Papers

New on arXiv

Tue 4 Nov 2025

332 papers

V-Agent: An Interactive Video Search System Using Vision-Language ModelsMultimodal LLMs / vision-language models → Video understanding with MLLMs +1
Evaluating Large Language Models for Workload Mapping and Scheduling in Heterogeneous HPC SystemsLLM reasoning → Test-time compute and efficient thinking
Mina: A Multilingual LLM-Powered Legal Assistant Agent for Bangladesh for Empowering Access to JusticeLLM agents and tool use → Social simulation and multi-agent debate
A Preliminary Study of RAG for Taiwanese Historical ArchivesRetrieval-augmented generation → RAG pipeline design and evaluation
Fine-Tuning Vision-Language Models for Multimodal Polymer Property PredictionMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Video Text Preservation with Synthetic Text-Rich VideosDiffusion models → Video generation
C3-Diff: Super-resolving Spatial Transcriptomics via Cross-modal Cross-content Contrastive Diffusion ModellingDiffusion models → Medical imaging and segmentation
In-Context Adaptation of VLMs for Few-Shot Cell Detection in Optical MicroscopyMultimodal LLMs / vision-language models → Medical vision-language models
M2S2L: Mamba-based Multi-Scale Spatial-temporal Learning for Video Anomaly DetectionState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
Lookahead Unmasking Elicits Accurate Decoding in Diffusion Language ModelsDiffusion models → Discrete and text diffusion
Effective Test-Time Scaling of Discrete Diffusion through Iterative RefinementLLM reasoning → Diffusion reasoning and test-time scaling +1
Diversified Flow Matching with Translation IdentifiabilityDiffusion models → Flow matching, time series and science
WorldPlanner: Monte Carlo Tree Search and MPC with Action-Conditioned Visual World ModelsWorld models → Embodied and robot world models
A Collaborative Reasoning Framework for Anomaly Diagnostics in Underwater RoboticsLLM reasoning → Spatial and embodied reasoning
No-Human in the Loop: Agentic Evaluation at Scale for RecommendationLLM agents and tool use → Agentic recommender systems
PublicAgent: Multi-Agent Design Principles From an LLM-Based Open Data Analysis FrameworkLLM agents and tool use → Web, search, and research agents
SLIP: Structural-aware Language-Image Pretraining for Vision-Language AlignmentMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Discrete Bayesian Sample Inference for Graph GenerationDiffusion models → Flow matching, time series and science
Error-Aware Knowledge Distillation via Targeted Revision for Customer-Service SummarizationKnowledge distillation → Knowledge distillation for LLMs
Evaluating Control Protocols for Untrusted AI AgentsLLM safety and security → Prompt injection and agent security
SCALE-VLP: Soft-Weighted Contrastive Volumetric Vision-Language Pre-training with Spatial-Knowledge SemanticsMultimodal LLMs / vision-language models → Medical vision-language models
Scalable Single-Cell Gene Expression Generation with Latent Diffusion ModelsDiffusion models → Flow matching, time series and science
Value of Information-Enhanced Exploration in Bootstrapped DQNDeep reinforcement learning methods → Deep RL algorithms and representations
Cache Mechanism for Agent RAG SystemsRetrieval-augmented generation → Agentic RAG, agents and memory +1
Adaptive and Robust Data Poisoning Detection and Sanitization in Wearable IoT Systems using Large Language ModelsLLM safety and security → Backdoor and data poisoning attacks
Agent-Omni: Test-Time Multimodal Reasoning via Model Coordination for Understanding AnythingLLM reasoning → Multimodal reasoning +2
In Good GRACEs: Principled Teacher Selection for Knowledge DistillationKnowledge distillation → General knowledge distillation methods
Kosmos: An AI Scientist for Autonomous DiscoveryWorld models → World models for language agents
MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement LearningLLM reasoning → Reasoning for agents, planning and code
When One Modality Sabotages the Others: A Diagnostic Lens on Multimodal ReasoningLLM reasoning → Multimodal reasoning +1
Diffusion Models are Robust PretrainersDiffusion models → Robustness, privacy and adversarial aspects
PoCo: Agentic Proof-of-Concept Exploit Generation for Smart ContractsLLM agents and tool use → Agent threats, guardrails and governance
When Visualizing is the First Step to Reasoning: MIRA, a Benchmark for Visual Chain-of-ThoughtLLM reasoning → Multimodal reasoning
XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion RepresentationsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Controlling Performance and Budget of a Centralized Multi-agent LLM System with Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents
Using Span Queries to Optimize for Cache and Attention LocalityLLM reasoning → Test-time compute and efficient thinking +1
Agentic World Modeling for 6G: Near-Real-Time Generative State-Space ReasoningWorld models → World models for language agents +2
CostBench: Evaluating Multi-Turn Cost-Optimal Planning and Adaptation in Dynamic Environments for LLM Tool-Use AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
VidEmo: Affective-Tree Reasoning for Emotion-Centric Video Foundation ModelsLLM reasoning → Audio-language model reasoning
Policy Gradient Methods for Information-Theoretic Opacity in Markov Decision ProcessesDeep reinforcement learning methods → RL theory and sample complexity
Curriculum Design for Trajectory-Constrained Agent: Compressing Chain-of-Thought Tokens in LLMsLLM reasoning → Reasoning for agents, planning and code +1
The Collaboration Gap: Exploration and Benchmarking of Open-World Agentic CooperationLLM agents and tool use → General LLM agent frameworks and benchmarks
Optimal Singular Damage: Efficient LLM Inference in Low Storage RegimesEfficient LLM inference and serving → Serving systems and request scheduling
Apriel-H1: Towards Efficient Enterprise Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking +1
Can Visual Input Be Compressed? A Visual Token Compression Benchmark for Large Multimodal ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
Natural-gas storage modelling by deep reinforcement learningDeep reinforcement learning methods → RL for energy, networks and operations
DecompSR: A dataset for decomposed analyses of compositional multihop spatial reasoningLLM reasoning → Spatial and embodied reasoning
The Realignment Problem: When Right becomes Wrong in LLMsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
UniChange: Unifying Change Detection with Multimodal Large Language ModelMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Adaptive GR(1) Specification Repair for Liveness-Preserving Shielding in Reinforcement LearningDeep reinforcement learning methods → Safe and constrained RL
CGES: Confidence-Guided Early Stopping for Efficient and Accurate Self-ConsistencyLLM reasoning → Test-time compute and efficient thinking
The ORCA Benchmark: Evaluating Real-World Calculation Accuracy in Large Language ModelsLLM reasoning → Mathematical reasoning
TAUE: Training-free Noise Transplant and Cultivation Diffusion ModelDiffusion models → Image generation, text-to-image and editing
Directional-Clamp PPODeep reinforcement learning methods → RL theory and sample complexity
Adaptive Neighborhood-Constrained Q Learning for Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
Agentic AI for Mobile Network RAN Management and OptimizationLLM reasoning → Reasoning for agents, planning and code +1
Adapting General-Purpose Foundation Models for X-ray Ptychography in Low-Data RegimesMultimodal LLMs / vision-language models → Medical vision-language models
BRAINS: A Retrieval-Augmented System for Alzheimer's Detection and MonitoringRetrieval-augmented generation → Medical and biomedical RAG
Lost in Code Generation: Reimagining the Role of Software Models in AI-driven Software EngineeringLLM agents and tool use → Software engineering coding agents
Modeling Hawkish-Dovish Latent Beliefs in Multi-Agent Debate-Based LLMs for Monetary Policy Decision ClassificationLLM agents and tool use → Financial trading and forecasting agents
Extending RLVR to Open-Ended Tasks via Verifiable Multiple-Choice ReformulationLLM reasoning → Reinforcement learning for reasoning
Improving DF-Conformer Using Hydra For High-Fidelity Generative Speech Enhancement on Discrete Codec TokenState-space sequence models (S4, Mamba) → Speech and audio SSMs
Merging Continual Pretraining Models for Domain-Specialized LLMs: A Case Study in FinanceLLM reasoning → Reasoning for agents, planning and code
Improving Unlearning with Model Updates Probably Aligned with GradientsLLM safety and security → Machine unlearning
From the Laboratory to Real-World Application: Evaluating Zero-Shot Scene Interpretation on Edge Devices for Mobile RoboticsLLM reasoning → Spatial and embodied reasoning
ReAcTree: Hierarchical LLM Agent Trees with Control Flow for Long-Horizon Task PlanningLLM agents and tool use → General LLM agent frameworks and benchmarks
Wasserstein Convergence of Critically Damped Langevin DiffusionsDiffusion models → Diffusion foundations and sampling
