Papers

New on arXiv

Wed 22 Jul 2026

380 papers

Neo-Classic: A Benchmark for Evaluating Linguistic-Aesthetic Reasoning in Classical Chinese PoetryLLM reasoning → Reasoning benchmarks and evaluation
FedGuide: Diffusion Prior Alignment and Value Baseline Guidance for Heterogeneous Federated Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
Who Are We Recommending To? Recommender Systems in the Agentic WebLLM agents and tool use → Agentic recommender systems
From Plots to Words: Model-Aware Multimodal Explanations as a Foundation for Accessible, Non-Visual InteractionMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Are you Talking Logic to Me? Assessing Language Models Syllogistic Reasoning CapabilitiesLLM reasoning → Reasoning benchmarks and evaluation
Are the Financial Reasoning from LLMs Credible? A Real World Test over Long-Horizon StatementsLLM reasoning → Reasoning for agents, planning and code
KernelGenBench: Can LLMs and Agents Write Efficient Kernels Across Operator Sources and Hardware Platforms?LLM agents and tool use → Software engineering coding agents
Cortex: Compact Behavior Cloning for Quake with Frozen Visual FeaturesDeep reinforcement learning methods → RL for robot locomotion and manipulation
Spatial Reasoning in LLM Game Agents: Impact of Causal Context and Multi-Step PlanningLLM reasoning → Spatial and embodied reasoning +1
Progress-conditioned Group Policy Optimization for Long-Horizon Agentic TasksLLM agents and tool use → Reinforcement learning for LLM agents
Visual Information Extraction from Documents via Classification-Guided Large Vision-Language ModelsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
GRACE: Gradient-Free Robot Action Generation via Combined Diffusion-MPPI Posterior Mean EstimationDiffusion models → Diffusion and flow policies for robotics
Progress Reward Modeling for Robotic Learning: A Comprehensive SurveyLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Molt: A Scalable PyTorch-Native Training Framework for Agentic Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents
3D-GIMP: When 3D Gaussian Inpainting Meets PatchMatchDiffusion models → 3D, point cloud and scene generation
Emergent Compositional Skills in Mixture-of-Experts VLAsMixture-of-experts → MoE in reinforcement learning and robotics
Rushes: A Human Preference Dataset for Pluralistic AlignmentLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
IssueTrojanBench: Benchmarking AI Coding Agents Against Malicious Issue RequestsLLM agents and tool use → Agent threats, guardrails and governance +1
Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token TimingEfficient LLM inference and serving → Speculative decoding
NVIDIA-labs OO Agents: Native Python Object-Oriented AgentsLLM agents and tool use → Software engineering coding agents
Perspective Latents as an Architectural Condition for Causal Emergence in Active Inference AgentsDeep reinforcement learning methods → Deep RL algorithms and representations
Learning to Detect UI Principle Violations via Reinforcement LearningMultimodal LLMs / vision-language models → Chart understanding and anomaly detection +2
Safe and Scalable Multi-Drone Payload Transport via CBF-based Reinforcement Learning with Zero-Shot Sim-to-Real TransferDeep reinforcement learning methods → RL for robot navigation and UAV control
Adaptive Multi-Horizon Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
SalesLoop: Reinforcement Learning from Performance Feedback for Sales Lead RankingDeep reinforcement learning methods → RL training of language and agent models
PhysCoRe: Physics-Corrected Residual World Models for Material-Aware Deformable DynamicsWorld models → Latent predictive world models
Scaling Interpretable Transformers with Parity Bottleneck LayersMechanistic interpretability → Sparse autoencoder training and evaluation
Frontier Financial Judgement: Can agents tell what might move a stock?LLM agents and tool use → Financial trading and forecasting agents
WaveformQA: Benchmarking LLM Temporal Reasoning on Digital WaveformsLLM reasoning → Reasoning for agents, planning and code
Demonstrating GenDB: Instance-Optimized and Customized Query Processing Code Generation via LLM AgentsLLM agents and tool use → Web, search, and research agents
RealVDeblur: One-Step Diffusion for Generalizable Real-World Video DeblurringDiffusion models → Inverse problems and image restoration
MedDiT4SR: Tri-Stream Joint Adaptation of Pre-Trained Diffusion Transformers for Medical Image Super-ResolutionDiffusion models → Inverse problems and image restoration
