Papers

New on arXiv

Fri 11 Jul 2025

288 papers

ToolRegistry: A Protocol-Agnostic Tool Management Library for Function-Calling LLMsLLM agents and tool use → Tool learning and function calling
ARPaCCino: An Agentic-RAG for Policy as Code ComplianceLLM agents and tool use → Agent threats, guardrails and governance
When and Where do Data Poisons Attack Textual Inversion?Diffusion models → Robustness, privacy and adversarial aspects
From Sequence to Structure: Uncovering Substructure Reasoning in TransformersLLM reasoning → Knowledge- and tool-augmented reasoning
OpenCodeReasoning-II: A Simple Test Time Scaling Approach via Self-CritiqueLLM reasoning → Reasoning for agents, planning and code
BlindSight: Harnessing Sparsity for Efficient Vision-Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency
SetupBench: Assessing Software Engineering Agents' Ability to Bootstrap Development EnvironmentsLLM agents and tool use → Software engineering coding agents +1
Action Chunking and Exploratory Data Collection Yield Exponential Improvements in Behavior Cloning for Continuous ControlOffline RL and imitation learning from data → Robot manipulation imitation learning
CALMA: A Process for Deriving Context-aligned Axes for Language Model AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Contrastive Conditional-Unconditional Alignment for Long-tailed Diffusion ModelDiffusion models → Diffusion foundations and sampling
On the Gradient Domination of the LQG ProblemDeep reinforcement learning methods → RL for games and combinatorial optimization
Accelerating Drug Discovery Through Agentic AI: A Multi-Agent Approach to Laboratory Automation in the DMTA CycleLLM agents and tool use → Medical and scientific domain agents
On Evaluating Performance of LLM Inference Serving SystemsEfficient LLM inference and serving → Serving systems and request scheduling
Enhancing RLHF with Human Gaze ModelingLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
VIP: Visual Information Protection through Adversarial Attacks on Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Learning Diffusion Models with Flexible Representation GuidanceDiffusion models → Diffusion foundations and sampling
PRISM: Reducing Spurious Implicit Biases in Vision-Language Models with LLM-Guided Embedding ProjectionMultimodal LLMs / vision-language models → Social bias and fairness in VLMs
Self-Improving Model SteeringSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
Improving Classifier-Free Guidance in Masked Diffusion: Low-Dim Theoretical Insights with High-Dim ImpactDiffusion models → Diffusion foundations and sampling
How to Train a Leader: Hierarchical Reasoning in Multi-Agent LLMsLLM reasoning → Reasoning for agents, planning and code +1
Bridging Literature and the Universe Via A Multi-Agent Large Language Model SystemLLM agents and tool use → Web, search, and research agents
Beyond Scores: Proximal Diffusion ModelsDiffusion models → Diffusion foundations and sampling
GraphRunner: A Multi-Stage Framework for Efficient and Accurate Graph-Based RetrievalRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Optimizing Sequential Multi-Step Tasks with Parallel LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
SEALGuard: Safeguarding the Multilingual Conversations in Southeast Asian Languages for LLM Software SystemsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
The Non-Linear Representation Dilemma: Is Causal Abstraction Enough for Mechanistic Interpretability?Mechanistic interpretability → Circuits and mechanistic analysis
Lumos-1: On Autoregressive Video Generation with Discrete Diffusion from a Unified Model PerspectiveDiffusion models → Video generation
KV Cache Steering for Controlling Frozen LLMsEfficient LLM inference and serving → Efficient reasoning and agent inference +1
One Token to Fool LLM-as-a-JudgeLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
Optimistic Exploration for Risk-Averse Constrained Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
From One to More: Contextual Part Latents for 3D GenerationDiffusion models → 3D, point cloud and scene generation
BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation SparsityMixture-of-experts → Expert routing strategies +1
Penalizing Infeasible Actions and Reward Scaling in Reinforcement Learning with Offline DataDeep reinforcement learning methods → Deep RL algorithms and representations
