prompt
stringlengths
41
511
target
stringlengths
1
25
keyword
stringclasses
697 values
full_sentence
stringlengths
48
1.25k
across 125 participants we found that interactive interfaces
significantly
task performance
across 125 participants we found that interactive interfaces significantly improved performance
the growing integration of artificial intelligence ai into human cognition raises a fundamental question does ai merely
improve
human-ai interaction
the growing integration of artificial intelligence ai into human cognition raises a fundamental question does ai merely improve efficiency or does it alter how we think
this study not only enriches the theoretical framework of evolutionary game theory but also provides a foundation for the management of
ecological
ecological interactions
this study not only enriches the theoretical framework of evolutionary game theory but also provides a foundation for the management of ecological systems and the design of cooperative mechanisms in society
recent alignment techniques such as reinforcement learning from human feedback have been widely adopted to align large
language
large language
recent alignment techniques such as reinforcement learning from human feedback have been widely adopted to align large language models with human preferences by learning and leveraging reward models
to characterize and measure these effects we probe vla s hidden representations and analyze attention maps further we design a set of targeted tasks and methods that contrast
vla
vla models
to characterize and measure these effects we probe vla s hidden representations and analyze attention maps further we design a set of targeted tasks and methods that contrast vla models with their counterpart vlms isolating changes in vl capabilities induced by action fine-tuning
this approach inherently handles closed chains in the same manner as tree-like structures eliminating the need for explicit
constraint
soft constraints
this approach inherently handles closed chains in the same manner as tree-like structures eliminating the need for explicit constraint force calculations or formulations based on differential-algebraic equations
leveraging this dataset we fine-tune llms to generate code
review
code review
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
there are two major approaches in policy learning the empirical
welfare
reinforcement learning
there are two major approaches in policy learning the empirical welfare maximization ewm approach and the plug-in approach
airborne assessment uncovers socioeconomic stratification of
urban
urban systems
airborne assessment uncovers socioeconomic stratification of urban nature in england
yet it remains unclear which correlations -- and how many -- are needed to predict
large-scale
functional connectivity
yet it remains unclear which correlations -- and how many -- are needed to predict large-scale neural activity
we investigate the impact of synthetic data generation and lightweight fine-tuning techniques on the ability of large language models
llms
large language models llms
we investigate the impact of synthetic data generation and lightweight fine-tuning techniques on the ability of large language models llms to recognize fallacious arguments using the missci dataset and framework
two spatial light modulators enable precise spectral shaping of both the pump and the local oscillator in amplitude and phase allowing investigation of the
spectral
coherent control
two spatial light modulators enable precise spectral shaping of both the pump and the local oscillator in amplitude and phase allowing investigation of the spectral properties of the generated states
gave a randomized tilde o m time algorithm for computing a phi
phi
-time algorithm
gave a randomized tilde o m time algorithm for computing a phi phi log 2 n -expander decomposition
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
collision avoidance
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
through fourier transform infrared spectroscopy measurement and full-wave photonic simulations we identified a
range
optical properties
through fourier transform infrared spectroscopy measurement and full-wave photonic simulations we identified a range of optical excitations in the rbs including three sphps two hyperbolic volume phonon polaritons hvphps and one epsilon-near-zero enz mode
this body of theory can accommodate a range of
social
social interactions
this body of theory can accommodate a range of social dilemmas or games as well as real-world complexities such as spatial structure or behaviors conditioned on reputations
the results confirm an average improvement and efficiency of the proposed method
compared
existing methods
the results confirm an average improvement and efficiency of the proposed method compared to the state-of-the-art approaches
this thesis delves into the world of non-invasive electrophysiological brain signals like electroencephalography
eeg
electroencephalography eeg
this thesis delves into the world of non-invasive electrophysiological brain signals like electroencephalography eeg and magnetoencephalography meg focusing on modelling and decoding such data
this paper presents a hierarchical path-planning and control framework that combines a high-level deep q-network dqn for discrete sub-goal selection with a low-level twin delayed deep deterministic
policy
motion planning
this paper presents a hierarchical path-planning and control framework that combines a high-level deep q-network dqn for discrete sub-goal selection with a low-level twin delayed deep deterministic policy gradient td3 controller for continuous actuation
drivers of variation in the optimal spatial
structure
spatial structure
drivers of variation in the optimal spatial structure of collective information gatherers