Are Single-Token Sparse Autoencoder Features Causally Necessary? Layer-Depth and SAE-Family EffectsMechanistic interpretability → SAE features and their applications
Geometric Configurations of Perturbed Jailbreak PromptsLLM safety and security → Jailbreaking
LKValues: Aligning Large Language Models with Sri Lankan Societal ValuesLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
SoftReason: A Fully Differentiable Neuro-Soft-Symbolic Deductive Reasoning Architecture over High-Dimensional Perceptual DataLLM reasoning → Reasoning benchmarks and evaluation
Self Gradient Forcing: Native Long Video ExtrapolationDiffusion models → Video generation
Look Less, Think Faster: Joint Token-Compute Adaptation for Multimodal LLMsLLM reasoning → Multimodal reasoning +1
Test-Time Training for Modality Order Consistency in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Closing the Lab-to-Store Gap: A Data-Efficient Post-Training and Experience-Driven Learning VLA Framework for Retail HumanoidsVision-language-action models → Generalist VLA models and benchmarks
PyroDash: Cost-Efficient Token-Level Small-Large Language Model Collaborative InferenceLLM reasoning → Test-time compute and efficient thinking
Evolving Cache Schedules for Fast Diffusion Policy InferenceDiffusion models → Diffusion foundations and sampling
Sound Probabilistic Safety Bounds for Large Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
Multimodal Large Language Models for Remote Sensing Image Understanding: Domain-Specific or General-Purpose?Multimodal LLMs / vision-language models → Embodied and 3D vision-language models
PoTRE: Test-Time Reasoning inspired by Cognitive HeterogeneityLLM reasoning → Test-time compute and efficient thinking
The Ethics of Autonomous AI Agents for Offensive SecurityLLM agents and tool use → Agent threats, guardrails and governance
User-Centric Modeling of Transactional Sequences with Explainable State Space ModelsState-space sequence models (S4, Mamba) → Core SSM architectures for language
SeededGrasp: Language-Guided Grasping in Complex Scenes with Multiple EmbodimentsLLM reasoning → Spatial and embodied reasoning
RS-RIE-Bench: Benchmarking Reasoning-Guided Remote Sensing Image EditingLLM reasoning → Spatial and embodied reasoning
On Optimization Complexity of Second-Order Certified UnlearningLLM safety and security → Machine unlearning
PerceptDrive: Perception Prior World-Action Modeling with Adaptive Expert Routing for End-to-End Autonomous DrivingWorld models → World models for autonomous driving +2
Audio-Zero: Label-Free Self-Evolution for Fine-Grained Audio ReasoningLLM reasoning → Audio-language model reasoning
Active Inference as a Convex Markov Decision ProcessDeep reinforcement learning methods → RL theory and sample complexity
SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPODMixture-of-experts → Efficient MoE training and inference systems
CUSUM-Shaped Inference-Time Monitoring and Targeted Re-Decoding for Quantized Small Language Model ReasoningLLM reasoning → Test-time compute and efficient thinking
HeadCast: Casting Attention Heads for Efficient Autoregressive Video GenerationDiffusion models → Video generation
OpenSkillRisk: Benchmarking Agent Safety When Using Real-World Risky Third-Party SkillsLLM agents and tool use → Agent threats, guardrails and governance
ENTRAP-VL: A Taxonomic Probe for Dual Contextual Entrainment in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Reinforcement Learning for Large Language Model Selective Evidence Adoption from Contaminated Retrieval ResultsRetrieval-augmented generation → Agentic RAG, agents and memory
PRO-LONG: Programmatic Memory Enables Long-Horizon ReasoningLLM reasoning → Reasoning for agents, planning and code +1
Solar Open 2 Technical ReportLLM agents and tool use → General LLM agent frameworks and benchmarks +1
Language-Specific versus Cross-Lingual Knowledge Graphs for Implicit Aspect Identification in Arabic: A Comparative Study of Reasoning and Adaptation StrategiesLLM reasoning → Audio-language model reasoning
Importance-Aware OBS Pruning for Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Generalized Kalman filter based temporal difference reinforcement learningDeep reinforcement learning methods → RL theory and sample complexity
STEREOFLOW: Progressive Stereo Matching with StereoDiT and Transition Flow MatchingDiffusion models → 3D, point cloud and scene generation
Towards Reliable C-to-Rust Translation with Rule-Guided Reasoning and Reinforcement LearningLLM reasoning → Reasoning for agents, planning and code
EvoThink: Evolving Thinking in Large Reasoning Models via Self-Pruning and Aha-Moment Preference OptimizationLLM reasoning → Test-time compute and efficient thinking +1