On the Effect of Regularization in Policy Mirror DescentDeep reinforcement learning methods → RL theory and sample complexity
SPLASH! Sample-efficient Preference-based inverse reinforcement learning for Long-horizon Adversarial tasks from Suboptimal Hierarchical demonstrationsOffline RL and imitation learning from data → Inverse reinforcement learning +1
elsciRL: Integrating Language Solutions into Reinforcement Learning Problem SettingsDeep reinforcement learning methods → RL training of language and agent models
ByDeWay: Boost Your multimodal LLM with DEpth prompting in a Training-Free WayMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Introspection of Thought Helps AI AgentsLLM reasoning → Reasoning for agents, planning and code +1
Safe Deep Reinforcement Learning for Resource Allocation with Peak Age of Information Violation GuaranteesDeep reinforcement learning methods → RL for energy, networks and operations
Leanabell-Prover-V2: Verifier-integrated Reasoning for Formal Theorem Proving via Reinforcement LearningLLM reasoning → Mathematical reasoning
DatasetAgent: A Novel Multi-Agent System for Auto-Constructing Datasets from Real-World ImagesLLM agents and tool use → Multimodal and embodied agents
Entangled Threats: A Unified Kill Chain Model for Quantum Machine Learning SecurityLLM safety and security → Machine unlearning
A comprehensive study of LLM-based argument classification: from LLAMA through GPT-4o to Deepseek-R1LLM reasoning → Reasoning for agents, planning and code
Agentic Large Language Models for Conceptual Systems Engineering and DesignLLM reasoning → Reasoning for agents, planning and code +1
Towards Collaborative Fairness in Federated Learning Under Imbalanced Covariate ShiftKnowledge distillation → Federated, robust and private distillation
AgentsNet: Coordination and Collaborative Reasoning in Multi-Agent LLMsLLM reasoning → Reasoning for agents, planning and code +1
BayesTTA: Continual-Temporal Test-Time Adaptation for Vision-Language Models via Gaussian Discriminant AnalysisMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
To Trade or Not to Trade: An Agentic Approach to Estimating Market Risk Improves Trading DecisionsLLM agents and tool use → Financial trading and forecasting agents
Large Multi-modal Model Cartographic Map Comprehension for Textual Locality GeoreferencingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
SAM2RL: Towards Reinforcement Learning Memory Control in Segment Anything Model 2Deep reinforcement learning methods → RL training of language and agent models
Recursive Reward AggregationDeep reinforcement learning methods → RL theory and sample complexity
Occlusion-Guided Feature Purification Learning via Reinforced Knowledge Distillation for Occluded Person Re-IdentificationKnowledge distillation → General knowledge distillation methods
Advancing Multimodal LLMs by Large-Scale 3D Visual Instruction Dataset GenerationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
SFedKD: Sequential Federated Learning with Discrepancy-Aware Multi-Teacher Knowledge DistillationKnowledge distillation → Federated, robust and private distillation
Efficient Deployment of Vision-Language Models on Mobile Devices: A Case Study on OnePlus 13RMultimodal LLMs / vision-language models → Visual token pruning and efficiency
From Language to Logic: A Bi-Level Framework for Structured ReasoningLLM reasoning → Reasoning benchmarks and evaluation
LLaPa: A Vision-Language Model Framework for Counterfactual-Aware Procedural PlanningLLM reasoning → Spatial and embodied reasoning +1
Evaluating SAE interpretability without explanationsMechanistic interpretability → SAE features and their applications
Clue-RAG: Towards Accurate and Cost-Efficient Graph-based RAG via Multi-Partite Graph and Query-Driven Iterative RetrievalRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
KGRAG-Ex: Explainable Retrieval-Augmented Generation with Knowledge Graph-based PerturbationsRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Finding Common Ground: Using Large Language Models to Detect Agreement in Multi-Agent Decision ConferencesLLM agents and tool use → Social simulation and multi-agent debate
