Entities · Labs
alphaXiv
632 articles tagged with this entity.
-
InductWave: Inductive Multi-Hop Logical Query Answering on Knowledge Graphs
-
Ego-Human Motion Prediction with 3D-Aware LLM
-
Latency-Constrained DNN Architecture Learning for Edge Systems using Zerorized Batch Normalization
-
A knowledge-augmented dataset of high-risk driving scenarios with LLM annotations for autonomous driving
-
Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity
-
HPR-SAM: Hierarchical Probabilistic Representation Learning for Prompt-free SAM-based Medical Image Segmentation
-
SynthAVE: Scalable Synthetic Labeling for E-Commerce with LLM-Arena Validation
-
Simulstream: Open-Source Toolkit for Evaluation and Demonstration of Streaming Speech-to-Text Translation Systems
-
HIVE: Understanding Post-Hallucination Reasoning in Vision Language Models
-
A Multi-Analyst LLM Pipeline for Auditable Rule Discovery Across 68 Public Physiological Corpora
-
From Atomic Actions to Standard Operating Procedures: Iterative Tool Optimization for Self-Evolving LLM Agents
-
A Good Initialization is All You Need for Faithful Visual Attribution
-
STAGformer: A Spatio-temporal Agent Graph Transformer for Micro Mobility Demand Forecasting
-
LEMUR 2: Unlocking Neural Network Diversity for AI
-
SPEAR: A Simulator for Photorealistic Embodied AI Research
-
PALS: Percentile-Aware Layerwise Sparsity for LLM Pruning
-
UI2App: Benchmarking Visual Interaction Inference in Executable Web Application Generation
-
ArtisanCAD: An Industrial-Level CAD Agent with Expert-Grounded Knowledge Distillation
-
Why does Deep Learning Improve Visual SLAM?
-
Lean-Quantum: Toward AI-Assisted Formalization of Quantum Information
-
Umm... With Transformers? Insights from Filled Pause Use across Four Slavic Parliaments
-
From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations
-
CurateEvo: Data-Curation Evolving for Agentic Post-Training
-
DynaKRAG: A Unified Framework for Learnable Evidence Control in Multi-Hop Retrieval-Augmented Generation
-
PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents
-
CanvasAgent: Enabling Complex Image Creation and Editing via Visual Tool Orchestration
-
AlayaWorld: Long-Horizon and Playable Video World Generation
-
Parameter-Free Encoders Remain Viable for RDB Foundation Models
-
Controlling Tool Use with Heading-Specific Activation Steering
-
Prompt Coach: An Empirical Evaluation of an Agentic Tutor for Learning Prompt Engineering in Software Development
-
Is Domain Adaptation Always Helpful? A Frozen-Backbone Study of Cross-Domain Sentiment Transfer
-
Measuring the practice of shared-decision making (OPTION12): An Investigation into Open-sourced Smaller LLMs (OS-sLLMs) for Better Privacy and Sustainability
-
Omni-RRM: Advancing Omni Reward Modeling via Automatic Rubric-Grounded Preference Synthesis
-
Base Models Know How to Reason, Thinking Models Learn When
-
PluraMath: Extending Mathematical Reasoning Evaluation Beyond High-Resource Languages
-
Estimating Uncertainty from Reasoning: A Large-Scale Study of Multi- and Crosslingual MCQA Performance in LLMs
-
RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications
-
Separating Representation from Reconstruction Enables Scalable Text Encoders
-
Scalable Semantic Steering of Embedding Projections
-
Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective
-
rePIRL: Learn PRM with Inverse RL for LLM Reasoning
-
ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evolution
-
Conditional Diffusion Guided Knowledge Transfer for Multi-Domain Knowledge Graph Completion
-
Rethinking Scientific Discovery in an Agentic Era
-
CARD: Cross-component Audio Representation Distillation for Encoder-Free Audio Captioning
-
STELLA: Efficient Sensor-to-LLM Translation for On-Device Human Activity Recognition
-
MPSelectTune: Prompt-type Selection for Fine-tuning improves Concept Unlearning in LLMs
-
Is Agentic Code Review Helpful? Mining Developers' Feedback to CodeRabbit Reviews in the Wild
-
PAST-TIDE: Prototype-Anchored Statement Tuning with Topic-Invariant Normalization for Stance Detection
-
A non-invasive video-based method for individual identification of wildlife using gait dynamics
-
EVA-Client: A Unified Data Collection, Inference, and Deployment Framework for Embodied Policies on Real Robots
-
IndustryNav: Exploring Spatial Reasoning of Embodied Agents in Dynamic Industrial Navigation
-
Beyond Static Rules: Automated Discovery of Latent Vulnerabilities in Text-to-SQL
-
GeoSAM-Lite: A Lightweight Foundation Model for Onboard Remote Sensing Segmentation
-
ResearchStudio-Idea: An Evidence-Grounded Research-Ideation Skill Suite from ML Conference Outcomes
-
Not Every Sync Is Safe: Calibrated DiLoCo Scheduling for Shared AI Infrastructure
-
FuseMamba-VD: Dual Branch VideoMamba with Gated Class Token Fusion for Violence Detection
-
They Infer What You Meant: Models Represent Communicative Intent More Reliably Than They Act On It
-
Probe, Don't Prompt: A Hidden-State Probe for Metadata Filtering in Multi-Meta-RAG
-
Turning Off-Policy Tokens On-Policy: A Plug-in Approach for Improving LLM Alignment
-
msPCA: An R Package for Sparse PCA with Multiple Components
-
QuantFlow: A Federated Mamba-Based Post-Transformer Foundation Model for Time-Series Forecasting
-
GRAFT: Grafted Reference Audio for Fine-grained Pronunciation in Zero-shot Text-to-Speech
-
PhenoNEST: A Neuro-Symbolic Framework for Ontology-Aware Multimodal Plant Phenotyping and Trait Discovery
-
WeightCLIP: Aligning Datasets and Models for Weight Space Learning
-
Adversarial LassoNet: Robust Feature Selection via Stability-Driven Sparse Learning
-
Dictionaries, Not Darwin: Set-Level Selection Beats LLM Evolution in Scientific Equation Discovery
-
Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memory
-
NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation
-
Fusion: A Framework for Unified Sequential Token AdaptatIon in VisiOn TraNsformers
-
DELTA-TTS: Adapting Autoregressive Model into Diffusion Language Model for Text-to-Speech
-
CONTRA: Red-Teaming Configurations of Personalizable Agents
-
Exploring SAM Supervision for Fine-Grained UAV Target Segmentation under Data Scarcity
-
RayTun3R: Online Camera Adaptation in 3D Foundation Models
-
RotateAttention: RoPE-Aware Rotation and Range Rectification for INT4 Quantized Attention in Video Generation
-
kAgent: An execution-guided crash resolution agent for the Linux kernel
-
FORGE: Research-Trajectory Hijacking Attacks on Deep Research Agents
-
When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games
-
Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems
-
