Papers

New on arXiv

Tue 20 Jan 2026

371 papers

Agentic SPARQL: Evaluating SPARQL-MCP-powered Intelligent Agents on the Federated KGQA BenchmarkLLM agents and tool use → Web, search, and research agents
Evaluating Few-Shot Temporal Reasoning of LLMs for Human Activity Prediction in Smart EnvironmentsLLM reasoning → Spatial and embodied reasoning
Modality as Heterogeneity: Node Splitting and Graph Rewiring for Multimodal Graph LearningMixture-of-experts → Multimodal MoE models
Simple Role Assignment is Extraordinarily Effective for Safety AlignmentLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
TextBFGS: A Case-Based Reasoning Approach to Code Optimization via Error-Operator RetrievalLLM reasoning → Reasoning for agents, planning and code
Variational Quantum Circuit-Based Reinforcement Learning for Dynamic Portfolio OptimizationDeep reinforcement learning methods → Quantum reinforcement learning
AMVICC: A Novel Benchmark for Cross-Modal Failure Mode Profiling for VLMs and IGMsLLM reasoning → Multimodal reasoning +1
ToolCaching: Towards Efficient Caching for LLM Tool-callingLLM agents and tool use → Software engineering coding agents
Empowering LLMs for Structure-Based Drug Design via Exploration-Augmented Latent InferenceLLM reasoning → Clinical and biomedical applications
RECAP: A Resource-Efficient Method for Adversarial Prompting in Large Language ModelsLLM safety and security → Jailbreaking
Predicting Retrieval Utility and Answer Quality in Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
PAS-Mamba: Phase-Amplitude-Spatial State Space Model for MRI ReconstructionState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Business Logic-Driven Text-to-SQL Data Synthesis for Business IntelligenceSynthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping
"Just in Time" World Modeling Supports Human Planning and ReasoningWorld models → World models for language agents
Can LLM Reasoning Be Trusted? A Comparative Study: Using Human Benchmarking on Statistical TasksLLM reasoning → Reasoning benchmarks and evaluation
Tokenomics: Quantifying Where Tokens Are Used in Agentic Software EngineeringLLM agents and tool use → Software engineering coding agents
Gaussian Based Adaptive Multi-Modal 3D Semantic Occupancy PredictionState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
VisTIRA: Closing the Image-Text Modality Gap in Visual Math Reasoning via Structured Tool IntegrationLLM reasoning → Multimodal reasoning +1
Agentic AI Meets Edge Computing in Autonomous UAV SwarmsLLM agents and tool use → Multimodal and embodied agents
CMind: An AI Agent for Localizing C Memory BugsLLM agents and tool use → Software engineering coding agents
Cosmo-FOLD: Fast generation and upscaling of field-level cosmological maps with overlap latent diffusionDiffusion models → Inverse problems and image restoration
Single-step Controllable Music Bandwidth Extension With Flow MatchingDiffusion models → Audio, speech and music
VJEPA: Variational Joint Embedding Predictive Architectures as Probabilistic World ModelsWorld models → Latent predictive world models
RoboBrain 2.5: Depth in Sight, Time in MindLLM reasoning → Spatial and embodied reasoning
MARBLE: Multi-Agent Reasoning for Bioinformatics Learning and EvolutionLLM reasoning → Clinical and biomedical applications +1
Turn-Based Structural Triggers: Structure-Conditioned Backdoors in Multi-Turn LLMsLLM safety and security → Backdoor and data poisoning attacks
CityCube: Benchmarking Cross-view Spatial Reasoning on Vision-Language Models in Urban EnvironmentsLLM reasoning → Spatial and embodied reasoning +1
LURE: Latent Space Unblocking for Multi-Concept Reawakening in Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
Yuan3.0 Ultra: A Trillion-Parameter Enterprise-Oriented MoE LLMMixture-of-experts → MoE large language models
SilentDrift: Exploiting Action Chunking for Stealthy Backdoor Attacks on Vision-Language-Action ModelsVision-language-action models → Safety and adversarial attacks +1
LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCRMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative AlignmentLLM reasoning → Prompting and chain-of-thought reasoning +1
XR: Cross-Modal Agents for Composed Image RetrievalLLM reasoning → Multimodal reasoning +1
Jet-RL: Enabling On-Policy FP8 Reinforcement Learning with Unified Training and Rollout Precision FlowLLM reasoning → Reinforcement learning for reasoning +1
APEX-AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
