Papers

New on arXiv

Sun 25 Jan 2026

171 papers

State Design Matters: How Representations Shape Dynamic Reasoning in Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Multi-source Heterogeneous Public Opinion Analysis via Collaborative Reasoning and Adaptive Fusion: A Systematically Integrated ApproachLLM reasoning → Audio-language model reasoning
Rethinking Soft Compression in Retrieval-Augmented Generation: A Query-Conditioned Selector PerspectiveLLM reasoning → Knowledge- and tool-augmented reasoning +1
Energy-Aware Reinforcement Learning for Robotic Manipulation of Articulated Components in Infrastructure Operation and MaintenanceDeep reinforcement learning methods → RL for robot locomotion and manipulation
R3G: A Reasoning-Retrieval-Reranking Framework for Vision-Centric Answer GenerationLLM reasoning → Multimodal reasoning
Autonomous Multi-Agent AI for High-Throughput Polymer Informatics: From Property Prediction to Generative Design Across Synthetic and Bio-PolymersLLM agents and tool use → Medical and scientific domain agents
The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context ReasoningLLM reasoning → Test-time compute and efficient thinking
Automated structural testing of LLM-based agents: methods, framework, and case studiesLLM agents and tool use → Software engineering coding agents
Differential Voting: Loss Functions For Axiomatically Diverse Aggregation of Heterogeneous PreferencesLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
RGFL: Reasoning Guided Fault Localization for Automated Program Repair Using Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
Sentipolis: Emotion-Aware Agents for Social SimulationsLLM agents and tool use → Social simulation and multi-agent debate
SD-E2: Semantic Exploration for Reasoning Under Token BudgetsLLM reasoning → Reinforcement learning for reasoning
Domain-Expert-Guided Hybrid Mixture-of-Experts for Medical AI: Integrating Data-Driven Learning with Clinical PriorsMixture-of-experts → Medical and biomedical MoE
Scaling Effects and Uncertainty Quantification in Neural Actor Critic AlgorithmsDeep reinforcement learning methods → RL theory and sample complexity
A Monosemantic Attribution Framework for Stable Interpretability in Clinical Neuroscience Transformer-Based Language ModelsMechanistic interpretability → Interpreting biomedical and science models
Agentic AI for Self-Driving Laboratories in Soft Matter: Taxonomy, Benchmarks,and Open ChallengesEnd-to-end autonomous driving → Large language models for driving +1
Benchmarking Direct Preference Optimization for Medical Large Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models +1
Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic DecodingEfficient LLM inference and serving → Diffusion LLM and MoE inference
Think Locally, Explain Globally: Graph-Guided LLM Investigations via Local Reasoning and Belief PropagationLLM reasoning → Reasoning for agents, planning and code +1
Prompt Injection Evaluations: Refusal Boundary Instability and Artifact-Dependent Compliance in GPT-4-Series ModelsLLM safety and security → Safety alignment, refusal and guardrails
Adaptive Weighting in Knowledge Distillation: An Axiomatic Framework for Multi-Scale Teacher Ensemble OptimizationKnowledge distillation → General knowledge distillation methods
When Personalization Legitimizes Risks: Uncovering Safety Vulnerabilities in Personalized Dialogue AgentsLLM agents and tool use → Agent memory systems
PEAfowl: Perception-Enhanced Multi-View Vision-Language-Action for Bimanual ManipulationVision-language-action models → 3D and spatial-aware VLA
SynMind: Reducing Semantic Hallucination in fMRI-Based Image ReconstructionDiffusion models → Medical imaging and segmentation
EFT-CoT: A Multi-Agent Chain-of-Thought Framework for Emotion-Focused TherapyLLM reasoning → Clinical and biomedical applications +1
An Effective and Cost-Efficient Agentic Framework for Ethereum Smart Contract AuditingLLM agents and tool use → Agent threats, guardrails and governance