The Classics at SemEval-2026 Task 3: Combining Transformer Models and LLM-Generated Annotations for Dimensional Aspect-Based Sentiment Analysis
-
Decentralized Aggregation of LLM Predictions via Wagering Mechanisms
-
Angry but Accurate: Detecting and Profiling the Counter-Misinformation Ecosystem on Twitter
-
QEDBENCH: Quantifying the Alignment Gap in Automated Evaluation of University-Level Mathematical Proofs
-
Global Logic and Local Search: Dual-Stream Multimodal In-Context Learning for Verifiable Industrial Anomaly Detection
-
WPG-MoE: Weak-Prior-Guided Dense Mixture-of-Experts for User-Level Social Media Depression Detection
-
FormalRx: Rectify and eXamine Semantic Failures in Autoformalization
-
You Frame It: How Conceptual Representations Shape LLM Detection and Reasoning about Antisemitism
-
HiMe: Hierarchical Embodied Memory for Long-Horizon Vision-Language-Action Control
-
CertMix: Certified, Data-Efficient Metamaterial Design by Affine Mixing of Aligned Neural-Implicit Weight Spaces
-
VERITAS: Towards a General-Purpose Replication Tool for Scientific Research
-
Grokking Is Conditional and Fragile: A Fully-Tractable, Multi-Seed Study at 12K Parameters
-
Localized LoRA-MoE: Block-wise Low-Rank Experts With Adaptive Routing
-
Learning Only What Valid Adapters Can Express: Subspace-Constrained Adaptation Against Fine-Tuning Poisoning
-
Silicon Sampling via Cross-Survey Transfer
-
Incentivizing Vision Language Models to Search for Long Video Question Answering
-
MTEB-PT: A Text Embedding Benchmark for Brazilian Portuguese
-
Language Models Represent and Transform Concepts with Shared Geometry
-
BVS: Bayesian Visual Search with Multimodal Large Language Model for Fine-grained Perception
-
FairFlow: Demystifying and Mitigating Stereotype Bias in Text-to-Image Diffusion Transformers
-
ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog
-
DSWAM: A Dual-System World Action Foundation Model for Fine-Grained Robot Manipulation
-
Progressive Disclosure for LLM-Maintained Wiki Knowledge Bases: a Preregistered Ablation
-
ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction
-
SHINE: A Scalable In-Context Hypernetwork for Mapping Context to LoRA in a Single Pass
-
DynaWM: A Base-VLA-Guided World Foundation Model for Moving-Object Manipulation
-
GuideMe: Multi-Domain Task Guidance and Intervention in Streaming Video
-
Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs
-
The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models
-
Framework of Thoughts: A Foundation Framework for Dynamic and Optimized Reasoning based on Chains, Trees, and Graphs
-
Legible-by-Construction: Attention and End-to-End Transformers
-
Variable Bit-width Quantization: Learning Per-Group Precision for "Bigger-but-Smaller" Language Models
-
MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution
-
Evaluating Agentic Harness Systems for Autonomous Computational Pathology
-
Conversational Human Audio-visual Talking Dialogue Generation
-
VISTA: Auditing Semantic Divergence in Vision-Language Models
-
How Do Diffusion Classifiers Decide? A Bias-Centric Evaluation
-
OmniLayout: A Schematic-Coupled Multimodal Benchmark for Constraint-Aware Geometric Reasoning in PCB Layout
-
NarrativeTrack: Evaluating Entity-Centric Reasoning for Narrative Understanding
-
Multi-Resolution Flow Matching: Training-Free Diffusion Acceleration via Staged Sampling
-
Anti-Prompt: Image Protection against Text-Guided Image-to-Video Generation
-
Scaling with Confidence: Calibrating Confidence of LLMs for Adaptive Test Time Scaling
-
UA-ChatDev: Uncertainty-Aware Multi-Agent Collaboration for Reliable Software Development
-
ADVENT: LLM-Driven Automatic Predicate Invention for ILP
-
On the Limits of Steering Vectors for Preference-Aligned Generation
-
Zeus: Towards Tuning-Free Foundation Model for Time Series Analysis
-
LLM-Empowered Multimodal Fusion Framework for Autonomous Driving: Semantic Enhancement and Channel-Adaptive Design
-
From Lab to Reality: A Practical Evaluation of Deep Learning Models and LLMs for Vulnerability Detection
-
Towards Interactive Global Geolocation Assistant
-
Generic Expert Coverage for Pruning SparseMixture-of-Experts Language Models
-
EPnG: Adaptive Expert Prune-and-Grow for Parameter-Efficient MoE Fine-tuning
-
Object Aligner: A Configurable JSON Schema Similarity Score for Graphs, Applied to LLM Prompt Optimization
-
Assessing VLM Reliability for Medical Image Quality Evaluation Under Corruption and Bias
-
Know Your Source: A Public Knowledge Store for Media Background Checks
-
Class-Grouped Normalized Momentum and Faster Hyperparameter Exploration to Tackle Class Imbalance in Federated Learning
-
DetailAnywhere: Fashion Detail Generation via Cross-Modal Feature Alignment Distillation
-
DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing
-
CreativityNeuro: Steering Language Model Weights to Improve Divergent Thinking and Reduce Mode Collapse
-
OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration
-
Safe and Adaptive Cloud Healing: Verifying LLM-Generated Recovery Plans with a Neural-Symbolic World Model
-
Autonomous discovery of traffic laws with AI traffic scientists
-
Safeguarding LLM Agents from Misalignment through Provenance Analysis
-
Mastermind: Strategy-grounded Learning for Repository-Scale Vulnerability Reproduction
-
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
-
Scaling Trends for Lie Detector Oversight in Preference Learning
-
Can Language Models Actually Retrieve In-Context? Drowning in Documents at Million Token Scale
-
Towards Metric-Agnostic Trajectory Forecasting
-
PixelEyes: Decoupling Perception and Reasoning for Pinpoint Visual Evidence Seeking
-
"Don't Say It!": Constraints, Compliance, and Communication when Language Models Play Taboo
-
Seahorse: A Unified Benchmarking Framework for Spatiotemporal Event Modeling
-
LongVQUBench: Benchmarking Long-Term Video Quality Understanding of Vision-Language Models
-
SenseWalk: Agent-Based Semantic Trajectory Simulation Powered by Large Language Models in Zoned Environments
-
YOMI-Bench: A Benchmark for Evaluating Kanji Reading and Phonological Understanding of LLMs for Japanese
-
Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates
-
GEAR-Seg: A Grounded Explainable Agent for Reasoning Segmentation and Data Engine
-
EquiSteer: Cross-Attention Steering Towards a Fairer Text-Guided Image Generation
-
Multi-scale Mixture of World Models for Embodied Agents in Evolving Environments
-
Agentic generation of verifiable rules for deterministic, self-expanding reaction classification
-
Real-Time Hard Negative Sampling via LLM-based Clustering for Large-Scale Two-Tower Retrieval
-