considering toy models and the mnist dataset we numerically illustrate how after learning the two
fisher
fisher information
considering toy models and the mnist dataset we numerically illustrate how after learning the two fisher information matrices match and essentially align with the category boundaries
the quantum dynamics of the two spins reveal entanglement resonances and kinks which can be identified from the energy spectrum when weak transverse field
strengths
quantum coherence
the quantum dynamics of the two spins reveal entanglement resonances and kinks which can be identified from the energy spectrum when weak transverse field strengths are considered
for training purposes we prepared a set of 1 730 annotated
images
computer vision
for training purposes we prepared a set of 1 730 annotated images that were captured under various traffic conditions
this motivates the development of a rigorous and readily applicable framework for studying properties of large
multiplex
multiplex networks
this motivates the development of a rigorous and readily applicable framework for studying properties of large multiplex networks
graph-theoretical mapping of resting-state
eeg
human brain
graph-theoretical mapping of resting-state eeg reveals neural signatures of creativity
3 we show that several fundamental 2d queries such as the 2d longest common extension rectangle sum and equality cannot be supported efficiently under hardness assumptions for rank and symbol occurrence
queries
compressed indexing
3 we show that several fundamental 2d queries such as the 2d longest common extension rectangle sum and equality cannot be supported efficiently under hardness assumptions for rank and symbol occurrence queries on 1d grammar-compressed strings
we demonstrate the inherent properties associated with the
mdp
markov decision
we demonstrate the inherent properties associated with the mdp formulation and the related optimal policy
quantitative electron magnetic circular dichroism emcd in transmission electron
microscopy
atomic force microscopy
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
these results confirm the effectiveness efficiency and robustness of our approach for low-resource
domain
llm agents
these results confirm the effectiveness efficiency and robustness of our approach for low-resource domain adaptation of llms
we apply this approach to three leading models i the general treatment model under unconfoundedness ii the negative control model and iii the long-term
causal
causal effects
we apply this approach to three leading models i the general treatment model under unconfoundedness ii the negative control model and iii the long-term causal inference model under unobserved confounding
the nct reframes ai evaluation from performance to persistence outlining conceptual requirements for future benchmarks and architectural designs that could sustain long-term identity and goal coherence in
generative
ai literacy
the nct reframes ai evaluation from performance to persistence outlining conceptual requirements for future benchmarks and architectural designs that could sustain long-term identity and goal coherence in generative models
we summarize published strategies for selecting
genetic
population genetics
we summarize published strategies for selecting genetic instruments and performing analyses when working with limited ancestry-specific data discussing the assumptions needed in each case for incorporating external data from different ancestry populations
we demonstrate this using bimanual manipulators humanoids and multi-fingered hands in optimal
control
optimal control
we demonstrate this using bimanual manipulators humanoids and multi-fingered hands in optimal control experiments for reaching desired geometric primitives and in teleoperation experiments using differential kinematics control
5 exhibits strong native multimodal capabilities including long-horizon vision-language generation any-to-image x2i
generation
image generation
5 exhibits strong native multimodal capabilities including long-horizon vision-language generation any-to-image x2i generation and complex text-rich image generation
tracing the evolution of brightest galaxies and
diffuse
galaxy cgm
tracing the evolution of brightest galaxies and diffuse light in galaxy groups
for reflection-based readout we leverage the spin-dependent cavity reflection contrast to generate the qubit
readout
qubit readout
for reflection-based readout we leverage the spin-dependent cavity reflection contrast to generate the qubit readout signal
large language models llms show strong potential to support creative
tasks
large language models llms
large language models llms show strong potential to support creative tasks but the role of the interface design is poorly understood
specifically the proposed evidential fusion paradigm transforms the features from different backbones into supporting evidences via a
set
image fusion
specifically the proposed evidential fusion paradigm transforms the features from different backbones into supporting evidences via a set of deep evidential networks
we examine the quantum dynamics of a large
spin
spin readout
we examine the quantum dynamics of a large spin in the presence of static and rotating magnetic fields
despite multiple efforts to understand the problem in these simple graph families the computational complexity of the problem had remained unsettled and our hardness results answer open questions by bhabak and harutyunyan and harutyunyan and hovhannisyan concerning the problem s
complexity
time complexity
despite multiple efforts to understand the problem in these simple graph families the computational complexity of the problem had remained unsettled and our hardness results answer open questions by bhabak and harutyunyan and harutyunyan and hovhannisyan concerning the problem s complexity in k -cycle and k -path graph...
