Papers

New on arXiv

Fri 23 Jan 2026

297 papers

Talking to Yourself: Defying Forgetting in Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Do Personality Traits Interfere? Geometric Limitations of Steering in Large Language ModelsMechanistic interpretability → Activation steering methods
Scaling the Scaling Logic: Agentic Meta-Synthesis of Logic ReasoningLLM reasoning → Reasoning for agents, planning and code +1
PENGUIN: General Vital Sign Reconstruction from PPG with Flow Matching State Space ModelDiffusion models → Medical imaging and segmentation
Beyond Translation: Cross-Cultural Meme Transcreation with Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
EDU-CIRCUIT-HW: Evaluating Multimodal Large Language Models on Real-World University-Level STEM Student Handwritten SolutionsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
ECCO: Evidence-Driven Causal Reasoning for Compiler OptimizationLLM reasoning → Reasoning for agents, planning and code
Semi-Supervised Domain Adaptation with Latent Diffusion for Pathology Image ClassificationDiffusion models → Medical imaging and segmentation
Retell, Reward, Repeat: Reinforcement Learning for Narrative Theory-Informed Story RetellingLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment +1
Parameter Inference and Uncertainty Quantification with Diffusion Models: Extending CDI to 2D Spatial ConditioningDiffusion models → Inverse problems and image restoration
Beyond Outcome Verification: Verifiable Process Reward Models for Structured ReasoningLLM reasoning → Reinforcement learning for reasoning +1
DF-RAG: Query-Aware Diversity for Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
Reasoning Beyond Literal: Cross-style Multimodal Reasoning for Figurative Language UnderstandingLLM reasoning → Multimodal reasoning +1
Implementing Tensor Logic: Unifying Datalog and Neural Reasoning via Tensor ContractionLLM reasoning → Reasoning benchmarks and evaluation
TrojanGYM: A Detector-in-the-Loop LLM for Adaptive RTL Hardware Trojan InsertionLLM agents and tool use → Agent threats, guardrails and governance
Interpreting Agentic Systems: Beyond Model Explanations to System-Level AccountabilityLLM agents and tool use → General LLM agent frameworks and benchmarks
Dynamic Role Assignment for Multi-Agent DebateMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Scaling medical imaging report generation with multimodal reinforcement learningDeep reinforcement learning methods → RL for clinical treatment and surgery
ConceptACT: Episode-Level Concepts for Sample-Efficient Robotic Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
iFSQ: Improving FSQ for Image Generation with 1 Line of CodeDiffusion models → Diffusion foundations and sampling
Acoustic Field Video for Multimodal Scene UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-ExpertsMixture-of-experts → Efficient MoE training and inference systems
Hybrid Mamba-Attention Neural Architecture for Channel EstimationState-space sequence models (S4, Mamba) → Core SSM architectures for language
Beyond Instrumental and Substitutive Paradigms: Introducing Machine Culture as an Emergent Phenomenon in Large Language ModelsLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
The Mouth is Not the Brain: Bridging Energy-Based World Models and Language GenerationWorld models → World models for language agents
GRASP: Guided Region-Aware Sparse Prompting for Adapting MLLMs to Remote SensingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic EvaluationsLLM agents and tool use → Social simulation and multi-agent debate
Do VLMs Have a Moral Backbone? A Study on the Fragile Morality of Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
VisGym: Diverse, Customizable, Scalable Environments for Multimodal AgentsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Auto-Regressive Masked Diffusion ModelsDiffusion models → Discrete and text diffusion
Spatial-Agent: Agentic Geo-spatial Reasoning with Scientific Core ConceptsLLM reasoning → Spatial and embodied reasoning +1
AgentDrive: An Open Benchmark Dataset for Agentic AI Reasoning with LLM-Generated Scenarios in Autonomous SystemsLLM reasoning → Chain-of-thought reasoning for driving +1
The Trajectory Alignment Coefficient in Two Acts: From Reward Tuning to Reward LearningLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