Diffusion-GR2: Diffusion Generative Reasoning Re-ranker
-
OmniMoE: An Efficient MoE by Orchestrating Atomic Experts at Scale
-
Beyond Activation Alignment:The Alignment-Diversity Tradeoff in Task-Aware LLM Quantization
-
Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation
-
Salt: Self-Consistent Distribution Matching with Cache-Aware Training for Fast Video Generation
-
Persona Without Substrate: Regime-Dependence and the LLM Individuation Problem
-
SemiScope: Disentangling Classifier Tuning and Joint Optimization in Semi-Supervised Security Classification
-
Is One Layer Enough? Training A Single Transformer Layer Can Match Full-Parameter RL Training
-
FusionFactory: Fusing LLM Capabilities with Multi-LLM Log Data
-
AGC-Bench: Measuring Artificial General Creativity
-
Concept Alignment Contrast and Long-Short Prompt Memory for Test-Time Adaptation of SAM3 in Medical Image Segmentation
-
Beyond the Prompt: Jailbreaking Function-Calling LLMs via Simulated Moderation Traces
-
From Technical Metrics to User Perception: A User Study of a Multimodal Human-Robot Interaction System for Object Detection and Grasping
-
Large language models replicate and predict human cooperation across experiments in game theory
-
Robust Text Watermarking for Large Language Models via Dual Semantic Embeddings
-
LOPA: Enhancing Spoken Language Assessment via Latent Ordinal Prototype Alignment
-
When LLMs Read Tables Carelessly: Measuring and Reducing Data Referencing Errors
-
GR2 Technical Report
-
SwiftAudio: Data-Efficient Caption-Only Distillation for One-Step Text-to-Audio Diffusion-based Generation
-
Seeing Through Multiple Views: Parameter-Efficient Fine-Tuning via Selective Neurons for Consistent Radiology Report Generation
-
Indi-RomCoM: Code-Mixed Benchmark for Evaluating LLMs on Romanized Indic-English Instructions
-
ALM2Vec: Learning Audio Embeddings for Universal Audio Retrieval with Large Audio-Language Models
-
FARS: A Fully Automated Research System Deployed at Scale
-
WIDER-FAIR: An Annotated Version of the WIDER-FACE Dataset for Fairness Evaluation
-
Fork-Think with Confidence
-
CLExEval: A Human-in-the-Loop Framework for Qualitative Evaluation of LLM Clinical Reasoning
-
Can LLMs Imagine Moral Alternatives Beyond Binary Dilemmas?
-
Self-Study Reconsidered: The Hidden Fragility of Learning from Self-Generated QA
-
AgRefactor: Self-Evolving Agentic Workflow for HLS Compatibility and Performance
-
MultiUAV-Plat: An LLM-Oriented Platform, Benchmark and Framework for Multi-UAV Collaborative Task Planning
-
ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents
-
Probing Stylistic Appropriation using Large Language Models: An Evaluation Framework for Copyright Infringement under EU Law
-
Horseshoe Priors for Spatial Small Area Estimation: Regular Variation, Tail Robustness, and Deep Learning
-
Fora: From Weight-Space to Function-Space Protection in Capability-Preserving Fine-Tuning
-
A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control
-
An Empirical Study of Security Calibration in Large Language Models for Code
-
Generalizing Numerical Reasoning in Table Data through Operation Sketches and Self-Supervised Learning
-
CLOUDADV: Decision-Aligned Instance Sizing with Zero-Shot Foundation Models under Drift
-
MOSAIC: Orchestrating Collaborative Knowledge Tracing with Hierarchical Semantic Alignment
-
The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables
-
Multimodal Mathematical Reasoning with Diverse Solving Perspective
-
Attractor States Emerge in Multi-Turn LLM Conversations
-
You Had One Job: Per-Task Quantization Using LLMs' Hidden Representations
-
Know Before You Fetch: Calibrated Retrieval-Budget Allocation for Retrieval-Augmented Generation
-
GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots
-
A Diagnostic Framework and Multi-Evaluator Audit of Evaluator-Driven Preference Dynamics in Self-Adapting LLM Agents
-
DataComp-VLM: Improved Open Datasets for Vision-Language Models
-
Reliability-Prioritized Fine-Grained Generation in Multimodal Large
-
Momentum Guidance: Plug-and-Play Guidance for Flow Models
-
LLM-Guided Planning for Multi-hop Reasoning over Multimodal Nuclear Regulatory Documents
-
Developmental Trajectories of Situation Modeling and Mentalizing in Transformer Language Models
-
The strength of clinical evidence is recoverable from language model representations but not from their stated grades
-
TriageRA-CCF: Source-Side Clinical Confidence and Coverage Signals for Adaptive Rank Budgeting in Medical LLMs
-
Can LLMs Hire Fairly? Racial Bias in Resume Screening
-
StrucTab: A Structured Optimization Framework for Table Parsing
-
Blackknife: Hard-Label Query-Limited Black-Box Attacks on Heterogeneous Graph Neural Networks
-
CAREBench: A Child-Safety Risk Benchmark for Language Models
-
DreamForge-World 0.1 Preview: A Low-Compute Real-Time Controllable World Model
-
AlgoSkill: Learning to Design Algorithms by Scheduling Human-Like Skills
-
RIPA: Sensory-Vector Prompt Injection Attacks on LLM-Controlled ROS 2 Robots
-
Reward-Free Code Alignment from Pretrained or Fine-Tuned LLM: Unpacking the Trade-offs for Code Generation
-
SemJoin: Semantic Join Optimization
-
CLIMP: Contrastive Language-Image Mamba Pretraining
-
APRIL-MedSeg: A Modular Medical Image Segmentation Toolbox Embracing Modern Paradigms
-
How much of an LLM-generated clinical corpus is actually new? A production-scale measurement of content redundancy for provenance classification
-
Mandol: An Agglomerative Agent Memory System for Long-Term Conversations
-
Self-Evolving World Models for LLM Agent Planning
-
Towards Evaluating Data Priors for Tabular Foundation Models
-
Legal Domain Adaptation of Modern BERT Models
-
FFAvatar: Feed-Forward 4D Head Avatar Reconstruction from Sparse Portrait Images
-
A Multi-Dataset Benchmark for Evaluating LLM Agents in Microservice Failure Diagnosis
-
Self-Evolving Agentic Image Restoration via Deliberate Planning and Intuitive Execution
-
An AI agent for treatment reasoning over a biomedical tool universe
-
Agentic Abstention: Do Agents Know When to Stop Instead of Act?
-
SFBench: The SciFy Scientific Feasibility Benchmark
-
MirrorCode: AI can rebuild entire programs from behavior alone
-
Optimizing Expert-Designed Crystal Graph Networks for Band-Gap Prediction with an Autonomous LLM Research Loop
-
A Machine-Verified Proof of a Quantum-Optimization Conjecture
-
Towards Spatial Trace with Reasoning in Vision-Language Models for Robotics
-
How Do LLMs Cite? A Mechanistic Interpretation of Attribution in Retrieval-Augmented Generation
-
Low-cost concept-based localized explanations: How far can we get with training-free approaches?