HijackKV: New Threat in Position-Independent KV Cache ReuseEfficient LLM inference and serving → KV cache optimization
When Does Knowledge Distillation Hurt? Reliability-Aware Distillation for Low-Resource Language SummarizationKnowledge distillation → Knowledge distillation for LLMs
ETPDesigner: Multi-Agent Orchestration for Interactive Multimodal Electronic Theater ProgramLLM agents and tool use → Multimodal and embodied agents
Efficient Chain-of-Modality Reasoning via Progressive Compression for Spoken Language ModelsLLM reasoning → Test-time compute and efficient thinking
Continuous-Time Reinforcement Learning for N-Player Stochastic Differential Games with Exploratory PoliciesDeep reinforcement learning methods → RL theory and sample complexity
WearWow: Native 2K Multi-Garment Virtual Try-On via Adaptive Token Packing and Preference AlignmentDiffusion models → 3D, point cloud and scene generation
MV-Bench: Benchmarking Multimodal Large Language Models for Coordinated Multi-View Interface ConstructionMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
OSVE: One Step Video Editing with One Step Diffusion ModelsDiffusion models → Video generation
Defense Against LLM Backdoors using Critical Neuron Isolation PruningLLM safety and security → Backdoor and data poisoning attacks
MTVDiff: Multimodal Conditional Latent Diffusion for Enhanced Thermal-to-Visible Face TranslationDiffusion models → Image generation, text-to-image and editing
KineBench: Benchmarking Embodied World Models via IDM-Free Kinematic GroundingWorld models → Embodied and robot world models
Memory-Augmented Multimodal Large Language Models for Small Object Understanding in Streaming Aerial VideosMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Asymptotically Optimal Regret for Reinforcement Learning without Horizon DependenceDeep reinforcement learning methods → RL theory and sample complexity
SOPD-SocialNav: Selective On-Policy Distillation for Vision-Language Social NavigationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Optimal Sensor Placement via Graph-constrained Flow MatchingDiffusion models → Flow matching, time series and science
VizRAG: Enhancing Retrieval-Augmented Generation with Hypergraph VisualizationRetrieval-augmented generation → Multimodal RAG
DARWIN: Evolving Jailbreak Adversary and Guardrail for LLM Safety Evaluation and ProtectionLLM safety and security → Jailbreaking
Rewarding Better Thinking for LLM Preference AlignmentLLM reasoning → Reinforcement learning for reasoning +1
Dreamer-CPC: Message Learning with World Models for Decentralized Multi-agent Reinforcement LearningWorld models → World models for language agents
OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent OptimizationLLM safety and security → Safety alignment, refusal and guardrails +1
CIR at iKAT SCAI 2026: Exploring Clarification Need Prediction in Agentic Conversational SearchLLM agents and tool use → Web, search, and research agents
TriAgent: Divergence-Aware Multi-Agent Committees for Cost-Efficient Financial Sentiment AnalysisLLM agents and tool use → Financial trading and forecasting agents
Silent Failures in Multimodal Agentic Search:A Diagnostic Taxonomy and Cross-Judge EvaluationLLM agents and tool use → Multimodal and embodied agents
Trace: A Taxonomy-Guided Environment for Multidomain Visual ReasoningLLM reasoning → Multimodal reasoning +1
Not Birds of a Feather: Personality-Based Partner Selection in LLM AgentsLLM agents and tool use → Social simulation and multi-agent debate
Defer to Plan: Adaptive Multi-Agent Fusion for End-to-End V2X DrivingEnd-to-end autonomous driving → Cooperative and V2X driving
The World Model Remembers, the Actor Forgets: Dream Rehearsal for Continual Model-Based RLWorld models → Latent dynamics models and model-based RL
Personalized Recommendation Tool Learning via Autonomous Language AgentsLLM reasoning → LLM reasoning for recommendation +1
Analytic Distribution of Classifier-Free Guidance for Schedule DesignDiffusion models → Diffusion foundations and sampling
Koopman Dreamer: Spectrally Constrained Latent Dynamics for Stable World-Model ImaginationWorld models → Latent predictive world models
SafeGen: Goal-Conditioned Video Diffusion of Safety-Critical Scenarios for VLM-Based Autonomous DrivingDiffusion models → Diffusion and flow policies for robotics +1
SLPO: Scaling Latent Reasoning via a Surrogate PolicyLLM reasoning → Test-time compute and efficient thinking
Multi-Mask Diffusion Language Models for Few-Step GenerationDiffusion models → Discrete and text diffusion

All 380