GRIP: Algorithm-Agnostic Machine Unlearning for Mixture-of-Experts via Geometric Router ConstraintsMixture-of-experts → Expert routing strategies +1
Evaluating Large Vision-language Models for Surgical Tool DetectionMultimodal LLMs / vision-language models → Medical vision-language models
From Atom to Community: Structured and Evolving Agent Memory for User Behavior ModelingLLM agents and tool use → Agentic recommender systems
Boosting Deep Reinforcement Learning with Semantic Knowledge for Robotic ManipulatorsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Mixture-of-Models: Unifying Heterogeneous Agents via N-Way Self-Evaluating DeliberationMixture-of-experts → MoE in reinforcement learning and robotics
Reasoning Promotes Robustness in Theory of Mind TasksLLM reasoning → Reasoning benchmarks and evaluation
AI builds, We Analyze: An Empirical Study of AI-Generated Build Code QualityLLM agents and tool use → Software engineering coding agents
Disentangling generalization and memorization in large language models using chessLLM reasoning → Reasoning for agents, planning and code
GTA: Generative Traffic Agents for Simulating Realistic Mobility BehaviorLLM agents and tool use → Multimodal and embodied agents
Flow Matching for Probabilistic Monocular 3D Human Pose EstimationDiffusion models → 3D, point cloud and scene generation
SWE-Pruner: Self-Adaptive Context Pruning for Coding AgentsLLM agents and tool use → Software engineering coding agents +1
LongCat-Flash-Thinking-2601 Technical ReportLLM reasoning → Reasoning for agents, planning and code +2
EMemBench: Interactive Benchmarking of Episodic Memory for VLM AgentsLLM agents and tool use → Agent memory systems
Adaptive Reinforcement and Model Predictive Control Switching for Safe Human-Robot Cooperative NavigationDeep reinforcement learning methods → RL for robot navigation and UAV control
AgentsEval: Clinically Faithful Evaluation of Medical Imaging Reports via Multi-Agent ReasoningLLM reasoning → Clinical and biomedical applications +1
Sim-to-Real Transfer via a Style-Identified Cycle Consistent Generative Adversarial Network: Zero-Shot Deployment on Robotic Manipulators through Visual Domain AdaptationDeep reinforcement learning methods → RL for robot locomotion and manipulation
ReViP: Mitigating False Completion in Vision-Language-Action Models with Vision-Proprioception RebalanceVision-language-action models → Failure recovery and robustness
Fast, faithful and photorealistic diffusion-based image super-resolution with enhanced Flow Map modelsDiffusion models → Inverse problems and image restoration +1
LUMINA: Long-horizon Understanding for Multi-turn Interactive AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
A Cognitive Framework for Autonomous Agents: Toward Human-Inspired DesignDeep reinforcement learning methods → Deep RL algorithms and representations
Edge-Aware Image Manipulation via Diffusion Models with a Novel Structure-Preservation LossDiffusion models → Image generation, text-to-image and editing
PanopMamba: Vision State Space Modeling for Nuclei Panoptic SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Omni-directional attention mechanism based on Mamba for speech separationState-space sequence models (S4, Mamba) → Speech and audio SSMs
Attention-MoA: Enhancing Mixture-of-Agents via Inter-Agent Semantic Attention and Deep Residual SynthesisLLM agents and tool use → General LLM agent frameworks and benchmarks
LLM is Not All You Need: A Systematic Evaluation of ML vs. Foundation Models for text and image based Medical ClassificationMultimodal LLMs / vision-language models → Medical vision-language models
CORD: Bridging the Audio-Text Reasoning Gap via Weighted On-policy Cross-modal DistillationLLM reasoning → Audio-language model reasoning
OnlineSI: Taming Large Language Model for Online 3D Understanding and GroundingLLM reasoning → Spatial and embodied reasoning +1
Curate-Train-Refine: A Closed-Loop Agentic Framework for Zero Shot ClassificationLLM agents and tool use → General LLM agent frameworks and benchmarks
Beyond Superficial Unlearning: Sharpness-Aware Robust Erasure of Hallucinations in Multimodal LLMsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
TangramPuzzle: Evaluating Multimodal Large Language Models with Compositional Spatial ReasoningLLM reasoning → Spatial and embodied reasoning +1
