prompt
stringlengths
41
511
target
stringlengths
1
25
keyword
stringclasses
697 values
full_sentence
stringlengths
48
1.25k
we characterize energy number location and hessian eigenspectra of
global
global minima
we characterize energy number location and hessian eigenspectra of global minima local minima and critical points as the landscape evolves
we hope these findings motivate a broader reconsideration of precision trade-offs in
rl
reinforcement learning rl
we hope these findings motivate a broader reconsideration of precision trade-offs in rl fine-tuning
we present a deep reinforcement learning framework based on proximal
policy
reinforcement learning
we present a deep reinforcement learning framework based on proximal policy optimization ppo for autonomous qos-aware load balancing implemented end-to-end in a lightweight pure-python simulation environment
while a multi-agent approach based on large
language
language models
while a multi-agent approach based on large language models llms represents a promising strategy to surpass the capabilities of single models its success is critically dependent on synergistic team composition
a logic-based algorithmic meta-theorem for
treedepth
tree edit
a logic-based algorithmic meta-theorem for treedepth single exponential fpt time and polynomial space
we suggest that interactions between parallel subnetworks in the brain may underlie such learning we present a model of representation learning by ensembles of neural networks where each network learns to encode stimuli into an abstract representation space by cross-supervising
interactions
brain-computer interface
we suggest that interactions between parallel subnetworks in the brain may underlie such learning we present a model of representation learning by ensembles of neural networks where each network learns to encode stimuli into an abstract representation space by cross-supervising interactions with other networks for inpu...
harnessing topological effects offers a promising route to protect quantum states of light from imperfections potentially enabling more robust platforms for
quantum
quantum technologies
harnessing topological effects offers a promising route to protect quantum states of light from imperfections potentially enabling more robust platforms for quantum information processing
this opens up a potential cognitive attack in which adversaries might create conditions that make an
ar
mobile ar
this opens up a potential cognitive attack in which adversaries might create conditions that make an ar user not recall certain potentially mission-critical objects
this survey aims to establish a structured foundation for advancing hgnn-based
anomaly
anomaly detection
this survey aims to establish a structured foundation for advancing hgnn-based anomaly detection toward scalable interpretable and practically deployable solutions
we then separately solve a dirichlet problem for laplace s equation to synthesize a safe textit guidance field that encodes variable levels of caution around
obstacles
collision avoidance
we then separately solve a dirichlet problem for laplace s equation to synthesize a safe textit guidance field that encodes variable levels of caution around obstacles -- by enforcing a tunable flux boundary condition
first hclfuse investigates the quantification theory of information mapping in unsupervised
fusion
image fusion
first hclfuse investigates the quantification theory of information mapping in unsupervised fusion networks which leads to the design of a multi-scale mask-regulated variational bottleneck encoder
despite their strong generative capabilities they face critical limitations in image editing tasks inaccurate inversion processes for mapping real images back into the latent space and gradient entanglement issues during
editing
image generation
despite their strong generative capabilities they face critical limitations in image editing tasks inaccurate inversion processes for mapping real images back into the latent space and gradient entanglement issues during editing often result in outputs that do not faithfully reflect the target prompt
simulations showed that grassmannian signaling provides competitive bit error rates ber at low signal-to-noise ratio snr regimes with low probability of detection at the unintended receiver compared to coherent schemes that use qpsk or qam modulation formats and need pilots to perform
channel
channel estimation
simulations showed that grassmannian signaling provides competitive bit error rates ber at low signal-to-noise ratio snr regimes with low probability of detection at the unintended receiver compared to coherent schemes that use qpsk or qam modulation formats and need pilots to perform channel estimation
second to guarantee pose-level safety pose correction based on penetration direction and line search is applied ensuring that each pose in the
trajectory
obstacle avoidance
second to guarantee pose-level safety pose correction based on penetration direction and line search is applied ensuring that each pose in the trajectory is collision-free and maximally clear from obstacles
in the human brain the allowed patterns of activity are constrained by the
correlations
functional connectivity
in the human brain the allowed patterns of activity are constrained by the correlations between brain regions
generative ai genai is rapidly transforming software engineering se practices influencing how se processes are executed as well as how
software
generative ai
generative ai genai is rapidly transforming software engineering se practices influencing how se processes are executed as well as how software systems are developed operated and evolved
results show that vibration reduces the crystallization onset temperature indicating enhanced
atomic
molecular dynamics