-
SABER-Math: Automated Benchmark for Information Retrieval Evaluation in Mathematics
-
FinInvest-GTCN: Explainable Graph-Temporal-Causal Modeling for Risk-Aware Investment Decision Optimization
-
Fine-Tuning General-Purpose Large Language Models for Agricultural Applications:A Reproducible Framework and Evaluation Protocol Based on Qwen3-8B
-
SCARCE: Scalable Cascade Analysis for Rare-event Characterisation via Embeddings
-
Latent Bridges for Multi-Table Question Answering
-
HARD-KV: Head-Adaptive Regularization for Decoding-time KV Compression
-
PromptGNN-sim: Deep Fusion and Alignment of GNN and LLMs for Text-Attributed Graph Learning
-
Online Data Selection for Instruction Tuning via Gaussian Processes
-
Diffusion Fine-tuning with Rewarded Moment Matching Distillation
-
TraceLab: Characterizing Coding Agent Workloads for LLM Serving
-
FlipGuard: Defending Large Language Models Against Quantization-Conditioned Backdoor Attacks
-
Democratic ICAI: Debating Our Way to Steering Principles from Preferences
-
CSD: Content-aware Speculative Decoding for Efficient Image Generation
-
Enhancing Numerical Prediction in LLMs via Smooth MMD Alignment
-
Grounded Iterative Language Planning: How Parameterized World Models Reduce Hallucination Propagation in LLM Agents
-
VASAE: Naming SAE Dictionary Directions with Vocabulary-Aligned Anchoring
-
Flexformer: Flexible Linear Transformer with Learnable Attention Kernel
-
Health-ORSC-Bench: A Benchmark for Measuring Over-Refusal and Safety Completion in Health Context
-
VLM-Aware Meta-Optic Front-End Design for Frozen Vision-Language Models
-
CascadeOcc: Rethinking 3D Occupancy World Models with Cascaded VQ Representations
-
The Context-Ready Transformer
-
From Signals to Transfer: A Factorised Study of Probe-Based Uncertainty Estimation in Large Language Models
-
Otter Weather: Skillful and Computationally Efficient Medium-Range Weather Forecasting
-
When Role-playing, Do Models Believe What They Say?
-
Where Larger Models Excel: The Primacy of Constraint-Guided Reasoning
-
DualEval: Joint Model-Item Calibration for Unified LLM Evaluation
-
ScheMatiQ: From Research Question to Structured Data through Interactive Schema Discovery
-
Helpfulness Hurts: Domain-Dependent Degradation of Mid-Trained Compassion Values Under Post-Training
-
Investigating LLM's Problem Solving Capability -- a Study on Statics Questions
-
Charting the Growth of Social-Physical HRI (spHRI): A Systematic Review Pipeline Augmented by Small Language Models
-
LAMP: Lane-Aligned Motion Primitives for Feasible Trajectory Prediction
-
Lacuna: A Research Map for Machine Learning
-
Disco-LoRA: Disentangled Composition of Content, Style, and Motion for Multi-concept Video Customization
-
Robust Onion: Peeling Open Vocab Object Detectors Under Noise
-
SamaVaani: Auditing and Debiasing Multilingual Clinical ASR for Indian Languages
-
ReaORE: Reasoning-Guided Progressive Open Relation Extraction Empowered by Large Reasoning Models
-
What Do Deepfake Benchmarks Measure? An Audit Using Frozen Self-Supervised Representations
-
TraMP-LLaMA: Generative Interpretability with Decoupled Instruction Tuning for Facial Expression Quality Assessment
-
OpenFinGym: A Verifiable Multi-Task Gym Environment for Evaluating Quant Agents
-
The Capability Frontier: Benchmarks Miss 82% of Model Performance
-
Know2Guess: A Contamination-Aware Multi-Zone Benchmark for Knowledge-Boundary Evaluation in Large Language Models
-
Joint Learning of Experiential Rules and Policies for Large Language Model Agents
-
Chai: Agentic Discovery of Cryptographic Misuse Vulnerabilities
-
ShareLock: A Stealthy Multi-Tool Threshold Poisoning Attack Against MCP
-
TEMPO-Diffusion: Temporally Exposed Malicious Poisoning of Diffusion Models
-
Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols
-
Zero-shot Tweet-Level Stance Detection Enhanced by External Knowledge and Reflective Chain-of-Thought Reasoning
-
Estimating Uncertainty in Classifier Performance with Applications to Large Language Models and Nested Data
-
RSPC: A Benchmark for Modeling Stress and Psychiatric Conditions in Digitally Mediated Relationships using Psychiatrist Annotations
-
Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets
-
Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decoding
-
ExTra: Exploratory Trajectory Optimization for Language Model Reinforcement Learning
-
Don't Go Breaking My LLM: The Impact of Pruning Attention Layers on Explanation Faithfulness and Confidence Calibration
-
EPTS: Elastic Post-Training Sparsity for Efficient Large Language Model Compression
-
MiniOpt: Reasoning to Model and Solve General Optimization Problems with Limited Resources
-
Distill on a Diet: Efficient Knowledge Distillation via Learnable Data Pruning
-
SeFi-Image: A Text-to-Image Foundation Model with Semantic-First Diffusion
-
TokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems
-
EchoStyle: Unlocking High-Fidelity Video Stylization with Reverse Data Synthesis
-
SurgAtlas: A Large-Scale Surgical Video-Language Dataset with 2,391 Hours of Open and Minimally Invasive Surgery
-
BiPACE: Bisimulation-Guided Policy Optimization with Action Counterfactual Estimation for LLM Agents
-
Riazi-8B: An Urdu Large Language Model for Mathematical Reasoning
-
SpeechEQ: Benchmarking Emotional Intelligence Quotient in Socially Aware Voice Conversational Models
-
Same Evidence, Different Answer: Auditing Order Sensitivity in Multimodal Large Language Models
-
Beyond One-Size-Fits-All: Diagnosis-Driven Online Reinforcement Learning with Offline Priors
-
Project Auto-World: Towards Automated Benchmarking of Neural Relational Reasoners
-
Business as Rulesual: A Benchmark and Framework for Business Rule Flow Modeling with LLMs
-
CANDLE: Character-level Arabic Noise Deduplication using Lightweight Encoder
-
Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning
-
Beyond Logprobs: A Multi-Signal Confidence Engine for LLM-Based Document Field Extraction
-
To Compare, or Not to Compare: On Methodological Practices in Evaluating Social Bias
-
NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?
-
Layer-wise Probing of wav2vec 2.0 and Whisper for Consonant Cluster Reduction in African American English
-
DiffusionBench: On Holistic Evaluation of Diffusion Transformers
-
SignNet-1M: Large-Scale Multilingual Sign Language Video Dataset with Downstream Benchmarks
-
MM-TRELLIS: Point-Cloud Guided Multi-Modal 3D Vehicle Generation in Autonomous Driving
-
HANCLIP: A Family of Hyperbolic Angular Negation Vision Language Models
-
Rule2Text: A Framework for Generating and Evaluating Natural Language Explanations of Knowledge Graph Rules
-
Spectral Evolution-Guided Token Pruning in Multimodal Large Language Models
-
On the Stability of Prompt Ranking in Large Language Model Evaluation
-
Evaluating the Interpretability of Sparse Autoencoders with Concept Annotations
-
Are We Ready For An Agent-Native Memory System?