Rethinking Large Language Models For Irregular Time Series Classification In Critical CareLLM reasoning → Clinical and biomedical applications
SALAD: Achieve High-Sparsity Attention via Efficient Linear Attention Tuning for Video Diffusion TransformerDiffusion models → Video generation
SafeThinker: Reasoning about Risk to Deepen Safety Beyond Shallow AlignmentLLM reasoning → Reasoning for agents, planning and code +2
LOGICAL-COMMONSENSEQA: A Benchmark for Logical Commonsense ReasoningLLM reasoning → Reasoning benchmarks and evaluation
MRAG: Benchmarking Retrieval-Augmented Generation for Bio-medicineRetrieval-augmented generation → Medical and biomedical RAG
EvoConfig: Self-Evolving Multi-Agent Systems for Efficient Autonomous Environment ConfigurationLLM agents and tool use → Software engineering coding agents +1
Timely Machine: Awareness of Time Makes Test-Time Scaling AgenticLLM reasoning → Reasoning for agents, planning and code +1
FlowSE-GRPO: Training Flow Matching Speech Enhancement via Online Reinforcement LearningDiffusion models → Audio, speech and music
TL-GRPO: Turn-Level RL for Reasoning-Guided Iterative OptimizationLLM reasoning → Reinforcement learning for reasoning
DeepEra: A Deep Evidence Reranking Agent for Scientific Retrieval-Augmented Generated Question AnsweringRetrieval-augmented generation → Agentic RAG, agents and memory +1
DeMark: A Query-Free Black-Box Attack on Deepfake Watermarking DefensesLLM safety and security → Watermarking generated content and models
Persona Jailbreaking in Large Language ModelsLLM safety and security → Jailbreaking
Finding What Matters: Anchoring Context Knowledge with Evolving Indices for Iterative RetrievalRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Mixing Expert Knowledge: Bring Human Thoughts Back To the Game of GoLLM reasoning → Reasoning for agents, planning and code
Endless Terminals: Scaling RL Environments for Terminal AgentsSynthetic data and self-improvement for LLMs → Self-evolving LLM agents
Learning Domain Knowledge in Multimodal Large Language Models through Reinforcement Fine-TuningMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Reinforcement Learning-Based Energy-Aware Coverage Path Planning for Precision AgricultureDeep reinforcement learning methods → RL for energy, networks and operations
Towards a Theoretical Understanding to the Generalization of RLHFLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Clarify or Answer: Reinforcement Learning for Agentic VQA with Context Under-specificationLLM agents and tool use → Multimodal and embodied agents
White-Box Sensitivity Auditing with Steering VectorsMechanistic interpretability → Probing deception, bias and beliefs
From Attribution to Abstention: Training-Free Attention-Based Auditing for Clinical SummarizationMultimodal LLMs / vision-language models → Medical vision-language models
ResAgent: Entropy-based Prior Point Discovery and Visual Reasoning for Referring Expression SegmentationLLM reasoning → Multimodal reasoning
Toward Agentic Software Project Management: A Vision and RoadmapLLM agents and tool use → Software engineering coding agents
Cross-Lingual Activation Steering for Multilingual Language ModelsMechanistic interpretability → Activation steering methods
Cognitively-Inspired Tokens Overcome Egocentric Bias in Multimodal ModelsLLM reasoning → Spatial and embodied reasoning +1
Prime Successor Irreducibility: Turing Machine Complexity, Kolmogorov Complexity, and Weakness-Based Formulations
Autonomous Editorial Systems and Computational Investigation with Artificial Intelligence
Three-dimensional Damage Visualization of Civil Structures via Gaussian Splatting-enabled Digital Twins
Narrative Theory-Driven LLM Methods for Automatic Story Generation and Understanding: A Survey
Large Language Models for Assisting American College Applications
IntelliAsk: Learning to Ask High-Quality Research Questions via RLVR
Can LLMs Assess Personality? Validating Conversational AI for Trait Profiling
Gated Tree Cross-Attention for Checkpoint-Compatible Syntax Injection in Decoder-Only LLMs
VeRA: Renewing Reasoning Benchmarks with Executable Specifications
The Turing Synthetic Radar Dataset: A dataset for pulse deinterleaving
Majorization-Minimization Networks for Inverse Problems: An Application to EEG Imaging
Measuring Individual User Fairness with User Similarity and Effectiveness Disparity

All 297