results show that vibration reduces the crystallization onset temperature indicating enhanced atomic mobility and nucleation kinetics
generating hand grasps with language instructions is a widely studied topic that benefits from
embodied
vision-language-action vla
generating hand grasps with language instructions is a widely studied topic that benefits from embodied ai and vr ar applications
492 we do find evidence of interacting molecular gas traced by oh providing further support for
oh
molecular gas
492 we do find evidence of interacting molecular gas traced by oh providing further support for oh s ability to trace hii region-molecular cloud interactions
photonic band-gap crystals radically modulate the rldos thereby controlling
spontaneous
spontaneous emission
photonic band-gap crystals radically modulate the rldos thereby controlling spontaneous emission
egoexo-con exploring view-invariant video
temporal
video generation
egoexo-con exploring view-invariant video temporal understanding
this work introduces a dynamic context-aware scene reasoning framework that leverages
vision-language
vision-language models vlms
this work introduces a dynamic context-aware scene reasoning framework that leverages vision-language alignment to address zero-shot real-world scenarios
we also explore applications of coherence to
graph
bipartite graphs
we also explore applications of coherence to graph problems
measurement technology employing optical interference phenomena such as a fringe pattern or
frequency
optical interference
measurement technology employing optical interference phenomena such as a fringe pattern or frequency shift has been evolving for more than a century
for fair and rigorous evaluation we curate goat-core a high-quality core split distilled from goat-bench tailored to multi-modal open-vocabulary multi-goal
visual
visual navigation
for fair and rigorous evaluation we curate goat-core a high-quality core split distilled from goat-bench tailored to multi-modal open-vocabulary multi-goal visual navigation
numerical analysis is provided to support the
optimality
numerical experiments
numerical analysis is provided to support the optimality result
most studies assume static power systems focus only on operational
emissions
optimal power flow
most studies assume static power systems focus only on operational emissions and overlook co-optimization
the state of brain emulation report 2025 provides a comprehensive reassessment of the field s progress since sandberg and bostrom s 2008 whole brain
emulation
electroencephalography eeg
the state of brain emulation report 2025 provides a comprehensive reassessment of the field s progress since sandberg and bostrom s 2008 whole brain emulation roadmap
we provide an inferential framework to assess variable importance for heterogeneous
treatment
treatment effect
we provide an inferential framework to assess variable importance for heterogeneous treatment effects
as a result of this compact structure the luminosity-radius relations for core and effective radii of
hii
hii regions
as a result of this compact structure the luminosity-radius relations for core and effective radii of hii regions depend sensitively on the adopted methodology
a prevailing theory in cognitive neuroscience proposes that the human brain operates through hierarchical predictive processing which underlies flexible
cognition
brain decoding
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
as llms occupy an increasingly important role in society they are more and more confronted with questions that require them not only to draw on their general
knowledge
llm inference
as llms occupy an increasingly important role in society they are more and more confronted with questions that require them not only to draw on their general knowledge but also to align with certain human value systems
experiments on synthetic and real-world datasets demonstrate that our method effectively mitigates spurious correlation issues and yields more robust
reward
reward models
experiments on synthetic and real-world datasets demonstrate that our method effectively mitigates spurious correlation issues and yields more robust reward models
classical newton-type multilevel methods mitigate this but like
gradient
gradient descent
classical newton-type multilevel methods mitigate this but like gradient descent achieve only linear convergence near the minimizer
user misconceptions of llm-based conversational
programming
large language models llms
user misconceptions of llm-based conversational programming assistants
out of the many deep reinforcement learning approaches for
autonomous
autonomous driving
out of the many deep reinforcement learning approaches for autonomous driving only few make use of the options or skills framework
our novel approach expands the capabilities of the
robot
obstacle avoidance
our novel approach expands the capabilities of the robot s inverse kinematics solver empowering it to acquire a sequential repertoire of actions using tools of varying lengths
this theory further promotes nonlinear transport as a probe of geometric effects and
phase
molecular dynamics
this theory further promotes nonlinear transport as a probe of geometric effects and phase transitions in quantum materials
unlike existing gaussian process-based approaches our method constructs an approximate
posterior
approximate posterior
unlike existing gaussian process-based approaches our method constructs an approximate posterior distribution using samples drawn from a gaussian process model fitted to the observed data which does not require any structural assumption about the underlying pde
in this work we conduct a systematic study of representation retention during
vla
vision-language-action vla
in this work we conduct a systematic study of representation retention during vla fine-tuning showing that naive action fine-tuning leads to degradation of visual representations
