar suite deriving line-of-sight integrated maps to measure median magnetic-field strength b specific energies thermal turbulent magnetic and cosmic-ray and star formation rate sfr star formation surface density sigma_ mathrm sfr and specific
sfr
star-forming region
we analyse 19 spiral disks from the cosmological rtnscrimhd azahar suite deriving line-of-sight integrated maps to measure median magnetic-field strength b specific energies thermal turbulent magnetic and cosmic-ray and star formation rate sfr star formation surface density sigma_ mathrm sfr and specific sfr ssfr
while demonstrated in orchard monitoring the approach can be applied to other outdoor domains requiring robust
multimodal
computer vision
while demonstrated in orchard monitoring the approach can be applied to other outdoor domains requiring robust multimodal perception
to explain this surprising redundancy we develop a cross-evaluation protocol in which we apply each linear
decoder
cross dual encoder network
to explain this surprising redundancy we develop a cross-evaluation protocol in which we apply each linear decoder operator to the subjects of every other relation
during inference slideagent selectively activates specialized
agents
reasoning capabilities
during inference slideagent selectively activates specialized agents for multi-level reasoning and integrates their outputs into coherent context-aware answers
furthermore we provide several characterizations of this class including nonsmooth and differentiable cases and derive key properties that fa -ci -li -ta -te the implementation of
first-order
zeroth-order methods
furthermore we provide several characterizations of this class including nonsmooth and differentiable cases and derive key properties that fa -ci -li -ta -te the implementation of first-order methods
purpose medical foundation models fms offer a
path
foundation models
purpose medical foundation models fms offer a path to build high-performance diagnostic systems
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
large language
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 to the physical world
using a lantern-inspired flexible electrode we achieve long-term stable recordings and robust memory decision decoding from within the ventricular system opening new directions for
bci
brain-computer interface
using a lantern-inspired flexible electrode we achieve long-term stable recordings and robust memory decision decoding from within the ventricular system opening new directions for bci technology and systems neuroscience
beating the winner s curse via inference-aware
policy
policy learning
beating the winner s curse via inference-aware policy optimization
from incremental transitive cover to strongly polynomial
maximum
maximum flow
from incremental transitive cover to strongly polynomial maximum flow
current tool-use large language models llms are trained on static datasets enabling them to interact with external tools and
perform
open-source models
current tool-use large language models llms are trained on static datasets enabling them to interact with external tools and perform multi-step tool-integrated reasoning which produces tool-call trajectories
we develop a direct debiased machine learning framework comprising neyman targeted estimation and generalized
riesz
riesz regression
we develop a direct debiased machine learning framework comprising neyman targeted estimation and generalized riesz regression
a total of 5 uavs with optimal location schemes would be sufficient to detect over 95 of the paths in the network considering both microscopic uncertainty regarding the intersection operation efficiency and the macroscopic
uncertainty
optimal uav
a total of 5 uavs with optimal location schemes would be sufficient to detect over 95 of the paths in the network considering both microscopic uncertainty regarding the intersection operation efficiency and the macroscopic uncertainty regarding the route choice of road users
yet the classic theoretical results of population
genetics
large population
yet the classic theoretical results of population genetics e
to address these issues this study introduces the
human-ai
artificial intelligence
to address these issues this study introduces the human-ai re synergy model hare-sm a conceptual framework that integrates ai-driven analysis with human oversight to improve requirements elicitation analysis and validation
consequently each iteration has a computational cost that is proportional to the number of spatial degrees of freedom producing a scalable method that preserves the established strong
convergence
zeroth-order methods
consequently each iteration has a computational cost that is proportional to the number of spatial degrees of freedom producing a scalable method that preserves the established strong convergence rates
discovering heuristics with large language
models
language models
discovering heuristics with large language models llms for mixed-integer programs single-machine scheduling
designing efficient quantum circuits is a central bottleneck to exploring the potential of quantum
computing
quantum algorithm
designing efficient quantum circuits is a central bottleneck to exploring the potential of quantum computing particularly for noisy intermediate-scale quantum nisq devices where circuit efficiency and resilience to errors are paramount
they may significantly affect the cosmological
environment
dark matter
they may significantly affect the cosmological environment of their host galaxy
instrumental variable methods are fundamental to
causal
causal effects
instrumental variable methods are fundamental to causal inference when treatment assignment is confounded by unobserved variables
the second was a hybrid approach that used the cnn as a feature extractor and then applied a support vector
machine
neural networks
the second was a hybrid approach that used the cnn as a feature extractor and then applied a support vector machine svm classifier
in this work we present a new randomized algorithm that improves the cut-query
complexity
randomized algorithm
in this work we present a new randomized algorithm that improves the cut-query complexity to widetilde o n 8 5
the ability to use random objects as tools in a generalizable manner is a missing piece in
robots
robotic manipulation
the ability to use random objects as tools in a generalizable manner is a missing piece in robots intelligence today to boost their versatility and problem-solving capabilities
furthermore the dual-bus scheme operates broadband spanning visible to mid-infrared across all four transmission channels highlighting its spectral richness and
platform
optical communication
furthermore the dual-bus scheme operates broadband spanning visible to mid-infrared across all four transmission channels highlighting its spectral richness and platform independence
the second part learns the neural stability
descriptor
deep neural
the second part learns the neural stability descriptor by iteratively training the nns with sample augmentation guided by the tailored conservativeness-aware loss function
these findings support a clear takeaway improving representation learning is a direct and useful path to robust
world
world models
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