Papers

New on arXiv

Fri 19 Dec 2025

335 papers

Keyword search is all you need: Achieving RAG-Level Performance without vector databases using agentic tool useRetrieval-augmented generation → Agentic RAG, agents and memory +1
WebAccessVL: Violation-Aware VLM for Web AccessibilityMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Consistency-Aware Editing for Entity-level Unlearning in Language ModelsLLM safety and security → Machine unlearning
A Multi-Stage Workflow for the Review of Marketing Content with Reasoning Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Reinforcement Learning for Chain of Thought Compression with One-Domain-to-All GeneralizationLLM reasoning → Test-time compute and efficient thinking
WRAVAL -- WRiting Assist eVALuationLLM reasoning → Reasoning benchmarks and evaluation
q3-MuPa: Quick, Quiet, Quantitative Multi-Parametric MRI using Physics-Informed Diffusion ModelsDiffusion models → Medical imaging and segmentation
Unbiased Visual Reasoning with Controlled Visual InputsLLM reasoning → Multimodal reasoning +1
Dominant vs. Dominated: Concept-Level Generative Collapse in Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Eidoku: A Neuro-Symbolic Verification Gate for LLM Reasoning via Structural Constraint SatisfactionLLM reasoning → Reasoning benchmarks and evaluation
Unifying Causal Reinforcement Learning: Survey, Taxonomy, Algorithms and ApplicationsDeep reinforcement learning methods → Deep RL algorithms and representations
SAM Audio: Segment Anything in AudioDiffusion models → Audio, speech and music
Rethinking Multi-Agent Intelligence Through the Lens of Small-World NetworksLLM reasoning → Reasoning for agents, planning and code +1
Faithful and Stable Neuron Explanations for Trustworthy Mechanistic InterpretabilityMechanistic interpretability → Circuits and mechanistic analysis
From Prompt to Product: A Human-Centered Benchmark of Agentic App Generation SystemsLLM agents and tool use → Software engineering coding agents
FPBench: A Comprehensive Benchmark of Multimodal Large Language Models for Fingerprint AnalysisMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
SurgiPose: Estimating Surgical Tool Kinematics from Monocular Video for Surgical Robot LearningOffline RL and imitation learning from data → Safe offline RL
Graph-based Nearest Neighbors with Dynamic Updates via Random WalksRetrieval-augmented generation → Vector search and ANN indexes
Securing Agentic AI Systems -- A Multilayer Security FrameworkLLM agents and tool use → Agent threats, guardrails and governance
Towards Benchmarking Privacy Vulnerabilities in Selective Forgetting with Large Language ModelsLLM safety and security → Machine unlearning
ReGal: A First Look at PPO-based Legal AI for Judgment Prediction and Summarization in IndiaLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment +1
Robotic VLA Benefits from Joint Learning with Motion Image DiffusionVision-language-action models → Efficient VLA inference and adaptation
Unifying Deep Predicate Invention with Pre-trained Foundation ModelsWorld models → Embodied and robot world models
MoE-TransMov: A Transformer-based Model for Next POI Prediction in Familiar & Unfamiliar MovementsMixture-of-experts → MoE for time series and graphs
Both Semantics and Reconstruction Matter: Making Representation Encoders Ready for Text-to-Image Generation and EditingDiffusion models → Diffusion foundations and sampling
Dexterous World ModelsWorld models → Embodied and robot world models +1
When Reasoning Meets Its LawsLLM reasoning → Test-time compute and efficient thinking
Distributionally Robust Imitation Learning: Layered Control Architecture for Certifiable AutonomyOffline RL and imitation learning from data → Imitation learning methods
RadarGen: Automotive Radar Point Cloud Generation from CamerasDiffusion models → 3D, point cloud and scene generation
XAgen: An Explainability Tool for Identifying and Correcting Failures in Multi-Agent WorkflowsLLM agents and tool use → Software engineering coding agents
Weighted Stochastic Differential Equation to Implement Wasserstein-Fisher-Rao Gradient FlowDiffusion models → Diffusion foundations and sampling
Preserving Spectral Structure and Statistics in Diffusion ModelsDiffusion models → Diffusion foundations and sampling