-
RAVEN: A Regime-Aware Variable-context Expert Network for Financial Time Series Forecasting
-
Holistic Data Scheduler for LLM Pre-training via Multi-Objective Reinforcement Learning
-
BehaviorBench: Benchmarking Foundation Models for Behavioral Science Tasks
-
RASC+: Retrieval-Constrained LLM Adjudication for Clinical Value Set Authoring
-
L3Cube-MahaPOS: A Marathi Part-of-Speech Tagging Dataset and BERT Models
-
ReMMD: Realistic Multilingual Multi-Image Agentic Verification for Multimodal Misinformation Detection
-
Can Scale Save Us From Plasticity Loss in Large Language Models?
-
CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference
-
LLMs Prompted for Legal Context Object More: Overrefusal from Small On-Premises LLMs in Criminal Legal Context
-
BluTrain: A C++/CUDA Framework for AI Systems
-
Beyond the Autoregressive Horizon: A Comprehensive Survey of Diffusion Models, World Modelling, and State Space Models for Code
-
Quantifying Prior Dominance in RAG Systems
-
PHANTOM: A Large-Scale Dataset of Multimodal Adversarial Attacks for Vision-Language Models
-
VeryTrace: Verifying Reasoning Traces through Compilable Formalism and Structured Verification
-
Qwen-AgentWorld: Language World Models for General Agents
-
SHERLOC: Structured Diagnostic Localization for Code Repair Agents
-
NoLimits.jl: Flexible and Composable Nonlinear Mixed-Effects Modeling in Julia
-
Scaling Laws for Task-Specific LLM Distillation
-
CALIBER: Calibrating Confidence Before and After Reasoning in Language Models
-
AgentArmor: A Framework, Evaluation, \& Mitigation of Coding Agent Failures
-
PsyScore: A Psychometrically-Aware Framework for Trait-Adaptive Essay Scoring and ZPD-Scaffolded Feedback
-
BAFIS: Dataset + Framework to assess occupational Bias and Human Preference in modern Text-to-image Models
-
AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models
-
VCG: A Multimodal Retrieval Framework for E-Commerce Video Feeds under Extreme Cold-Start Conditions
-
MiqraBERT: Regression-Based Sentence-BERT Finetuning for Biblical Hebrew Parallel Detection
-
Characterizing Narrative Content in Web-scale LLM Pretraining Data
-
CARE: Competence-Aware Reward Shaping for Adaptive Reasoning Length in Video-MLLMs
-
ViCoStream: Streaming VideoLLMs Can Run Beyond 100 FPS with Stage-Wise Coordinated Inference
-
IdealGPT: Iteratively Decomposing Vision and Language Reasoning via Large Language Models
-
SAM3 Self-Distillation for Fine-Grained GOOSE 2D Semantic Segmentation
-
ProMUSE: Progressive Multi-modal Uncertainty-guided Staged Evidential Alzheimer Disease Classification
-
Interpretable and Verifiable Hardware Generation with LLM-Driven Stepwise Refinement
-
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
-
Techniques for Peak Memory Reduction for LoRA Fine-tuning of LLMs on Edge Devices
-
Improving Code-Switching ASR with Code-Mixing Guided Synthetic Speech
-
Confidence Calibration for Multimodal LLMs: An Empirical Study through Medical VQA
-
The Hidden Evolution of Disguised Visual Context inside the VLM
-
Spectral Retrieval-Augmented Time-Series Forecasting
-
Insulin4RL: Real-Time Insulin Management in the Intensive Care Unit for Offline Reinforcement Learning
-
FloatDoor: Platform-Triggered Backdoors in LLMs
-
Where Does Social Reasoning Come From? Capability Provenance in Language Models
-
Probe-and-Refine Tuning of Repository Guidance for Coding Agents
-
Pixel-Level Residual Diffusion Transformer: Scalable 3D CT Volume Generation
-
GLARE: A Natural Language Interface for Querying Global Explanations
-
Conflict-Aware Retriever Editing for Knowledge Injection Attacks on LLM-Based RAG Systems
-
LLMs Struggle to Measure What Distinguishes Students of Different Proficiency Levels: A Study of Item Discrimination in Reading Comprehension Assessment
-
FLiP: Towards understanding and interpreting multimodal multilingual sentence embeddings
-
Hierarchical Multi-Modal Retrieval for Knowledge-Grounded News Image Captioning
-
Beyond Scalar Scores: Exploring LLM-based Metrics for Clinical Significance Evaluation in Radiology Reports
-
Diffusion-Proof: Recipe for Formal Theorem Proving Beyond Auto-Regressive Generation
-
Steerable Cultural Preference Optimization of Reward Models
-
Leadership as Coordination Control: Behavioral Signatures and the Recovery-Advantage Boundary in Multi-Agent LLM Teams
-
Improve Large Language Model Systems with User Logs
-
InfoPO: Information-Driven Policy Optimization for User-Centric Agents
-
VISUALSKILL: Multimodal Skills for Computer-Use Agents
-
Enhancing Multilingual Reasoning via Steerable Model Merging
-
Dango: A Strictly L1-Only Large Language Model for Studying Second Language Acquisition
-
Adv-TGD: Adversarial Text-Guided Diffusion for Face Recognition Impersonation Attacks
-
Vibe Coding Ate My Homework: An evaluation of AI approaches to greenfield software engineering and programming
-
Be Your Own Teacher: Steering Protein Language Models via Unsupervised Reward Optimization
-
EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts
-
OrthoReg: Orthogonal Regularization for Hybrid Symbolic-Neural Dynamical Systems
-
CAOA -- Completion-Assisted Object-CAD Alignment
-
Freeing the Law with LOCUS: A Local Ordinance Corpus for the United States
-
Confidence is Not Reliability: Rethinking MC Dropout in Brain Tumour Segmentation
-
See First, Answer Later: Visual Evidence Pre-Alignment via Sufficiency-Driven RL
-
When Rules Learn: A Self-Evolving Agent for Legal Case Retrieval
-
Dissecting model behavior through agent trajectories
-
SEAGym: An Evaluation Environment for Self-Evolving LLM Agents
-
DecoSearch: Complexity-Aware Routing and Plan-Level Repair for Text-to-SQL
-
Feynman Kac Reweighted Schr\"odinger Bridge Matching for Surface-Based Tau PET Harmonization
-
Agentic Discovery of Non-Canonical Antimicrobial Peptides with AMPGAN v3
-
The Benchmark Illusion: Pruned LLMs Can Pass Multiple Choice but Fail to Answer
-
Reading between the Lines: Leveraging Large Language Models for Global Dementia and Depression Assessment from Clinical Interviews
-
ZeroSyl: Simple Zero-Resource Syllable Tokenization for Spoken Language Modeling
-
RadSEM: A Finding-by-Finding Metric for Clinical Consistency in Radiology Reports
-
AnchorKV: Safety-Aware KV Cache Compression via Soft Penalty with a Refusal Anchor
-
LiveStarPro: Proactive Streaming Video Understanding with Hierarchical Memory for Long-Horizon Streams
-
When LLMs Analyze Scars: From Images to Clinically-Meaningful Features
-
S4oP: Operator-level Pruning of Structured State Space Models for Resource-Constrained Devices
-
TaFD: Threat-Aware Frequency Decoupling for Adversarial Robustness against Heterogeneous Attacks
-
EventDrive: Event Cameras for Vision-Language Driving Intelligence
-
Advances in 4D Representation: Geometry, Motion, and Interaction
-
SCC-Loc: A Unified Semantic Cascade Consensus Framework for UAV Thermal Geo-Localization
-
LLMs Infer Cultural Context but Fail to Apply It When Responding
-
Decoding Hidden Deception in Reasoning LLMs: Activation Explainers for Deception Auditing
-
Nothing from Something: Can a Language Model Discover 0?