SRA 2: Variational Autoencoder Self-Representation Alignment for Efficient Diffusion TrainingDiffusion models → Diffusion foundations and sampling
Linguistic and Argument Diversity in Synthetic Data for Function-Calling AgentsLLM agents and tool use → Tool learning and function calling
OwlerLite: Scope- and Freshness-Aware Web Retrieval for LLM AssistantsRetrieval-augmented generation → RAG pipeline design and evaluation
ViTCoP: Accelerating Large Vision-Language Models via Visual and Textual Semantic Collaborative PruningMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Multi-Agent Collaborative Intrusion Detection for Low-Altitude Economy IoT: An LLM-Enhanced Agentic AI FrameworkLLM agents and tool use → Agent threats, guardrails and governance
Less Is More: Scalable Visual Navigation from Limited DataOffline RL and imitation learning from data → Language-conditioned policies and VLAs
MMR-Bench: A Comprehensive Benchmark for Multimodal LLM RoutingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Beyond a Single Perspective: Text Anomaly Detection with Multi-View Language RepresentationsLLM safety and security → Safety alignment, refusal and guardrails
Multi-Agent End-to-End Vulnerability Management for Mitigating Recurring VulnerabilitiesLLM agents and tool use → Agent threats, guardrails and governance
HyperGraphPro: Progress-Aware Reinforcement Learning for Structure-Guided Hypergraph RAGLLM reasoning → Knowledge- and tool-augmented reasoning +2
Learning Sewing Patterns via Latent Flow Matching of Implicit FieldsDiffusion models → 3D, point cloud and scene generation
The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video GenerationLLM agents and tool use → Multimodal and embodied agents
ReFuGe: Feature Generation for Prediction Tasks on Relational Databases with LLM AgentsLLM agents and tool use → Web, search, and research agents
EntWorld: A Holistic Environment and Benchmark for Verifiable Enterprise GUI AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
A Survey on Evaluating Quality and Trustworthiness in LLM-Generated DataSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Do Reasoning Models Ask Better Questions? A Formal Information-Theoretic Analysis on Multi-Turn LLM GamesLLM reasoning → Reasoning for agents, planning and code +1
A Computational Approach to Visual MetonymyMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
SQL-Trail: Multi-Turn Reinforcement Learning with Interleaved Feedback for Text-to-SQLDeep reinforcement learning methods → RL training of language and agent models
LegalMALR:Multi-Agent Query Understanding and LLM-Based Reranking for Chinese Statute RetrievalRetrieval-augmented generation → Legal and financial RAG +1
Agentic reinforcement learning empowers next-generation chemical language models for molecular design and synthesisLLM agents and tool use → Medical and scientific domain agents
∞-MoE: Generalizing Mixture of Experts to Infinite ExpertsMixture-of-experts → MoE theory and gating functions
Align to the Pivot: Dual Alignment with Self-Feedback for Multilingual Math ReasoningLLM reasoning → Mathematical reasoning +1
Fast KVzip: Efficient and Accurate LLM Inference with Gated KV EvictionEfficient LLM inference and serving → KV cache optimization
AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and ThinkingLLM reasoning → Audio-language model reasoning
Do Multimodal RAG Systems Leak Data? A Comprehensive Evaluation of Membership Inference and Image Caption Retrieval AttacksRetrieval-augmented generation → Privacy and security of RAG +1
Health-ORSC-Bench: A Benchmark for Measuring Over-Refusal and Safety Completion in Health ContextLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Results-Actionability Gap: Understanding How Practitioners Evaluate LLM Products in the Wild
TabularMath: Evaluating Computational Extrapolation in Tabular Learning via Program-Verified Synthesis
IMU-1: Sample-Efficient Pre-training of Small Language Models