those models give structure to expectations around walking
behavior
emergent behaviors
those models give structure to expectations around walking behavior of groups from pedestrian walking independently to the emergence of a collective intention where group members tended to maintain bounded distance between each other
we present a generalization of the so-called fractional stirap f-stirap demonstrating precise control over the mixing ratio of
quantum
quantum dot
we present a generalization of the so-called fractional stirap f-stirap demonstrating precise control over the mixing ratio of quantum states in the wave packet
a convexity-dependent two-phase training algorithm for deep
neural
neural network
a convexity-dependent two-phase training algorithm for deep neural networks
then we report how 4d-stem data enable to reconstruct high-resolution images with
electron
electron microscopy
then we report how 4d-stem data enable to reconstruct high-resolution images with electron ptychography
existing reinforcement learning from verifiable rewards rlvr methods such as group relative
policy
policy learning
existing reinforcement learning from verifiable rewards rlvr methods such as group relative policy optimization grpo have achieved remarkable progress in improving the reasoning capabilities of large reasoning models lrms
simulated galaxies with higher halo masses higher median cgm gas density and higher star formation
rates
stellar mass
simulated galaxies with higher halo masses higher median cgm gas density and higher star formation rates produce brighter and more widespread o vi emission in their cgm
meanwhile the resource allocation problem is solved using a successive
convex
power allocation
meanwhile the resource allocation problem is solved using a successive convex approximation sca -based algorithm
the study is done in the ambit of some suitably chosen inequalities in i linear
networks
quantum networks
the study is done in the ambit of some suitably chosen inequalities in i linear networks and ii star-shaped networks
immersive applications call for synthesizing
spatiotemporal
image fusion
immersive applications call for synthesizing spatiotemporal 4d content from casual videos without costly 3d supervision
first we find that combining english and multilingual data does not necessarily degrade the in-language performance of either group provided that
languages
multilingual data
first we find that combining english and multilingual data does not necessarily degrade the in-language performance of either group provided that languages have a sufficient number of tokens included in the pretraining corpus
we apply our method to variational quantum
algorithm
quantum key distribution
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
5 exhibits strong native multimodal capabilities including long-horizon
vision-language
vision-language models vlms
5 exhibits strong native multimodal capabilities including long-horizon vision-language generation any-to-image x2i generation and complex text-rich image generation
we investigate the impact of synthetic data generation and lightweight
fine-tuning
large language
we investigate the impact of synthetic data generation and lightweight fine-tuning techniques on the ability of large language models llms to recognize fallacious arguments using the missci dataset and framework
aci advances the detection of instantaneous causal relationships and the intermittent reversal of
causal
causal inference
aci advances the detection of instantaneous causal relationships and the intermittent reversal of causal roles over time
ais have made rapid progress on research-oriented benchmarks of knowledge and
reasoning
reasoning capabilities
ais have made rapid progress on research-oriented benchmarks of knowledge and reasoning but it remains unclear how these gains translate into economic value and automation
contextual inference facilitates the creation learning and reuse of low-rank rnn components as new tasks are introduced sequentially enabling
continual
continual learning
contextual inference facilitates the creation learning and reuse of low-rank rnn components as new tasks are introduced sequentially enabling continual learning without catastrophic forgetting
this paper studies an extended rd design where
assignment
covariate balancing
this paper studies an extended rd design where assignment rules simultaneously involve two or more continuous covariates
entanglement superactivation in multiphoton
distillation
multipartite entanglement
entanglement superactivation in multiphoton distillation networks
we study the problem of computing a minimum s -- t cut in an unweighted undirected
graph
bipartite graphs
we study the problem of computing a minimum s -- t cut in an unweighted undirected graph via emph cut queries
in summary this work advances the modeling of epidemics to a more local scope offering a more expressive tool for epidemiological research and public
health
public health
in summary this work advances the modeling of epidemics to a more local scope offering a more expressive tool for epidemiological research and public health planning
finally we prove that even if an estimator had access to the test distribution mu the convergence rate of its expected risk over mu could not be faster than that of our pretrained transformers thereby providing a more appropriate optimality guarantee than
minimax
minimax optimal
finally we prove that even if an estimator had access to the test distribution mu the convergence rate of its expected risk over mu could not be faster than that of our pretrained transformers thereby providing a more appropriate optimality guarantee than minimax lower bounds
we confirm that galaxies with star formation efficiencies lower than the milky
way
dwarf galaxies
we confirm that galaxies with star formation efficiencies lower than the milky way have high probably indicating a stronger efficiency of the delayed sources of r-process at low metallicities