Spatiotemporal Wildfire Prediction and Reinforcement Learning for Helitack SuppressionDeep reinforcement learning methods → RL for energy, networks and operations
Q-learning with Adjoint MatchingDiffusion models → Diffusion and flow policies for robotics +1
KAGE-Bench: Fast Known-Axis Visual Generalization Evaluation for Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
MASCOT: Multi-Agent Socio-Collaborative Companion SystemsLLM agents and tool use → Social simulation and multi-agent debate
Rerank Before You Reason: Analyzing Reranking Tradeoffs through Effective Token Cost in Deep Search AgentsLLM agents and tool use → Web, search, and research agents
InT: Self-Proposed Interventions Enable Credit Assignment in LLM ReasoningLLM reasoning → Reinforcement learning for reasoning
Toward Efficient Agents: Memory, Tool learning, and PlanningLLM agents and tool use → General LLM agent frameworks and benchmarks
IIR-VLM: In-Context Instance-level Recognition for Large Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Domain-Adaptation through Synthetic Data: Fine-Tuning Large Language Models for German LawSynthetic data and self-improvement for LLMs → LLM synthetic data generation
TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-TransformersMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Toward architecting self-coding information systemsLLM agents and tool use → Software engineering coding agents
The Side Effects of Being Smart: Safety Risks in MLLMs' Multi-Image ReasoningLLM reasoning → Reasoning for agents, planning and code +1
Style Transfer as Bias Mitigation: Diffusion Models for Synthetic Mental Health Text for ArabicDiffusion models → Image generation, text-to-image and editing
A Systematic Analysis of Chunking Strategies for Reliable Question AnsweringRetrieval-augmented generation → Multimodal RAG
When Backdoors Meet Partial Observability: Attacking Real-World Reinforcement LearningDeep reinforcement learning methods → Adversarial robustness and RL in finance
Optimizing Energy and Data Collection in UAV-aided IoT Networks using Attention-based Multi-Objective Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Zero-shot adaptable task planning for autonomous construction robots: a comparative study of lightweight single and multi-AI agent systemsLLM agents and tool use → General LLM agent frameworks and benchmarks
DermaBench: A Clinician-Annotated Benchmark Dataset for Dermatology Visual Question Answering and ReasoningLLM reasoning → Clinical and biomedical applications +1
XCR-Bench: Benchmarking Cross-Cultural Reasoning in LLMs via Culture-Specific Items and Hall's TriadLLM reasoning → Reasoning benchmarks and evaluation
POCI-Diff: 3D-Layout Guided Diffusion for Controllable Synthetic Surveillance Data GenerationDiffusion models → 3D, point cloud and scene generation
LLMOrbit: A Circular Taxonomy of Large Language Models -From Scaling Walls to Agentic AI SystemsLLM reasoning → Reasoning benchmarks and evaluation +2
Vision Also You Need: Navigating Out-of-Distribution Detection with Multimodal Large Language ModelMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Understanding Multilingualism in Mixture-of-Experts LLMs: Routing Mechanism, Expert Specialization, and Layerwise SteeringMixture-of-experts → Expert routing strategies
Weather-R1: Logically Consistent Reinforcement Fine-Tuning for Multimodal Reasoning in MeteorologyLLM reasoning → Multimodal reasoning +1
Top 10 Open Challenges Steering the Future of Diffusion Language Model and Its VariantsDiffusion models → Discrete and text diffusion
RM-Distiller: Exploiting Generative LLM for Reward Model DistillationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Likelihood-Separable Diffusion Inference for Multi-Image MRI Super-ResolutionDiffusion models → Medical imaging and segmentation
Numina-Lean-Agent: An Open and General Agentic Reasoning System for Formal MathematicsLLM agents and tool use → Educational agents and math problem solving
Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language ModelsMechanistic interpretability → Knowledge and representations inside LLMs
"The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation FrameworkLLM reasoning → Prompting and chain-of-thought reasoning
VirtualCrime: Evaluating the Criminal Potential and Agentic Behaviors of Large Language Models via Sandbox SimulationLLM agents and tool use → Agent threats, guardrails and governance +1
FantasyVLN: Unified Multimodal Chain-of-Thought Reasoning for Vision-Language NavigationLLM reasoning → Spatial and embodied reasoning