Scaled Dot-Product Attention implements projection of inputs onto a common surface
HYPE-EDIT-1: Benchmark for Measuring Reliability in Frontier Image Editing Models
RIR-Mega-Speech: A Reverberant Speech Corpus with Comprehensive Acoustic Metadata and Reproducible Evaluation
Time series forecasting with Hahn Kolmogorov-Arnold networks
CanaryBench: Stress Testing Privacy Leakage in Cluster-Level Conversation Summaries
Agentic Business Process Management Systems
How Much Temporal Modeling is Enough? A Systematic Study of Hybrid CNN-RNN Architectures for Multi-Label ECG Classification
CP Loss: Channel-wise Perceptual Loss for Time Series Forecasting
IPBC: An Interactive Projection-Based Framework for Human-in-the-Loop Semi-Supervised Clustering of High-Dimensional Data
VAE with Hyperspherical Coordinates: Improving Anomaly Detection from Hypervolume-Compressed Latent Space
Leveraging Persistence Image to Enhance Robustness and Performance in Curvilinear Structure Segmentation
SpatialEmb: Extract and Encode Spatial Information for 1-Stage Multi-channel Multi-speaker ASR on Arbitrary Microphone Arrays
An Experimental Comparison of Cognitive Forcing Functions for Execution Plans in AI-Assisted Writing: Effects On Trust, Overreliance, and Perceived Critical Thinking
Multimodal Machine Learning for Soft High-k Elastomers under Data Scarcity
Spelling Bee Embeddings for Language Modeling
CommonLID: Re-evaluating State-of-the-Art Language Identification Performance on Web Data
MlPET: A Localized Neural Network Approach for Probabilistic Post-Reconstruction PET Image Analysis Using Informed Priors
A System for Name and Address Parsing with Large Language Models
Evaluating Semantic and Syntactic Understanding in Large Language Models for Payroll Systems
Strip-Fusion: Spatiotemporal Fusion for Multispectral Pedestrian Detection
Memory-Efficient FPGA Implementation of Stochastic Simulated Annealing
PEAR: Pairwise Evaluation for Automatic Relative Scoring in Machine Translation
Flow-based Extremal Mathematical Structure Discovery
MorphXAI: An Explainable Framework for Morphological Analysis of Parasites in Blood Smear Images
Coding-Enforced Resilient and Secure Aggregation for Hierarchical Federated Learning
AI-based approach to burnout identification from textual data
A Cherry-Picking Approach to Large Load Shaping for More Effective Carbon Reduction
Systematic Characterization of Minimal Deep Learning Architectures: A Unified Analysis of Convergence, Pruning, and Quantization
Federated learning for unpaired multimodal data through a homogeneous transformer model
Boosting methods for interval-censored data with regression and classification
LLMs as Cultural Archives: Cultural Commonsense Knowledge Graph Extraction
TensorLens: End-to-End Transformer Analysis via High-Order Attention Tensors
UPLiFT: Efficient Pixel-Dense Feature Upsampling with Local Attenders
FlowMorph: Physics-Consistent Self-Supervision for Label-Free Single-Cell Mechanics in Microfluidic Videos
Credit Fairness: Online Fairness In Shared Resource Pools
LLM-Based SQL Generation: Prompting, Self-Refinement, and Adaptive Weighted Majority Voting
DTC: A Deformable Transposed Convolution Module for Medical Image Segmentation
FedGraph-VASP: Privacy-Preserving Federated Graph Learning with Post-Quantum Security for Cross-Institutional Anti-Money Laundering
From Specialist to Generalist: Unlocking SAM's Learning Potential on Unlabeled Medical Images
Dissipative Learning: A Framework for Viable Adaptive Systems
RemEdit: Efficient Diffusion Editing with Riemannian Geometry
Learning Transferable Skills in Action RPGs via Directed Skill Graphs and Selective Adaptation
ShapLoRA: Allocation of Low-rank Adaption on Large Language Models via Shapley Value Inspired Importance Estimation
UniPACT: A Multimodal Framework for Prognostic Question Answering on Raw ECG and Structured EHR
Causal Pre-training Under the Fairness Lens: An Empirical Study of TabPFN

All 171