recalling previously experienced movements is essential for a range of activities including sports music and rehabilitation yet little is known about the accuracy and decay of proprioceptive
working
brain activity
recalling previously experienced movements is essential for a range of activities including sports music and rehabilitation yet little is known about the accuracy and decay of proprioceptive working memory
the emergence of spite driven by prejudice is also found to persist when one considers long-term
evolutionary
evolutionary game
the emergence of spite driven by prejudice is also found to persist when one considers long-term evolutionary dynamics in the mutation-selection dominated regime
these findings suggest that cd-based acs some of which show the multi-step magnetic transitions provide a new platform for investigating exotic
magnetic
magnetic properties
these findings suggest that cd-based acs some of which show the multi-step magnetic transitions provide a new platform for investigating exotic magnetic properties that cannot be understood within the conventional framework of hybridization at ef
we argue that a central cause is representation quality exteroceptive
inputs
representation learning
we argue that a central cause is representation quality exteroceptive inputs e
we demonstrate that for the majority of rjgs the
rejuvenating
molecular gas
we demonstrate that for the majority of rjgs the rejuvenating gas is originally in the galaxy rather than accreted gas
these paradoxes highlight the tension between quantum
theory
quantum correlations
these paradoxes highlight the tension between quantum theory and our intuitions about reality being observer-independent
adaptive context length optimization with
low-frequency
context length
adaptive context length optimization with low-frequency truncation for multi-agent reinforcement learning
reinforcement learning rl fine-tuning of large
language
large language
reinforcement learning rl fine-tuning of large language models llms often suffers from instability due to the numerical mismatch between the training and inference policies
while pruning and low-rank approximation have each demonstrated promising performance individually their synergy for llms
remains
large language models llms
while pruning and low-rank approximation have each demonstrated promising performance individually their synergy for llms remains underexplored
extensions to longitudinal data dynamic treatment
regimes
dynamic treatment
extensions to longitudinal data dynamic treatment regimes and multiplicative instrumental variables are further developed
we present the first known pivot gray code for spanning trees of complete
graphs
tree embedding
we present the first known pivot gray code for spanning trees of complete graphs listing all spanning trees such that consecutive trees differ by pivoting a single edge around a vertex
extracting spectral diffusion in two-dimensional coherent
spectra
spectral range
extracting spectral diffusion in two-dimensional coherent spectra via the projection slice theorem
in this study we used a data-driven network approach to examine whether resting-state
eeg
electroencephalography eeg
in this study we used a data-driven network approach to examine whether resting-state eeg connectivity patterns differentiate individuals according to their creative abilities
in this environment second-order minimizing methods such as the conjugate
gradient
gradient descent
in this environment second-order minimizing methods such as the conjugate gradient cg give a guaranteed superlinear convergence
along with the natural language embeddings the representations are trained by an haoi manipulation language model to align the grasping process with its
language
vision-language models
along with the natural language embeddings the representations are trained by an haoi manipulation language model to align the grasping process with its language description in a shared representation space
while deep learning dominates recent mtl research support vector machines svms and twin
svms
deep learning
while deep learning dominates recent mtl research support vector machines svms and twin svms twsvms remain relevant due to their interpretability theoretical rigor and effectiveness with small datasets
to tackle this challenge we propose braincognizer a novel brain decoding model inspired by human visual cognition which explores multi-level
semantics
encoding models
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
we verify our proposal through exact calculations on
ground
ground state
we verify our proposal through exact calculations on ground states constructed using real space constructions
we outline a multi-phase research methodology focused on preparing re datasets fine-tuning ai models and designing collaborative
human-ai
ai literacy
we outline a multi-phase research methodology focused on preparing re datasets fine-tuning ai models and designing collaborative human-ai workflows
we study a bi-virus susceptible-infected-susceptible sis epidemic model in which individuals are either susceptible or
infected
infectious individuals
we study a bi-virus susceptible-infected-susceptible sis epidemic model in which individuals are either susceptible or infected with one of two virus strains and consider mutation-driven transitions between strains
specifically we show that in the presence of non-hermitian degeneracies known as exceptional
points
exceptional points
specifically we show that in the presence of non-hermitian degeneracies known as exceptional points the expected mode hierarchy can be dramatically altered
we apply our method to variational quantum
algorithm
open quantum
we apply our method to variational quantum algorithm vqa ansatz design for molecular ground state estimation max-cut and image classification key challenges in near-term quantum computing