since no dataset annotated with both pts and drs exists we took the semeval 2023 task 3 dataset labelled with 19 pts as a starting point and developed llm-based classifiers to label each instance of the
dataset
training data
since no dataset annotated with both pts and drs exists we took the semeval 2023 task 3 dataset labelled with 19 pts as a starting point and developed llm-based classifiers to label each instance of the dataset with one of the 22 pdtb 3
experimenting with llama-3 and qwen-3 models of different sizes and popular supervised fine-tuning sft and preference optimization datasets and algorithms we find that the sft phase generally establishes a model s values and subsequent
preference
preference optimization
experimenting with llama-3 and qwen-3 models of different sizes and popular supervised fine-tuning sft and preference optimization datasets and algorithms we find that the sft phase generally establishes a model s values and subsequent preference optimization rarely re-aligns these values
we present a merge-free algorithm for multi-way
co-ranking
compressed indexing
we present a merge-free algorithm for multi-way co-ranking the problem of computing cut indices i_1 dots i_m that partition each of the m sorted sequences such that all prefix segments together contain exactly k elements
moreover it gives access at the same time to the spatial distribution within the monolayer of the individual spin-segregated states responsible for the hidden
spin
spin-momentum locking
moreover it gives access at the same time to the spatial distribution within the monolayer of the individual spin-segregated states responsible for the hidden spin textures not provided by other techniques
through these principles our method achieves image reconstructions from
fmri
fmri data
through these principles our method achieves image reconstructions from fmri that faithfully reconstruct the seen images and surpass current sota approaches both visually and by standard objective metrics
the distinct structural attributes coupled with suitable electronic band structure promotes the electron
transport
transport properties
the distinct structural attributes coupled with suitable electronic band structure promotes the electron transport properties
our numerical results explain how localized interventions affect the spread of the
disease
disease transmission
our numerical results explain how localized interventions affect the spread of the disease across cities
stakeholders share common perceptions on biodiversity impacts despite geographic disparity but they differentiate between
climate
environmental change
stakeholders share common perceptions on biodiversity impacts despite geographic disparity but they differentiate between climate and land use impacts
we review studies that have examined nccs at the level of single neurons and populations of neurons and evaluate their implications on the debates between
cognitive
surrogate brain
we review studies that have examined nccs at the level of single neurons and populations of neurons and evaluate their implications on the debates between cognitive and sensory theories of consciousness
reliable estimation of network-wide traffic states is essential for urban
traffic
traffic states
reliable estimation of network-wide traffic states is essential for urban traffic management
we believe this survey establishes a solid foundation and offers insights into the growing field of multimodal
spatial
spatial reasoning
we believe this survey establishes a solid foundation and offers insights into the growing field of multimodal spatial reasoning
in this paper we consider a full-duplex fd integrated sensing and communication
isac
communication isac
in this paper we consider a full-duplex fd integrated sensing and communication isac system in which the base station bs performs downlink and uplink communications with multiple users while simultaneously sensing multiple targets
here we present a continuous-wave synthetic wavelength interferometry technique that employs digitally tunable electro-optic
frequency
optical interference
here we present a continuous-wave synthetic wavelength interferometry technique that employs digitally tunable electro-optic frequency combs
surface phonon polaritons sphps are promising candidates for enhanced light--matter
interactions
light-matter interactions
surface phonon polaritons sphps are promising candidates for enhanced light--matter interactions due to their efficient and low-loss light confinement features
we investigate a minimal algebraic setup that allows us to study a notion of differentiability suitable for newton-type methods called
newton
first-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
consistent with this the fraction of barred galaxies among agn
hosts
host galaxy
consistent with this the fraction of barred galaxies among agn hosts decreases with increasing l_ rm disc
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
convergence guarantees
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
however its potential is constrained by the difficulty of acquiring accurate user-to-ris channel state information csi due to the cascaded
channel
channel state information
however its potential is constrained by the difficulty of acquiring accurate user-to-ris channel state information csi due to the cascaded channel structure and the high pilot overhead of non-parametric methods
jogs joint optimization of pose estimation and 3d
gaussian
point tracking
jogs joint optimization of pose estimation and 3d gaussian splatting
the basic reproduction number mathcal r_0 calculated from models that assume homogeneous mixing or single-layer contact structures have limited
applicability
reproduction number
the basic reproduction number mathcal r_0 calculated from models that assume homogeneous mixing or single-layer contact structures have limited applicability to complex social systems
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
artificial intelligence
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...