InfSplign: Inference-Time Spatial Alignment of Text-to-Image Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Planning as Descent: Goal-Conditioned Latent Trajectory Synthesis in Learned Energy LandscapesDeep reinforcement learning methods → Deep RL algorithms and representations
NeuRehab: A Reinforcement Learning and Spiking Neural Network-Based Rehab Automation FrameworkDeep reinforcement learning methods → Spiking and neuromorphic RL
ReX-MLE: The Autonomous Agent Benchmark for Medical Imaging ChallengesLLM agents and tool use → Medical and scientific domain agents
UrbanDIFF: A Denoising Diffusion Model for Spatial Gap Filling of Urban Land Surface Temperature Under Dense Cloud CoverDiffusion models → Inverse problems and image restoration
DEER: A Benchmark for Evaluating Deep Research Agents on Expert Report GenerationLLM agents and tool use → Web, search, and research agents
AdaptPrompt: Parameter-Efficient Adaptation of VLMs for Generalizable Deepfake DetectionDiffusion models → Image generation, text-to-image and editing
An Empirical Study of Sampling Hyperparameters in Diffusion-Based Super-ResolutionDiffusion models → Inverse problems and image restoration
Vidarc: Embodied Video Diffusion Model for Closed-loop ControlDiffusion models → Diffusion and flow policies for robotics
Generative Human-Object Interaction Detection via Differentiable Cognitive Steering of Multi-modal LLMsMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
About Time: Model-free Reinforcement Learning with Timed Reward MachinesDeep reinforcement learning methods → Interpretable and logic-guided RL
Trust-Region Adaptive Policy OptimizationLLM reasoning → Reinforcement learning for reasoning
PathFLIP: Fine-grained Language-Image Pretraining for Versatile Computational PathologyMultimodal LLMs / vision-language models → Medical vision-language models
Learning Safe Autonomous Driving Policies Using Predictive Safety RepresentationsEnd-to-end autonomous driving → Reinforcement learning for driving +1
SkinGenBench: Generative Model and Preprocessing Effects for Synthetic Dermoscopic Augmentation in Melanoma DiagnosisDiffusion models → Medical imaging and segmentation
Enabling Disaggregated Multi-Stage MLLM Inference via GPU-Internal Scheduling and Resource SharingMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Kinematics-Aware Diffusion Policy with Consistent 3D Observation and Action Space for Whole-Arm Robotic ManipulationDiffusion models → Diffusion and flow policies for robotics
The HydroGym Reinforcement Learning Platform for Fluid DynamicsDeep reinforcement learning methods → RL with diffusion and flow models
Robust-R1: Degradation-Aware Reasoning for Robust Visual UnderstandingLLM reasoning → Multimodal reasoning +1
InsertAnywhere: Geometrically Grounded and Optics-Aware Video Object InsertionDiffusion models → 3D, point cloud and scene generation
UPMRI: Unsupervised Parallel MRI Reconstruction via Projected Conditional Flow MatchingDiffusion models → Medical imaging and segmentation
Xiaomi MiMo-VL-Miloco Technical ReportLLM reasoning → Spatial and embodied reasoning +1
ImagineNav++: Prompting Vision-Language Models as Embodied Navigator through Scene ImaginationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Are Vision Language Models Cross-Cultural Theory of Mind Reasoners?LLM reasoning → Reasoning benchmarks and evaluation +1
AdvJudge-Zero: Binary Decision Flips in LLM-as-a-Judge via Adversarial Control TokensLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification
TakeAD: Preference-based Post-optimization for End-to-end Autonomous Driving with Expert Takeover DataEnd-to-end autonomous driving → Imitation-learning end-to-end planners +2
Adversarially Robust Detection of Harmful Online Content: A Computational Design Science ApproachLLM safety and security → Safety alignment, refusal and guardrails
Democratising Pathology Co-Pilots: An Open Pipeline and Dataset for Whole-Slide Vision-Language ModellingMultimodal LLMs / vision-language models → Medical vision-language models
DESSERT: Diffusion-based Event-driven Single-frame Synthesis via Residual TrainingDiffusion models → Video generation
A Benchmark for Ultra-High-Resolution Remote Sensing MLLMsLLM reasoning → Spatial and embodied reasoning +1
Auxiliary Descriptive Knowledge for Few-Shot Adaptation of Vision-Language ModelMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