a case study in south yorkshire uk illustrates how regional forecasts can be translated into local design responses connecting quantitative modelling with 3d
spatial
urban systems
a case study in south yorkshire uk illustrates how regional forecasts can be translated into local design responses connecting quantitative modelling with 3d spatial planning
the sga method outperforms the gradient method by including
second-order
accelerated gradient
the sga method outperforms the gradient method by including second-order mixed derivatives computed at each iterate which requires considerably larger computational effort
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
optimal control
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
large language models llms face significant
inference
llm inference
large language models llms face significant inference latency challenges stemming from their autoregressive design and large size
in particular we show that narrower degree distributions contain longer shortest
loops
complex networks
in particular we show that narrower degree distributions contain longer shortest loops as a universal property in a wide class of random networks
26 10 1 m _ sun yr -1 kpc -2 our results indicate a rapid building of inner stellar mass and
bulge
bulge stars
26 10 1 m _ sun yr -1 kpc -2 our results indicate a rapid building of inner stellar mass and bulge assembly within these young systems
this alignment anchors the reasoning of a judge large
language
natural language
this alignment anchors the reasoning of a judge large language model llm in structured information and helps reduce the burden of regulatory interpretation and event parsing enabling a focus on the core reasoning step
this work introduces steervlm a lightweight steering module designed to guide
vision-language
vision-language models
this work introduces steervlm a lightweight steering module designed to guide vision-language models vlms towards outputs that better adhere to desired instructions
practical hybrid decoding scheme for parity-encoded
spin
spin readout
practical hybrid decoding scheme for parity-encoded spin systems
on the positive side we present polynomial-time approximation schemes ptass for k -cycle and k -path graphs improving over the best existing approximation factors of 2 for k -cycle
graphs
regular graphs
on the positive side we present polynomial-time approximation schemes ptass for k -cycle and k -path graphs improving over the best existing approximation factors of 2 for k -cycle graphs and an approximation factor of 4 for k -path graphs
we further show that in this case the truncated random return follows a positively weighted non-central chi-square distribution if the random disturbances admits gaussian and its cumulative distribution function is log-concave if the probability density function of the
random
truncated random return
we further show that in this case the truncated random return follows a positively weighted non-central chi-square distribution if the random disturbances admits gaussian and its cumulative distribution function is log-concave if the probability density function of the random disturbances is log-concave
but this method often fails to converge due to the system
s
first-order methods
but this method often fails to converge due to the system s high dimensionality and nonlinearity
when evaluated against the clinical standard used to train the ai both ai and
human-ai
human-machine teaming
when evaluated against the clinical standard used to train the ai both ai and human-ai teams significantly outperform unaided experts with collaboration also reducing inter-rater variability
the informative sampling method was further divided into sequential method which ranked pens by descending viral load probability and cluster and random method which selected pens at random from high and low
viral
viral replication
the informative sampling method was further divided into sequential method which ranked pens by descending viral load probability and cluster and random method which selected pens at random from high and low viral load clusters
however in recent years with the rise of generative ai especially large
language
language models
however in recent years with the rise of generative ai especially large language models llm and particularly with the widespread popularity of the chatgpt model that concern became practical
one of the main challenges in isac lies in resource allocation which becomes computationally demanding in dynamic environments requiring
real-time
communication isac
one of the main challenges in isac lies in resource allocation which becomes computationally demanding in dynamic environments requiring real-time adaptation
exploring complementarity and explainability in cnns for
periocular
deep network
exploring complementarity and explainability in cnns for periocular verification across acquisition distances
importantly we only train and evaluate on pairs of frames where one contains the query point effectively removing any
temporal
temporal understanding
importantly we only train and evaluate on pairs of frames where one contains the query point effectively removing any temporal context
our framework thus provides a unified foundation for analyzing and improving the efficiency of fundamental string-processing problems through the lens of
prefix
suffix array
our framework thus provides a unified foundation for analyzing and improving the efficiency of fundamental string-processing problems through the lens of prefix queries
we investigate a minimal algebraic setup that allows us to study a notion of differentiability suitable for newton-type methods called
newton
zeroth-order methods
we investigate a minimal algebraic setup that allows us to study a notion of differentiability suitable for newton-type methods called newton differentiability