xpSHACL: Explainable SHACL Validation using Retrieval-Augmented Generation and Large Language ModelsRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Training-free Mixed-Resolution Latent Upsampling for Spatially Accelerated Diffusion TransformersDiffusion models → Diffusion foundations and sampling
Multi-modal Mutual-Guidance Conditional Prompt Learning for Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Multi-Agent LLMs as Ethics Advocates for AI-Based SystemsLLM agents and tool use → Software engineering coding agents
Inference-Time Scaling of Diffusion Language Models via Trajectory RefinementLLM reasoning → Diffusion reasoning and test-time scaling +1
Online Pre-Training for Offline-to-Online Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
The Curious Case of Factuality Finetuning: Models' Internal Beliefs Can Improve FactualitySynthetic data and self-improvement for LLMs → Self-improving MLLMs and synthetic data
Understanding Driving Risks using Large Language Models: Toward Elderly Driver AssessmentMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
Intelligent Control of Spacecraft Reaction Wheel Attitude Using Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Exploring Design of Multi-Agent LLM Dialogues for Research IdeationLLM agents and tool use → Web, search, and research agents
What Factors Affect LLMs and RLLMs in Financial Question Answering?LLM reasoning → Reasoning for agents, planning and code
Distillation versus Contrastive Learning: How to Train Your RerankersKnowledge distillation → Knowledge distillation for LLMs
Token-Based Audio Inpainting via Discrete DiffusionDiffusion models → Audio, speech and music
CRMAgent: A Multi-Agent LLM System for E-Commerce CRM Message Template GenerationLLM agents and tool use → Social simulation and multi-agent debate
Improving MLLM's Document Image Machine Translation via Synchronously Self-reviewing Its OCR ProficiencyMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
M2-Reasoning: Empowering MLLMs with Unified General and Spatial ReasoningLLM reasoning → Spatial and embodied reasoning +1
KAT-V1: Kwai-AutoThink Technical ReportLLM reasoning → Test-time compute and efficient thinking +1
Invariant-based Robust Weights Watermark for Large Language ModelsLLM safety and security → Watermarking generated content and models
Lightweight Safety Guardrails via Synthetic Data and RL-guided Adversarial TrainingSynthetic data and self-improvement for LLMs → Safety of self-evolving systems +1
Agent Safety Alignment via Reinforcement LearningLLM agents and tool use → Agent threats, guardrails and governance +2
A Practical Two-Stage Recipe for Mathematical LLMs: Maximizing Accuracy with SFT and Efficiency with Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning
Abductive Computational Systems: Creative Abduction and Future DirectionsLLM reasoning → Reasoning benchmarks and evaluation
InsightBuild: LLM-Powered Causal Reasoning in Smart Building SystemsLLM reasoning → Causal reasoning in LLMs
Making VLMs More Robot-Friendly: Self-Critical Distillation of Low-Level Procedural ReasoningLLM reasoning → Spatial and embodied reasoning
Gearshift Fellowship: A Next-Generation Neurocomputational Game Platform to Model and Train Human-AI Adaptability
Page image classification for content-specific data processing
Building crypto portfolios with agentic AI
A Sparsity Predicting Approach for Large Language Models via Activation Pattern Clustering
Feature Bank Enhancement for Distance-based Out-of-Distribution Detection
Understanding Two-Layer Neural Networks with Smooth Activation Functions
Enhancing Breast Cancer Detection with Vision Transformers and Graph Neural Networks
Transformer-Based Framework for Motion Capture Denoising and Anomaly Detection in Medical Rehabilitation
H-NeiFi: Non-Invasive and Consensus-Efficient Multi-Agent Opinion Guidance
Fairness Is Not Enough: Auditing Competence and Intersectional Bias in AI-powered Resume Screening
MH-FSF: A Unified Framework for Overcoming Benchmarking and Reproducibility Limitations in Feature Selection Evaluation
Repairing Language Model Pipelines by Meta Self-Refining Competing Constraints at Runtime
Comparative Analysis of Vision Transformers and Traditional Deep Learning Approaches for Automated Pneumonia Detection in Chest X-Rays
Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing
AutoRAG-LoRA: Hallucination-Triggered Knowledge Retuning via Lightweight Adapters

All 288