the proliferation of collaborative robots across diverse
tasks
multi-robot collaboration
the proliferation of collaborative robots across diverse tasks and embodiments presents a central challenge achieving lifelong adaptability scalable coordination and robust scheduling in multi-agent systems
we argue that such abstraction leads to oversimplification of reasoning methodologies from nlp ml and results in a distortion of
llms
language models
we argue that such abstraction leads to oversimplification of reasoning methodologies from nlp ml and results in a distortion of llms empirically studied capabilities and un known limitations
3 we show this dependency is almost optimal to get a poly n log
lambda
polynomial time
3 we show this dependency is almost optimal to get a poly n log lambda -time complexity
our estimator also admits a semiparametric sensitivity analysis for violations of the key identifying assumption as well as asymptotically valid confidence intervals for
local
nonparametric identification
our estimator also admits a semiparametric sensitivity analysis for violations of the key identifying assumption as well as asymptotically valid confidence intervals for local unit-level estimates under additional assumptions
to make inverse optimal issf controllers robust to gain variation we propose a
gain
inverse optimal
to make inverse optimal issf controllers robust to gain variation we propose a gain margin improvement approach at the expense of an increased control effort
while pulsed laser ablation in liquids offers unparalleled advantages in terms of nanoparticle purity and material versatility enhancing the size control and productivity require modifications of the standard pulsed
laser
pulsed laser
while pulsed laser ablation in liquids offers unparalleled advantages in terms of nanoparticle purity and material versatility enhancing the size control and productivity require modifications of the standard pulsed laser ablation in liquid technique such as the incorporation of beam shaping techniques
in this work we extend our previous work on the extrapolation of single-time observables to demonstrate an efficient scheme for
computing
quantum correlations
in this work we extend our previous work on the extrapolation of single-time observables to demonstrate an efficient scheme for computing two-time impurity correlation functions by combining the non-markovian quantum mpemba effect nmqmpe with a dynamical map-based framework for open quantum systems
this weighted resampling strategy ensures the diffusion-generated samples are
distributed
diffusion models
this weighted resampling strategy ensures the diffusion-generated samples are distributed according to our desired multi-target boltzmann distribution
accurate estimation of motion information is crucial in diverse computational imaging and
computer
optical flow
accurate estimation of motion information is crucial in diverse computational imaging and computer vision applications
their structural configuration as well as on the presence and nature of
atomic
atomic force
their structural configuration as well as on the presence and nature of atomic defects
to address this we analyze one year of smartphone position data from over one million users in the tokyo metropolitan area to extract high-resolution
commuting
travel information
to address this we analyze one year of smartphone position data from over one million users in the tokyo metropolitan area to extract high-resolution commuting trajectories
furthermore for specific pairs of models and riesz representer estimation methods we can automatically obtain the covariate balancing property without explicitly solving the covariate
balancing
covariate balancing
furthermore for specific pairs of models and riesz representer estimation methods we can automatically obtain the covariate balancing property without explicitly solving the covariate balancing objective
across classification regression and generative benchmarks himae consistently outperforms state of the
art
consistently outperforms
across classification regression and generative benchmarks himae consistently outperforms state of the art foundation models that collapse scale while being orders of magnitude smaller
our novel approach expands the capabilities of the
robot
robotic manipulation
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
in this work we answer it positively by providing both computational and statistical
convergence
convergence rate
in this work we answer it positively by providing both computational and statistical convergence guarantees of sgd
interpolation in generative models allows for controlled
generation
generative ai
interpolation in generative models allows for controlled generation model inspection and more