ChartM3: A Multi-Stage Code-Driven Pipeline for Constructing Multi-Dimensional and Multi-Step Visual Reasoning Data in Chart ComprehensionLLM reasoning → Multimodal reasoning +1
IllumFlow: Illumination-Adaptive Low-Light Enhancement via Conditional Rectified Flow and Retinex DecompositionDiffusion models → Inverse problems and image restoration
Towards Iterative End-to-End Software Development: A Feature-Driven Multi-Agent FrameworkLLM agents and tool use → Software engineering coding agents
AutoAdv: Automated Adversarial Prompting for Multi-Turn Jailbreaking of Large Language ModelsLLM safety and security → Jailbreaking
LUMA-RAG: Lifelong Multimodal Agents with Provably Stable Streaming AlignmentRetrieval-augmented generation → Multimodal RAG +1
The Pervasive Blind Spot: Benchmarking VLM Inference Risks on Everyday Personal VideosMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
LaRe: Latent Refocusing for Multimodal ReasoningLLM reasoning → Multimodal reasoning
ASTRA: An Automated Framework for Strategy Discovery, Retrieval, and Evolution for Jailbreaking LLMsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
SWE-Sharp-Bench: A Reproducible Benchmark for C# Software Engineering TasksLLM agents and tool use → Software engineering coding agents
LTD-Bench: Evaluating Large Language Models by Letting Them DrawLLM reasoning → Spatial and embodied reasoning
RoME: Domain-Robust Mixture-of-Experts for MILP Solution Prediction across DomainsMixture-of-experts → Expert routing strategies
The Sequential Edge: Inverse-Entropy Voting Beats Parallel Self-Consistency at Matched ComputeLLM reasoning → Test-time compute and efficient thinking
Unlocking the Power of Multi-Agent LLM for Reasoning: From Lazy Agents to DeliberationLLM reasoning → Reasoning for agents, planning and code +1
FP8-Flow-MoE: A Casting-Free FP8 Recipe without Double Quantization ErrorMixture-of-experts → Efficient MoE training and inference systems
Reinforcement learning based data assimilation for unknown state modelDeep reinforcement learning methods → RL theory and sample complexity
VFocus: Better Verilog Generation from Large Language Model via Focused ReasoningLLM reasoning → Reasoning for agents, planning and code
SAIL-RL: Guiding MLLMs in When and How to Think via Dual-Reward RL TuningLLM reasoning → Reinforcement learning for reasoning +1
From Models to Operators: Rethinking Autoscaling Granularity for Large Generative ModelsLLM reasoning → Diffusion reasoning and test-time scaling
When Modalities Conflict: How Unimodal Reasoning Uncertainty Governs Preference Dynamics in MLLMsLLM reasoning → Multimodal reasoning +1
LACY: A Vision-Language Model-based Language-Action Cycle for Self-Improving Robotic ManipulationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Deep Ideation: Designing LLM Agents to Generate Novel Research Ideas on Scientific Concept NetworkLLM agents and tool use → Web, search, and research agents
Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without RetrainingMixture-of-experts → Expert routing strategies
An Evaluation of Interleaved Instruction Tuning on Semantic Reasoning Performance in an Audio MLLMLLM reasoning → Audio-language model reasoning +1
Continuum: Efficient and Robust Multi-Turn LLM Agent Scheduling with KV Cache Time-to-LiveLLM agents and tool use → Software engineering coding agents +1
Learning Interactive World Model for Object-Centric Reinforcement LearningWorld models → Latent predictive world models +1
TabDSR: Decompose, Sanitize, and Reason for Complex Numerical Reasoning in Tabular DataLLM reasoning → Knowledge- and tool-augmented reasoning
Optimizing Multi-Lane Intersection Performance in Mixed Autonomy EnvironmentsDeep reinforcement learning methods → RL for autonomous driving and traffic
Adaptive Cooperative Transmission Design for Ultra-Reliable Low-Latency Communications via Deep Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
Training Proactive and Personalized LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
Optimal-Agent-Selection: State-Aware Routing Framework for Efficient Multi-Agent CollaborationLLM agents and tool use → General LLM agent frameworks and benchmarks
Open the Oyster: Empirical Evaluation and Improvement of Code Reasoning Confidence in LLMsLLM reasoning → Reasoning for agents, planning and code
Personalized Decision Modeling: Utility Optimization or Textualized-Symbolic ReasoningLLM reasoning → Reasoning benchmarks and evaluation
MM-UNet: Morph Mamba U-shaped Convolutional Networks for Retinal Vessel SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis

All 332