CodeDance: A Dynamic Tool-integrated MLLM for Executable Visual ReasoningLLM reasoning → Multimodal reasoning
Cryptanalysis of LDPC-Based Pseudorandom Error-Correcting CodesLLM safety and security → Watermarking generated content and models
Large Language Models as Pok'emon Battle Agents: Strategic Play and Content GenerationLLM agents and tool use → Reinforcement learning for LLM agents
Deep But Reliable: Advancing Multi-turn Reasoning for Thinking with ImagesLLM reasoning → Multimodal reasoning +1
EMAG: Self-Rectifying Diffusion Sampling with Exponential Moving Average GuidanceDiffusion models → Diffusion foundations and sampling
ProCache: Constraint-Aware Feature Caching with Selective Computation for Diffusion Transformer AccelerationDiffusion models → Diffusion foundations and sampling
Robust TTS Training via Self-Purifying Flow Matching for the WildSpoof 2026 TTS TrackDiffusion models → Audio, speech and music
Vision-Language Model Guided Image RestorationMultimodal LLMs / vision-language models → Image editing and quality assessment
WDFFU-Mamba: A Wavelet-guided Dual-attention Feature Fusion Mamba for Breast Tumor Segmentation in Ultrasound ImagesState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Verifiability-First Agents: Provable Observability and Lightweight Audit Agents for Controlling Autonomous LLM SystemsLLM agents and tool use → Agent threats, guardrails and governance +1
Mitty: Diffusion-based Human-to-Robot Video GenerationDiffusion models → Diffusion and flow policies for robotics
Cooperative Energy Scheduling of Multi-Microgrids Based on Risk-Sensitive Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
Video Detective: Seek Critical Clues Recurrently to Answer Question from Long VideosMultimodal LLMs / vision-language models → Video understanding with MLLMs
Learning When to Look: A Disentangled Curriculum for Strategic Perception in Multimodal ReasoningLLM reasoning → Multimodal reasoning +1
DAVE: A VLM Vision Encoder for Document Understanding and Web AgentsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
Mindscape-Aware Retrieval Augmented Generation for Improved Long Context UnderstandingRetrieval-augmented generation → RAG pipeline design and evaluation
CheXPO-v2: Preference Optimization for Chest X-ray VLMs with Knowledge Graph ConsistencyLLM reasoning → Clinical and biomedical applications +2
Reasoning Palette: Modulating Reasoning via Latent Contextualization for Controllable Exploration for (V)LMsLLM reasoning → Test-time compute and efficient thinking
MMRAG-RFT: Two-stage Reinforcement Fine-tuning for Explainable Multi-modal Retrieval-augmented GenerationRetrieval-augmented generation → Agentic RAG, agents and memory
Anatomical Region-Guided Contrastive Decoding: A Plug-and-Play Strategy for Mitigating Hallucinations in Medical VLMsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Conservative Bias in Multi-Teacher Learning: Why Agents Prefer Low-Reward AdvisorsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
Can Synthetic Images Serve as Effective and Efficient Class Prototypes?Multimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
PhysFire-WM: A Physics-Informed World Model for Emulating Fire Spread DynamicsWorld models → Latent predictive world models
Biosecurity-Aware AI: Agentic Risk Auditing of Soft Prompt Attacks on ESM-Based Variant PredictorsLLM agents and tool use → Agent threats, guardrails and governance
Perceptions of AI-CBT: Trust and Barriers in Chinese Postgrads
Triples and Knowledge-Infused Embeddings for Clustering and Classification of Scientific Documents
Hierarchical Sparse Plus Low Rank Compression of LLM
Sports Business Administration and New Age Technology: Role of AI
Advances and Challenges in Semantic Textual Similarity: A Comprehensive Survey
The Instruction Gap: LLMs get lost in Following Instruction
OpenAI GPT-5 System Card
Spike-Timing-Dependent Plasticity for Bernoulli Message Passing
AI-Enhanced Virtual Biopsies for Brain Tumor Diagnosis in Low Resource Settings
Enhancing Medical Data Analysis through AI-Enhanced Locally Linear Embedding: Applications in Medical Point Location and Imagery
Interpretable Link Prediction in AI-Driven Cancer Research: Uncovering Co-Authorship Patterns
iOS as Acceleration
Latent Sculpting for Zero-Shot Generalization: A Manifold Learning Approach to Out-of-Distribution Anomaly Detection

All 335