under sufficient differentiability the new method achieves an asymptotic average
convergence
zeroth-order methods
under sufficient differentiability the new method achieves an asymptotic average convergence order of at least the square root of 5 per iteration surpassing the quadratic order of the original algorithm
x-ray and variability selected agn have higher average star formation
rates
star-forming region
x-ray and variability selected agn have higher average star formation rates than those selected with optical narrow line spectroscopic diagrams
quantum computation with d -level quantum systems also known as qudits benefits from the possibility to use a richer
computational
quantum dot
quantum computation with d -level quantum systems also known as qudits benefits from the possibility to use a richer computational space compared to qubits
we further extended jne with sample-specificity jne-ss revealing stimulus-selective nonlinear response
patterns
brain regions
we further extended jne with sample-specificity jne-ss revealing stimulus-selective nonlinear response patterns in functionally specialized brain regions
this dramatic shift underscores the enhanced robustness of hub-mediated spin correlations under competitive coupling leading to asymmetric order parameters between layers and novel
phase
phase transition
this dramatic shift underscores the enhanced robustness of hub-mediated spin correlations under competitive coupling leading to asymmetric order parameters between layers and novel phase transition phenomena
large language models are influencing the
education
large language
large language models are influencing the education landscape with students relying on them in their learning process
we conclude that ly alpha radiation pressure severely limits a possible extremely efficient feedback-free phase of star formation in
dense
star clusters
we conclude that ly alpha radiation pressure severely limits a possible extremely efficient feedback-free phase of star formation in dense metal-poor clouds
allowing the order of quantum operations to exist in superposition is known to
open
open quantum
allowing the order of quantum operations to exist in superposition is known to open new routes for thermodynamic tasks
urban form and function shape mobility more profoundly than structure even though structure often exhibits higher correlations as observed in
cities
urban systems
urban form and function shape mobility more profoundly than structure even though structure often exhibits higher correlations as observed in cities such as singapore new delhi london chicago and moscow
in linear time-invariant systems the sensitivity function to
disturbances
bounded disturbances
in linear time-invariant systems the sensitivity function to disturbances is designed under a sensitivity tradeoff known as the waterbed effect
based on our characterization we design a policy optimization algorithm that uses machine learning to predict counterfactual outcomes and then plugs in these predictions to estimate the pareto frontier then the decision-maker can select the policy that optimizes their desired tradeoff after which
policy
policy optimization
based on our characterization we design a policy optimization algorithm that uses machine learning to predict counterfactual outcomes and then plugs in these predictions to estimate the pareto frontier then the decision-maker can select the policy that optimizes their desired tradeoff after which policy evaluation can ...
this research develops a composite liveability index for greater london based on metrics related to the proximity density and diversity of pois along with population density and investigates how neighbourhood liveability relates to active
travel
human mobility
this research develops a composite liveability index for greater london based on metrics related to the proximity density and diversity of pois along with population density and investigates how neighbourhood liveability relates to active travel behaviour
to overcome the potential non-smoothness of the hyper-objective and the computational challenges associated with the hessian matrix we utilize penalty and augmented lagrangian methods to reformulate the original
problem
bilevel optimization
to overcome the potential non-smoothness of the hyper-objective and the computational challenges associated with the hessian matrix we utilize penalty and augmented lagrangian methods to reformulate the original problem as a single-level one
in difference-in-differences did settings with
categorical
categorical outcomes
in difference-in-differences did settings with categorical outcomes such as voting occupation or major choices treatments often affect both total counts e
twin-field quantum key distribution tf-qkd has emerged as a potential protocol for long distance secure communication overcoming the rate-distance limitations of conventional quantum
key
fault-tolerant quantum
twin-field quantum key distribution tf-qkd has emerged as a potential protocol for long distance secure communication overcoming the rate-distance limitations of conventional quantum key distribution without requiring trusted repeaters
the near-ghz soliton repetition frequency
exhibits
ghz ghz
the near-ghz soliton repetition frequency exhibits a single-sideband phase noise of -137 dbc hz at a 100 khz offset surpassing state-of-the-art microwave generators
we hope our work inspires further exploration into maximizing the
potential
training data
we hope our work inspires further exploration into maximizing the potential of data utilization in large-scale model training
reality distortion room a study of user locomotion responses to spatial augmented
reality
physical virtual
reality distortion room a study of user locomotion responses to spatial augmented reality effects