Papers

New on arXiv

Thu 22 Jan 2026

268 papers

KD4MT: A Survey of Knowledge Distillation for Machine TranslationKnowledge distillation → Knowledge distillation for LLMs
CodeGuard: Improving LLM Guardrails in CS EducationLLM safety and security → Safety alignment, refusal and guardrails
CARE-RFT: Confidence-Anchored Reinforcement Finetuning for Reliable Reasoning in Large Language ModelsLLM reasoning → Reinforcement learning for reasoning
SPARC-RAG: Adaptive Sequential-Parallel Scaling with Context Management for Retrieval-Augmented GenerationRetrieval-augmented generation → Agentic RAG, agents and memory
A Mechanistic View on Video Generation as World Models: State and DynamicsWorld models → Video-generation world models and simulators
ThinkTank-ME: A Multi-Expert Framework for Middle East Event ForecastingLLM reasoning → Time series reasoning
FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge DevicesMixture-of-experts → Efficient MoE training and inference systems
The Behavioral Fabric of LLM-Powered GUI Agents: Human Values and Interaction OutcomesLLM agents and tool use → Social simulation and multi-agent debate
DSGym: A Holistic Framework for Evaluating and Training Data Science AgentsLLM agents and tool use → Web, search, and research agents
DMAVA: Distributed Multi-Autonomous Vehicle Architecture Using AutowareEnd-to-end autonomous driving → Cooperative and V2X driving
Teaching and Evaluating LLMs to Reason About Polymer Design Related TasksLLM reasoning → Clinical and biomedical applications
Memory-V2V: Memory-Augmented Video-to-Video Diffusion for Consistent Multi-Turn EditingDiffusion models → Video generation
SemanticALLI: Caching Reasoning, Not Just Responses, in Agentic SystemsLLM reasoning → Reasoning for agents, planning and code +1
When Agents Fail to Act: A Diagnostic Framework for Tool Invocation Reliability in Multi-Agent LLM SystemsLLM agents and tool use → Software engineering coding agents
Better as Generators Than Classifiers: Leveraging LLMs and Synthetic Data for Low-Resource Multilingual ClassificationSynthetic data and self-improvement for LLMs → LLM synthetic data generation
GR3EN: Generative Relighting for 3D EnvironmentsDiffusion models → 3D, point cloud and scene generation
CamPilot: Improving Camera Control in Video Diffusion Model with Efficient Camera Reward FeedbackDiffusion models → Video generation +1
Scaling Text-to-Image Diffusion Transformers with Representation AutoencodersDiffusion models → Diffusion foundations and sampling
IVRA: Improving Visual-Token Relations for Robot Action Policy with Training-Free Hint-Based GuidanceVision-language-action models → 3D and spatial-aware VLA
Computer Environments Elicit General Agentic Intelligence in LLMsLLM agents and tool use → General LLM agent frameworks and benchmarks
Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and PlanningDiffusion models → Diffusion and flow policies for robotics
Learning to Watermark in the Latent Space of Generative ModelsLLM safety and security → Watermarking generated content and models
Replicating Human Motivated Reasoning Studies with LLMsLLM reasoning → Reasoning benchmarks and evaluation
Efficiently Learning Robust Torque-based Locomotion Through Reinforcement with Model-Based SupervisionDeep reinforcement learning methods → RL for robot locomotion and manipulation
Multimodal Climate Disinformation Detection: Integrating Vision-Language Models with External Knowledge SourcesMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Clustering-Guided Spatial-Spectral Mamba for Hyperspectral Image ClassificationState-space sequence models (S4, Mamba) → Hyperspectral image analysis with Mamba
Controlling Long-Horizon Behavior in Language Model Agents with Explicit State DynamicsLLM agents and tool use → Social simulation and multi-agent debate
DSFedMed: Dual-Scale Federated Medical Image Segmentation via Mutual Distillation Between Foundation and Lightweight ModelsKnowledge distillation → Federated, robust and private distillation
DTP: A Simple yet Effective Distracting Token Pruning Framework for Vision-Language Action ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +2
