Papers

New on arXiv

Mon 7 Jul 2025

371 papers

Exploring Agentic Artificial Intelligence Systems: Towards a Typological FrameworkLLM agents and tool use → General LLM agent frameworks and benchmarks
A Query-Aware Multi-Path Knowledge Graph Fusion Approach for Enhancing Retrieval-Augmented Generation in Large Language ModelsRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
wd1: Weighted Policy Optimization for Reasoning in Diffusion Language ModelsLLM reasoning → Diffusion reasoning and test-time scaling +1
Helix Parallelism: Rethinking Sharding Strategies for Interactive Multi-Million-Token LLM DecodingEfficient LLM inference and serving → KV cache optimization
Towards Solving More Challenging IMO Problems via Decoupled Reasoning and ProvingLLM reasoning → Mathematical reasoning
SPARC: Concept-Aligned Sparse Autoencoders for Cross-Model and Cross-Modal InterpretabilityMechanistic interpretability → Interpreting vision and multimodal models
Red Teaming AI Red TeamingLLM safety and security → Prompt injection and agent security
Conversational Education at Scale: A Multi-LLM Agent Workflow for Procedural Learning and Pedagogic Quality AssessmentLLM agents and tool use → Educational agents and math problem solving
Architecting Clinical Collaboration: Multi-Agent Reasoning Systems for Multimodal Medical VQALLM reasoning → Clinical and biomedical applications +1
LEGO Co-builder: Exploring Fine-Grained Vision-Language Modeling for Multimodal LEGO Assembly AssistantsLLM reasoning → Spatial and embodied reasoning +1
Cloud Diffusion Part 1: Theory and MotivationDiffusion models → Diffusion foundations and sampling
Deep Research Comparator: A Platform For Fine-grained Human Annotations of Deep Research AgentsLLM agents and tool use → Web, search, and research agents
Epistemically-guided forward-backward explorationDeep reinforcement learning methods → Deep RL algorithms and representations
2048: Reinforcement Learning in a Delayed Reward EnvironmentDeep reinforcement learning methods → Deep RL algorithms and representations
A Systematization of Security Vulnerabilities in Computer Use AgentsLLM agents and tool use → Agent threats, guardrails and governance
"Lost-in-the-Later": Framework for Quantifying Contextual Grounding in Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Learn Globally, Speak Locally: Bridging the Gaps in Multilingual ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Neural-Driven Image EditingDiffusion models → Medical imaging and segmentation
pFedMMA: Personalized Federated Fine-Tuning with Multi-Modal Adapter for Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
On the Bias of Next-Token Predictors Toward Systematically Inefficient Reasoning: A Shortest-Path Case StudyLLM reasoning → Test-time compute and efficient thinking
A Careful Examination of Large Behavior Models for Multitask Dexterous ManipulationOffline RL and imitation learning from data → Robot manipulation imitation learning
MindFlow: Revolutionizing E-commerce Customer Support with Multimodal LLM AgentsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
Going Beyond Heuristics by Imposing Policy Improvement as a ConstraintDeep reinforcement learning methods → Deep RL algorithms and representations
AGACCI : Affiliated Grading Agents for Criteria-Centric Interface in Educational Coding ContextsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
Beyond Simple Edits: X-Planner for Complex Instruction-Based Image EditingDiffusion models → Image generation, text-to-image and editing
Spatio-Temporal LLM: Reasoning about Environments and ActionsLLM reasoning → Spatial and embodied reasoning +1
Evaluating Memory in LLM Agents via Incremental Multi-Turn InteractionsLLM agents and tool use → Agent memory systems
Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual ReasoningLLM reasoning → Multimodal reasoning
Action Space Reduction Strategies for Reinforcement Learning in Autonomous DrivingEnd-to-end autonomous driving → Reinforcement learning for driving +1
Response Attack: Exploiting Contextual Priming to Jailbreak Large Language ModelsLLM safety and security → Jailbreaking
When Chain of Thought is Necessary, Language Models Struggle to Evade MonitorsLLM reasoning → Prompting and chain-of-thought reasoning
StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context ModelingEfficient LLM inference and serving → Visual token pruning in multimodal models
NavigScene: Bridging Local Perception and Global Navigation for Beyond-Visual-Range Autonomous DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