-
findsylls: A Language-Agnostic Toolkit for Syllable-Level Speech Tokenization and Embedding
-
DeepInsight: A Unified Evaluation Infrastructure Across the Physical AI Stack
-
ParkingTransformer: LLM-Enhanced End-to-End Trajectory Planning for Autonomous Parking
-
HRDX: A Large-Scale Vector HD-Map Dataset
-
Mask-Proof: An LLM-based Automated Data Curation Pipeline on Mathematical Proofs
-
Mask Proposal Voting Based on Geodesic Framework for Robust Image Segmentation
-
On the Adversarial Robustness of Multimodal LLM Judges
-
Deep Residual Injection for Full-Spectrum Forensic Signal Perception in Multimodal Large Language Models
-
GRACE: Boosting Video MLLMs with Grounded Action-Centric Evidence for Viewer Sentiment Prediction
-
Chronological Blindness: Benchmarking Temporal Reasoning in Vision-Language Models with CHRONOSIGHT
-
LLM-Based Visual Explanation Evaluation Framework for Assessing the Explainability of Facial Skin Disease Classification Models
-
Towards Verifiable Agentic Data Science: Solving Irregular TSQA Via Tool-Grounded Reasoning
-
MVEB: Massive Video Embedding Benchmark
-
GradPower: Powering Gradients for Faster Language Model Pre-Training
-
Constitutional Value Potentials: reading and steering internal priority margins in language models
-
Beyond the Blood Draw: Explainable Machine Learning for Non-Invasive Dysglycemia Risk Screening
-
Pre-Training for Simulation-Based Science: A Study on Jet Foundation Model Training Objectives
-
Distilling latent electrostatics from foundation machine learning interatomic potentials
-
Not All Skills Help: Measuring and Repairing Agent Knowledge
-
EIBench: A Simulator-Based Benchmark and Turn-Credit RL for Emotion Management
-
TuneJury: An Open Metric for Improving Music Generation Preference Alignment
-
Risk-Aware LLM Agents for Geospatial Data Retrieval: Design and Preliminary Adversarial Evaluation
-
ACC: Compiling Agent Trajectories for Long-Context Training
-
The Value Axis: Language Models Encode Whether They're on the Right Track
-
LLM Judges Have Dark Current: A Psychometric Datasheet for LLM-as-a-Judge Evaluation
-
The Truth Stays in the Family: Enhancing Contextual Grounding via Inherited Truthful Heads in Model Lineages
-
Pepti-Agent: An AI Agent for Peptide Design and Optimization
-
Metric Match: A Subset Selection Approach to Evaluating LLM Judge Reliability
-
LatentGym: A Testbed For Cross-Task Experiential Learning With Controllable Latent Structure
-
EdgeZSAD: Practical Zero-Shot Anomaly Detection on Edge Devices
-
OneFocus: Enabling Real-World X-ray Security Screening with a Unified Vision-Language Model
-
Latent Thought Flow: Efficient Latent Reasoning in Large Language Models
-
MIRAGE: Auditing Anti-Muslim Bias in Frontier LLMs Across Reasoning, Agentic, and Time-Coupled Conditions
-
A Systematic Evaluation of Large Language Models for PTSD Severity Estimation: The Role of Contextual Knowledge and Modeling Strategies
-
DoubtProbe: Black-Box Jailbreak Defense via Structural Verification and Semantic Auditing
-
Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving
-
BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation
-
PaperJury: Due-Process Review for Bounded LaTeX Revision
-
A Large-Scale Multi-Dimensional Empirical Study of LLMs for Conversation Summarization
-
Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMs
-
MosaicQuant: Inlier-Outlier Disaggregation for Unified 4-Bit LLM Quantization
-
LLM-as-Code Agentic Programming for Agent Harness
-
SciText2Eq: Assessing LLMs for Explainable Equation Generation for Scientific Creativity
-
Sub-Semantic Image Segmentation
-
Scaling LLM Reasoning from Minimal Labels: A Semi-Supervised Framework with a Lightweight Verifier
-
Few-Shot Biomedical Relation Extraction with Large Language Models: A Viable Alternative to Supervised Learning?