ProGiDiff: Prompt-Guided Diffusion-Based Medical Image SegmentationDiffusion models → Medical imaging and segmentation
DextER: Language-driven Dexterous Grasp Generation with Embodied ReasoningLLM reasoning → Spatial and embodied reasoning
Universal Refusal Circuits Across LLMs: Cross-Model Transfer via Trajectory Replay and Concept-Basis ReconstructionMechanistic interpretability → Steering and probing for LLM safety
Deja Vu in Plots: Leveraging Cross-Session Evidence with Retrieval-Augmented LLMs for Live Streaming Risk AssessmentRetrieval-augmented generation → Privacy and security of RAG
PhysicsMind: Sim and Real Mechanics Benchmarking for Physical Reasoning and Prediction in Foundational VLMs and World ModelsWorld models → Latent predictive world models +2
HyperAlign: Hypernetwork for Efficient Test-Time Alignment of Diffusion ModelsDiffusion models → Image generation, text-to-image and editing +1
Decoupling Return-to-Go for Efficient Decision TransformerOffline RL and imitation learning from data → Decision transformers and sequence modeling +1
NeuroMamba: Multi-Perspective Feature Interaction with Visual Mamba for Neuron SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
RadJEPA: Radiology Encoder for Chest X-Rays via Joint Embedding Predictive ArchitectureMultimodal LLMs / vision-language models → Medical vision-language models
Contrast-X: A Multi-Modal Contrast Image Synthesis Benchmark and Universal Modality Flow MatchingDiffusion models → Medical imaging and segmentation
EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic ExperienceLLM agents and tool use → General LLM agent frameworks and benchmarks
SoK: Challenges in Tabular Membership Inference AttacksLLM safety and security → Membership inference attacks
PF-D2M: A Pose-free Diffusion Model for Universal Dance-to-Music GenerationDiffusion models → Audio, speech and music
Towards Realistic Remote Sensing Dataset Distillation with Discriminative Prototype-guided DiffusionDiffusion models → Inverse problems and image restoration
ExDR: Explanation-driven Dynamic Retrieval Enhancement for Multimodal Fake News DetectionRetrieval-augmented generation → Multimodal RAG
Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided VerificationLLM reasoning → Knowledge- and tool-augmented reasoning +2
Attributing and Exploiting Safety Vectors through Global Optimization in Large Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
Assessing Situational and Spatial Awareness of VLMs with Synthetically Generated VideoLLM reasoning → Spatial and embodied reasoning +1
Diffusion Model-Based Data Augmentation for Enhanced Neuron SegmentationDiffusion models → Medical imaging and segmentation
Agentic Confidence CalibrationLLM agents and tool use → General LLM agent frameworks and benchmarks
Off-Policy Actor-Critic with Sigmoid-Bounded Entropy for Real-World Robot LearningOffline RL and imitation learning from data → Offline RL algorithms +1
DualShield: Safe Model Predictive Diffusion via Reachability Analysis for Interactive Autonomous DrivingDiffusion models → Diffusion and flow policies for robotics
Towards Automated Kernel Generation in the Era of LLMsLLM agents and tool use → Software engineering coding agents
VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool ReasoningLLM reasoning → Multimodal reasoning +1
Communication-efficient Federated Graph Classification via Generative Diffusion ModelingDiffusion models → Robustness, privacy and adversarial aspects
Zero-Shot Product Attribute Labeling with Vision-Language Models: A Three-Tier Evaluation FrameworkMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
AgentSM: Semantic Memory for Agentic Text-to-SQLLLM agents and tool use → Web, search, and research agents
Persona Switch: Mixing Distinct Perspectives in Decoding TimeLLM reasoning → Reasoning benchmarks and evaluation
D-Optimality-Guided Reinforcement Learning for Efficient Open-Loop Calibration of a 3-DOF Ankle Rehabilitation RobotDeep reinforcement learning methods → RL for robot locomotion and manipulation
Agentic Uncertainty QuantificationLLM agents and tool use → General LLM agent frameworks and benchmarks