EmbodieDreamer: Advancing Real2Sim2Real Transfer for Policy Training via Embodied World ModelingWorld models → Embodied and robot world models
Pre-Trained Policy Discriminators are General Reward ModelsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
QMoE: A Quantum Mixture of Experts Framework for Scalable Quantum Neural NetworksMixture-of-experts → MoE theory and gating functions
From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity ExplanationsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
CREW-WILDFIRE: Benchmarking Agentic Multi-Agent Collaborations at ScaleLLM agents and tool use → General LLM agent frameworks and benchmarks
Critique of World ModelWorld models → World models for language agents
SV-DRR: High-Fidelity Novel View X-Ray Synthesis Using Diffusion ModelDiffusion models → Medical imaging and segmentation
VERITAS: Verification and Explanation of Realness in Images for Transparency in AI SystemsDiffusion models → Image generation, text-to-image and editing
SMART: Simulated Students Aligned with Item Response Theory for Question Difficulty PredictionLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
VOTE: Vision-Language-Action Optimization with Trajectory Ensemble VotingVision-language-action models → Generalist VLA models and benchmarks
CLIP-Guided Backdoor Defense through Entropy-Based Poisoned Dataset SeparationLLM safety and security → Backdoor and data poisoning attacks
The Hidden Threat in Plain Text: Attacking RAG Data LoadersRetrieval-augmented generation → Privacy and security of RAG
MoDiT: Learning Highly Consistent 3D Motion Coefficients with Diffusion Transformer for Talking Head GenerationDiffusion models → Audio, speech and music
ICAS: Detecting Training Data from Autoregressive Image Generative ModelsLLM safety and security → Membership inference attacks
Replacing thinking with tool usage enables reasoning in small language modelsLLM reasoning → Reasoning for agents, planning and code
INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance SamplingMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
MoLink: Distributed and Efficient Serving Framework for Large ModelsEfficient LLM inference and serving → Serving systems and request scheduling
A Generative Diffusion Model for Amorphous MaterialsDiffusion models → Flow matching, time series and science
DARIL: When Imitation Learning outperforms Reinforcement Learning in Surgical Action PlanningWorld models → Medical and surgical world models +2
TLB-VFI: Temporal-Aware Latent Brownian Bridge Diffusion for Video Frame InterpolationDiffusion models → Video generation
Can Video LLMs Refuse to Answer? Alignment for Answerability in Video Large Language ModelsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Taming the Tri-Space Tension: ARC-Guided Hallucination Modeling and Control for Text-to-Image GenerationDiffusion models → Image generation, text-to-image and editing
ReLoop: "Seeing Twice and Thinking Backwards" via Closed-loop Training to Mitigate Hallucinations in Multimodal understandingLLM reasoning → Hallucination detection and mitigation +1
Object-centric Denoising Diffusion Models for Physical ReasoningDiffusion models → Discrete and text diffusion
HumanVideo-MME: Benchmarking MLLMs for Human-Centric Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
When do World Models Successfully Learn Dynamical Systems?World models → Latent predictive world models
MARBLE: A Multi-Agent Rule-Based LLM Reasoning Engine for Accident Severity PredictionLLM reasoning → Reasoning for agents, planning and code +1
Fine-tuning on simulated data outperforms prompting for agent tone of voiceLLM agents and tool use → General LLM agent frameworks and benchmarks
Building Open-Retrieval Conversational Question Answering Systems by Generating Synthetic Data and Decontextualizing User QuestionsSynthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping
Beyond Training-time Poisoning: Component-level and Post-training Backdoors in Deep Reinforcement LearningDeep reinforcement learning methods → Adversarial robustness and RL in finance
SHARP: Shared State Reduction for Efficient Matching of Sequential PatternsRetrieval-augmented generation → Generation and context integration
NTSFormer: A Self-Teaching Graph Transformer for Multimodal Isolated Cold-Start Node ClassificationMixture-of-experts → MoE for time series and graphs
Music Boomerang: Reusing Diffusion Models for Data Augmentation and Audio ManipulationDiffusion models → Audio, speech and music