-
Let LLMs Judge Each Other: Multi-Agent Peer-Reviewed Reasoning for Medical Question Answering
-
LLM-Assisted Stance Detection in Scientific Discourse: A Test Case in Bayesian Cognitive Science
-
Koshur Diacritizer: A Byte-Level Sequence-to-Sequence Model for Kashmiri Diacritic Restoration
-
PVminerLLM2: Improving Structured Extraction of Patient Voice via Preference Optimization
-
Long-Context Modeling via GSS-Transformer Hybrid Architecture with Learnable Mixing
-
UXBench: Measuring the Actionability of LLM-Generated UX Critiques
-
AgentLeak: A Benchmark for Internal-Channel Privacy Leakage in Multi-Agent LLM Systems
-
Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents
-
FLaRA: Predicting Future Latent Representations for Accident Anticipation
-
VideoWeave: Unlocking Geometric Consistency in Video Generation via Joint Geometry-Video Modeling
-
RT-VLA: Real-Time Vision-Language-Action Models via Knowledge Distillation
-
The Culture Funnel: You Can't Align What isn't in the Data
-
EiCAP: Beyond Fluency, Probing and Improving Emotional Intelligence in LLMs via Psychologically Grounded Multi-Turn Dialogue
-
Hybrid Classical-Quantum Variational Autoencoder for Neural Topic Modeling
-
A General Framework for Decision Trees via Bregman Divergences
-
From Prompts to Responses: Dual-Sided Data Leakage and Defense in Split Large Language Models
-
MeEvo: Metacognitive Evolution Combined with Natural Evolution for Automatic Heuristic Design
-
ADORE: Iterative Query Expansion with Retrieval-Grounded Relevance Feedback
-
Clay-CNN Hybrids: Leveraging Geo-Foundational Models as Auxiliary Context for Landslide Detection
-
AudioDER: A Deduplication-Enhanced Reasoning Dataset for Post-Training Large Audio-Language Models
-
DLawBench: Evaluating LLMs Through Multi-Turn Legal Consultation
-
Harsher on Male? Evaluating LLMs on Gender-Asymmetric Moral Framing Across Diverse Conflict Scenarios
-
The Curse and Blessing of Mean Bias in FP4-Quantized LLM Training
-
Beyond Perplexity: UTF-8 Validity in Byte-aware Language Models
-
SciDef: Datasets and Tools for Automated Definition Extraction from Scientific Literature with LLMs
-
Revisiting Vehicle Color Recognition in Long-Tailed Surveillance Scenarios
-
Causal Inference with Generative Artificial Intelligence: Application to Texts as Treatments
-
Does AI Reviewer See the Full Picture? Attacking and Defending Multimodal Peer Review
-
Shopping Reasoning Bench: An Expert-Authored Benchmark for Multi-Turn Conversational Shopping Assistants
-
Small LLMs for Biomedical Claim Verification: Cost-Effective Fine-Tuning, Structural Dataset Shortcuts, and Cross-Domain Generalization
-
FENCE: A Financial and Multimodal Jailbreak Detection Dataset
-
Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents
-
One Polluted Page Is Enough: Evaluating Web Content Pollution in Generative Recommenders
-
IVIE: A Neuro-symbolic Approach to Incremental and Validated Generation of Interactive Fiction Worlds
-
SkMTEB: Slovak Massive Text Embedding Benchmark and Model Adaptation
-
When Smaller Wins: Dual-Stage Distillation and Pareto-Guided Compression of Liquid Neural Networks for Edge Battery Prognostics
-
When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering
-
HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents
-
EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
-
Benchmarking AI Agents for Addressing Scientific Challenges Across Scales
-
ToolSense: A Diagnostic Framework for Auditing Parametric Tool Knowledge in LLMs
-
MLUBench: A Benchmark for Lifelong Unlearning Evaluation in MLLMs
-
Topical Phase Transitions in Artificial Intelligence Research: Large-Scale Evidence and an Early-Warning Signature for Emerging Topics
-
DailyReport: An Open-ended Benchmark for Evaluating Search Agents on Daily Search Tasks
-
Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach
-
Nous: An Attempt to Extract and Inject the Cognition Behind Prediction-Market Behavior
-
Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning
-
Agents-K1: Towards Agent-native Knowledge Orchestration
-
LLM-Powered Personalized Glycemic Assessment in Type 2 Diabetes with Wearable Sensor Data
-
Stubborn: A Streamlined and Unified Reinforcement Learning Framework for Robust Motion Tracking and Fall Recovery for Humanoids
-
LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling
-
EDEN: A Large-Scale Corpus of Clinical Notes for Italian
-
KCSAT-ML: Probing Reasoning Models with Nationwide-Cohort Human Difficulty
-
The Order Matters: Sequential Fine-Tuning of LLaMA for Coherent Automated Essay Scoring
-
Density Field State Space Models: 1-Bit Distillation, Efficient Inference, and Knowledge Organization in Mamba-2
-
ConvMemory v2: A Recall-Preserving Top-10 Evidence Reranker for Conversational Memory Retrieval
-
A Navigable Manifold of Hypothesized Consciousness-Spectrum States in Language Model Representations
-
An Industrial-Scale Insurance LLM Achieving Verifiable Domain Mastery and Hallucination Control without Competence Trade-offs
-
Calibrating Overconfidence Without Sacrificing Confidence: Probe-Conditioned Head Intervention for LLMs
-
When Design Rules Break: Benchmark Composition Determines Whether Label Informativeness Predicts GNN Aggregator Choice
-
PL-KKT-hPINN: Enforcing Nonlinear Equality Constraints on Neural Networks via Piecewise-Linear Projection
-
Task Robustness via Re-Labelling Vision-Action Robot Data
-
Pushing the Limits of LLM Tool Calling via Experiential Knowledge Integration and Activation
-
Impatient Users Confuse AI Agents: High-fidelity Simulations of Human Traits for Testing Agents
-
PhantomBench: Benchmarking the Non-existential Threat of Language Models
-
TinyTroupe: An LLM-powered Multiagent Persona Simulation Toolkit
-
Rotate2Think: Geometric Priming via Orthogonal Rotation to Improve Language Model Reasoning
-
Less Context, More Accuracy: A Bi-Temporal Memory Engine for LLM Agents Where a Lean Retrieved Context Beats the Full History
-
When Attribution Patching Lies: Diagnosis and a Second-Order Correction
-
IDP-Bench: Benchmarking ability of LLMs to protect personal information in interdependent privacy contexts
-
When RL Fails after SFT: Rejuvenating Model Plasticity for Robust SFT-to-RL Handoff
-
3SPO: State-Score-Supervised Policy Optimization for LLM Agents
-
Generative Explainability for Next-Generation Networks: LLM-Augmented XAI with Mutual Feature Interactions
-
Gaming AI-Assisted Peer Reviews Poses New Risks to the Scientific Community
-
FedSteer: Taming Extreme Gradient Staleness in Federated Learning with Corrective Projections and Caching
-
BiWM: Advancing Open-Source Interactive Video World Models with Bidirectional Autoregression
-
KG-SoftMAP: Soft Knowledge-Graph Priors for Bayesian Network Structure Learning from Sparse Discrete Data
-
Speech Meets ELF: Audio Conditional Continuous-Target Diffusion for Speech Recognition and Translation
-
UPLOTS: A Unified Pretrained Language Model for Constrained Time-series Generation
-
Stop Early, Spend Less: Hidden-State Probes as a Practical Recipe for Streaming Moderation of LLM Outputs
-
SD-GRPO: Verifiable Segment Decomposition for Long-Form Vision-Language Generation
-
Conformal Prediction for Neural Operators: Distribution-Free Uncertainty Quantification in Physics Simulation
-
RealMath-Eval: Why SOTA Judges Struggle with Real Human Reasoning
-
Predicting Future Behaviors in Reasoning Models Enables Better Steering
-
STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios
-
What Fits (Into Few Tokens) Doesn't Overfit: Compression and Generalization in ML Research Agents
-
The Interlocutor Effect: Why LLMs Leak More Personal Data to Agents Than Humans
-
EstRTL: Functional Estimation Guided RTL Code Generation
-
Towards Diverse Scientific Hypothesis Search with Large Language Models
-
Blurry Window Attention
-
Self-EmoQ: Plutchik-Guided Value-based Planning to Drive Streaming Emotional TTS
-
Bellman-Taylor Score Decoding for Markov Decision Processes with State-Dependent Feasible Action Sets
-
FADA: Accessible fetal ultrasound interpretation and annotation with a selectively distilled unified vision-language model
-
The 1st PortraitCraft Challenge: A CVPR 2026 Workshop Competition on Portrait Composition Understanding and Generation
-
Dissect and Prune: Enhancing Robustness in AI-Generated Image Detection
-
POISE: Position-Aware Undetectable Skill Injection on LLM Agents
-
What Makes Video World Model Latents Action-Relevant: Prediction over Reconstruction
-
SRT: Super-Resolution for Time Series via Disentangled Rectified Flow
-
Liberating LLM Capabilities in Full-Duplex Speech Models
-
ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research
-
Principled Agent Debate: Adversarial Arbitration for Sycophancy Reduction in Large Language Models
-
Post-training is (Massive) Supervised Learning
-
Bridging Traditional Explainability Methods and Multimodal Multilingual Models: An XAI-Based Analysis
-
MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation
-
Evaluation Cards: An Interpretive Layer for AI Evaluation Reporting
-
ZIPP:Zero-shot Image Personalization from Personas
-
FF-JEPA: Long-Horizon Planning in World Models with Latent Planners
-
Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs
-
DN-Hypo-Pipeline: An AI-Driven Workflow for Hypothesis Generation via Large Language Models and Scientific Explanations
-
VESTA: A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents
-
Zero-Shot Learning in Industrial Scenarios: New Large-Scale Benchmark, Challenges and Baseline
-
Traxia: A Framework for Verifiable, Agent-Native Scientific Publishing
-
OmniMem: Perturbation-aware Memory Compression for Streaming Audio-Visual LLMs
-
Why Limit the Residual Stream to Layers and Not Tokens? Persistent Memory for Continuous Latent Reasoning
-
Observability for Delegated Execution in Agentic AI Systems
-
Code Is More Than Text: Uncertainty Estimation for Code Generation
-
Defending Against Malicious Finetuning by Scaling Train-time Adversarial Attacks
-
Jas: AI-Paired Engineering as a Revival of N-Version Programming
-
TABVERSE: Benchmarking Cross-Format Table Understanding in LLMs and VLMs
-
Implicit Causal Graph Construction in Text via Chain Discovery
-
CoVEBench: Can Video Editing Models Handle Complex Instructions?