Beyond Visual Safety: Jailbreaking Multimodal Large Language Models for Harmful Image Generation via Semantic-Agnostic InputsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
Performance-guided Reinforced Active Learning for Object DetectionDeep reinforcement learning methods → RL theory and sample complexity
Connect the Dots: Knowledge Graph-Guided Crawler Attack on Retrieval-Augmented Generation SystemsRetrieval-augmented generation → Privacy and security of RAG
Enhancing guidance for missing data in diffusion-based sequential recommendationDiffusion models → Diffusion recommenders
Integrating Knowledge Distillation Methods: A Sequential Multi-Stage FrameworkKnowledge distillation → General knowledge distillation methods
Event-VStream: Event-Driven Real-Time Understanding for Long Video StreamsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Agentic AI Governance and Lifecycle Management in HealthcareLLM agents and tool use → Medical and scientific domain agents
Robust Tool Use via Fission-GRPO: Learning to Recover from Execution ErrorsLLM reasoning → Reinforcement learning for reasoning +1
AION: Aerial Indoor Object-Goal Navigation Using Dual-Policy Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Data-Free Privacy-Preserving for LLMs via Model Inversion and Selective UnlearningLLM safety and security → Machine unlearning
Parallelism and Generation Order in Masked Diffusion Language Models: Limits Today, Potential TomorrowDiffusion models → Discrete and text diffusion
YuFeng-XGuard: A Reasoning-Centric, Interpretable, and Flexible Guardrail Model for Large Language ModelsLLM reasoning → Reasoning for agents, planning and code +1
Relative Classification Accuracy: A Calibrated Metric for Identity Consistency in Fine-Grained K-pop Face GenerationDiffusion models → Image generation, text-to-image and editing
VIOLA: Towards Video In-Context Learning with Minimal AnnotationsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Deep Sketch-Based 3D Modeling: A Survey
GLEAN: Grounded Lightweight Evaluation Anchors for Contamination-Aware Tabular Reasoning
EduResearchBench: A Hierarchical Atomic Task Decomposition Benchmark for Full-Lifecycle Educational Research
High Convergence Rates of CMOS Invertible Logic Circuits Based on Many-Body Hamiltonians
When to Think Fast and Slow? AMOR: Adaptive Entropy Gate for Hybrid Models
BotzoneBench: Scalable LLM Evaluation via Graded AI Anchors
Precoding-Oriented CSI Feedback Design with Mutual Information Regularized VQ-VAE
Learning-augmented smooth integer programs with PAC-learnable oracles
Impact of LLMs news Sentiment Analysis on Stock Price Movement Prediction
Why LoRA Resists Label Noise: A Theoretical Framework for Noise-Robust Parameter-Efficient Fine-Tuning
Design and Empirical Study of a Large Language Model-Based Multi-Agent Investment System for Chinese Public REITs
The GT-Score: A Robust Objective Function for Reducing Overfitting in Data-Driven Trading Strategies
Embedding Compression via Spherical Coordinates
Prediction Markets as Bayesian Inverse Problems: Uncertainty Quantification, Identifiability, and Information Gain from Price-Volume Histories under Latent Types
Lightweight Quantum-Enhanced ResNet for Coronary Angiography Classification: A Hybrid Quantum-Classical Feature Enhancement Framework
Trademark Search, Artificial Intelligence and the Role of the Private Sector
Superpixel-Based Image Segmentation Using Squared 2-Wasserstein Distances
Multi-Agent Deep Reinforcement Learning Under Constrained Communications
Between Search and Platform: ChatGPT Under the DSA
A Computer Vision Pipeline for Iterative Bullet Hole Tracking in Rifle Zeroing
Initial results of the Digital Consciousness Model
Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs
Post-Training Neural Network Pruning using Graph Curvature
Experience with Single Domain Generalization in Real World Medical Imaging Deployments
Identity, Cooperation and Framing Effects within Groups of Real and Simulated Humans
NOIR: Privacy-Preserving Generation of Code with Open-Source LLMs
Regional Bias in Large Language Models

All 268