Semantically Consistent Discrete Diffusion for 3D Biological Graph ModelingDiffusion models → Medical imaging and segmentation
RIPE: Reinforcement Learning on Unlabeled Image Pairs for Robust Keypoint ExtractionDeep reinforcement learning methods → RL for robot locomotion and manipulation
Discrete Diffusion Trajectory Alignment via Stepwise DecompositionDiffusion models → Flow matching, time series and science +1
Training-free Generation of Temporally Consistent Rewards from VLMsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Reason to Rote: Rethinking Memorization in ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Efficient Unlearning with Privacy GuaranteesLLM safety and security → Machine unlearning
FurniMAS: Language-Guided Furniture Decoration using Multi-Agent SystemLLM agents and tool use → Multimodal and embodied agents
ABench-Physics: Benchmarking Physical Reasoning in LLMs via High-Difficulty and Dynamic Physics ProblemsLLM reasoning → Spatial and embodied reasoning
Activation Steering for Chain-of-Thought CompressionLLM reasoning → Test-time compute and efficient thinking +1
Vision-Language Models Can't See the ObviousMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
ChipSeek: Optimizing Verilog Generation via EDA-Integrated Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
An analysis of vision-language models for fabric retrievalMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
CueLearner: Bootstrapping and local policy adaptation from relative feedbackDeep reinforcement learning methods → RL training of language and agent models
Backdoors in Conditional Diffusion: Threats to Responsible Synthetic Data PipelinesDiffusion models → Robustness, privacy and adversarial aspects +1
Who's the Mole? Modeling and Detecting Intention-Hiding Malicious Agents in LLM-Based Multi-Agent SystemsLLM agents and tool use → Agent threats, guardrails and governance
UrbanMind: Towards Urban General Intelligence via Tool-Enhanced Retrieval-Augmented Generation and Multilevel OptimizationRetrieval-augmented generation → Agentic RAG, agents and memory
A Visual Leap in CLIP Compositionality Reasoning through Generation of Counterfactual SetsLLM reasoning → Multimodal reasoning +1
Interpretable Reward Modeling with Active Concept BottlenecksLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Structure-Guided Diffusion Models for High-Fidelity Portrait Shadow RemovalDiffusion models → 3D, point cloud and scene generation
Identify, Isolate, and Purge: Mitigating Hallucinations in LVLMs via Self-Evolving DistillationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
Trojan Horse Prompting: Jailbreaking Conversational Multimodal Models by Forging Assistant MessageLLM safety and security → Jailbreaking
VectorLLM: Human-like Extraction of Structured Building Contours vis Multimodal LLMsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
DRAE: Dynamic Retrieval-Augmented Expert Networks for Lifelong Learning and Task Adaptation in RoboticsRetrieval-augmented generation → Agentic RAG, agents and memory +1
SOSAE: Self-Organizing Sparse AutoEncoderMechanistic interpretability → Sparse autoencoder training and evaluation
MODA: MOdular Duplex Attention for Multimodal Perception, Cognition, and Emotion UnderstandingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
PRISM: Pointcloud Reintegrated Inference via Segmentation and Cross-attention for ManipulationOffline RL and imitation learning from data → Robot manipulation imitation learning
Can Prompt Difficulty be Online Predicted for Accelerating RL Finetuning of Reasoning Models?LLM reasoning → Reinforcement learning for reasoning
Learning Robust Stereo Matching in the Wild with Selective Mixture-of-ExpertsMixture-of-experts → Vision mixture-of-experts
Learn 3D VQA Better with Active Selection and ReannotationLLM reasoning → Spatial and embodied reasoning
Information-Guided Diffusion Sampling for Dataset DistillationDiffusion models → Diffusion foundations and sampling
PRIME: Large Language Model Personalization with Cognitive Dual-Memory and Personalized Thought ProcessLLM reasoning → Reinforcement learning for reasoning
Accelerated Online Reinforcement Learning using Auxiliary Start State DistributionsDeep reinforcement learning methods → Deep RL algorithms and representations
S2Edit: Text-Guided Image Editing with Precise Semantic and Spatial ControlDiffusion models → Image generation, text-to-image and editing
Accelerating GenAI Workloads by Enabling RISC-V Microkernel Support in IREE

All 371