-
See More, Think Deeper: Query-Expanded Visual Evidence and Answer-Clue Guided Reflection for Long Video Understanding
-
To Nuke or Not to Nuke: LLMs' (Missing) Ethical Reasoning and Actions in a High-Stakes Decision-Making Simulation
-
Efficient Skill Grounding via Code Refactoring with Small Language Models
-
GIFT: LLM-Guided State-Reward Interface for Financial Reinforcement Learning
-
InA-Probe: Instruction-Aware Active Probing for Time Series Forecasting with LLMs
-
DALE-CT: Depth-Aware Foundation Models for Computed Tomography
-
DyCo-RL: Dynamic Cross-Modal Coordination for Visual Reasoning
-
OmniTryOn: Video Try-On Anything at Once!
-
SSAFE: Simple and Strong AI-Generated Image Detection via Frozen Vision Encoders
-
HDRAgent: An Agentic Framework for Multi-Exposure HDR Imaging
-
Claude Code-Driving Scenario Mining for the Argoverse 2 Challenge
-
Latent Spatial Memory for Video World Models
-
POTATR: A Lightweight Image-to-Graph Model for Page-Level Table Extraction
-
Training-Free Generalized Few-Shot Segmentation through Open-Vocabulary Semantic Arbitration
-
Visual Template Inference for Data Extraction from Documents
-
Activation Steering Induces Emergent Misalignment: A More Comprehensive Evaluation
-
Sparse Memory Finetuning as a Low-Forgetting Alternative to LoRA and Full Finetuning
-
A large-scale nanocrystal database with aligned synthesis and properties enabling generative inverse design
-
AttentionCap: Transformer Based Capacitance Matrix Learning Toward Full-Chip Extraction
-
Illusions of the Gold Standard: A Large-scale Analysis of Human Evaluation Protocols for Long-form Text Generation
-
VideoWeaver: Evaluating and Evolving Skills for Agentic Long Video Generation
-
Explaining Data Mixing Scaling Laws
-
Contemporary AI lacks the imagination to diverge or negate in science
-
Stage-1 Controls the Entropy Regime, Not the Outcome
-
NutriMLLM: Multimodal Large Language Models for Dietary Micronutrient Analysis
-
Real-time body pose non-verbal communication with a consistency-based reliability measure
-
SEF-CLGC at SemEval-2026 Task 11: Logical Notation Impact on Language Model Performance
-
FMplex: Model Virtualization for Serving Extensible Foundation Models
-
NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning
-
GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting
-
Accelerating Reproducible Research in Synthetic EHR Generation
-
MalTree: Tracing Malware Evolution from Embeddings at Scale
-
AI-Driven Test Case Generation from Natural Language Requirements: A Survey of Techniques and Research Gaps
-
ShallowBench: Benchmarking Generative Drug Design Models on Shallow-Pocket Targets
-
OpenHalDet: A Unified Benchmark for Hallucination Detection across Diverse Generation Scenarios
-
Inside the Visual Mind: Neuroscience-Motivated Concept Circuits for Interpreting and Steering Vision Transformers
-
SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating
-
CULTURESCORE: Evaluating Cultural Faithfulness in Video Generation Models
-
When Large Language Models Fail in Healthcare: Evaluating Sensitivity to Prompt Variations
-
An Expanded Synthetic Conversation Dataset for Multi-Turn Smishing Detection
-
Beyond Rubrics: Exploration-Guided Evaluation Skills for Reward Modeling
-
HKVM-RAG: Key-Value-Separated Hypergraph Evidence Organization for Multi-Hop RAG
-
Seeing Without Exposing: Adaptive Privacy Control for Open-World, Context-Hungry MLLMs
-
OpenGlass: Open-Source Smart Glasses for On-Device Event-Based Gesture Recognition
-
Evidence Graph Consistency in Retrieval-Augmented Generation: A Model-Dependent Analysis of Hallucination Detection
-
Analysing Differences in Persuasive Language in LLM-Generated Text: Uncovering Stereotypical Gender Patterns
-
MCERF: Advancing Multimodal LLM Evaluation of Engineering Documentation with Enhanced Retrieval
-
GOPAgen: Motion-Aware and Efficient Agentic Long-Video Understanding with Structural Memory and Hierarchical Reasoning
-
EASE-TTT: Evidence-Aligned Selective Test-Time Training for Long-Context Question Answering
-
Act As a Real Researcher: A Suite of Benchmarks Evaluating Frontier LLMs and Agentic Harnesses in Research Lifecycle
-
LLM Agent-Assisted Reverse Engineering with Quantitative Readability Metrics
-
Evidence-Based Intelligent Diagnostic and Therapeutic Visualization System with Large Language Models: Multi-Turn Interaction and Multimodal Treatment Plan Generation
-
Are Large Language Models Suitable for Graph Computation? Progress and Prospects
-
Extracting Recurring Vulnerabilities from Black-Box LLM-Generated Software
-
Improving Cross-Lingual Factual Recall via Consistency-Driven Reinforcement Learning
-
Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents
-
A Wolf in Sheep's Clothing: Targeted Routing Hijacking in Federated RAG