RL-BioAug: Label-Efficient Reinforcement Learning for Self-Supervised EEG Representation LearningDeep reinforcement learning methods → Deep RL algorithms and representations
Glance-or-Gaze: Incentivizing LMMs to Adaptively Focus Search via Reinforcement LearningMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
VulnResolver: A Hybrid Agent Framework for LLM-Based Automated Vulnerability Issue ResolutionLLM agents and tool use → Agent threats, guardrails and governance
Multi-Objective Hierarchical Optimization with Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Human Simulation Computation: A Human-Inspired Framework for Adaptive AI SystemsLLM reasoning → Reasoning benchmarks and evaluation
Revisiting Multi-Task Visual Representation LearningMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
LifeAgentBench: Benchmarking LLMs for Long-Horizon, Cross-Dimensional Lifestyle Health ReasoningLLM reasoning → Clinical and biomedical applications
Chain-of-Thought Compression Should Not Be Blind: V-Skip for Efficient Multimodal Reasoning via Dual-Path AnchoringLLM reasoning → Multimodal reasoning +2
Pedagogical Alignment for Vision-Language-Action Models: A Comprehensive Framework for Data, Architecture, and Evaluation in EducationVision-language-action models → Generalist VLA models and benchmarks
DisasterVQA: A Visual Question Answering Benchmark Dataset for Disaster ScenesMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
FutureOmni: Evaluating Future Forecasting from Omni-Modal Context for Multimodal LLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Channel Estimation in MIMO Systems Using Flow Matching ModelsDiffusion models → Diffusion for compression and wireless
Knowledge Graph-Assisted LLM Post-Training for Enhanced Legal ReasoningLLM reasoning → Reasoning for agents, planning and code +1
HoverAI: An Embodied Aerial Agent for Natural Human-Drone InteractionLLM agents and tool use → Multimodal and embodied agents
PREGEN: Uncovering Latent Thoughts in Composed Video RetrievalLLM reasoning → Multimodal reasoning
Principled Latent Diffusion for Graphs via Laplacian AutoencodersDiffusion models → Flow matching, time series and science
DARC: Decoupled Asymmetric Reasoning Curriculum for LLM EvolutionLLM reasoning → Reasoning for agents, planning and code
Finding RELIEF: Shaping Reasoning Behavior without Reasoning Supervision via Belief EngineeringLLM reasoning → Test-time compute and efficient thinking
Hearing Between the Lines: Unlocking the Reasoning Power of LLMs for Speech EvaluationLLM reasoning → Audio-language model reasoning
Breaking the Data Barrier in Learning Symbolic Computation: A Case Study on Variable Ordering Suggestion for Cylindrical Algebraic DecompositionLLM reasoning → Mathematical reasoning
OP-Bench: Benchmarking Over-Personalization for Memory-Augmented Personalized Conversational AgentsLLM agents and tool use → Agent memory systems
Hierarchical Long Video Understanding with Audiovisual Entity Cohesion and Agentic SearchMultimodal LLMs / vision-language models → Video understanding with MLLMs +1
Reasoning or Pattern Matching? Probing Large Vision-Language Models with Visual PuzzlesLLM reasoning → Multimodal reasoning +1
Dr. Assistant: Enhancing Clinical Diagnostic Inquiry via Structured Diagnostic Reasoning Data and Reinforcement LearningLLM reasoning → Clinical and biomedical applications
HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM InferenceEfficient LLM inference and serving → KV cache optimization
Dynamic Differential Linear Attention: Enhancing Linear Diffusion Transformer for High-Quality Image GenerationDiffusion models → Video generation
CodeContests-O: Powering LLMs via Feedback-Driven Iterative Test Case GenerationLLM reasoning → Reasoning for agents, planning and code
CommunityBench: Benchmarking Community-Level Alignment across Diverse Groups and TasksLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Reinforcement Learning for Opportunistic Routing in Software-Defined LEO-Terrestrial SystemsDeep reinforcement learning methods → RL for robot navigation and UAV control
Communication-Free Collective Navigation for a Swarm of UAVs via LiDAR-Based Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
TimeART: Towards Agentic Time Series Reasoning via Tool-AugmentationLLM reasoning → Time series reasoning +1
Sample Complexity of Average-Reward Q-Learning: From Single-agent to Federated Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
EGM: Efficient Visual Grounding Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning

All 371