Papers

New on arXiv

Tue 18 Nov 2025

391 papers

Orchestrating Heterogeneous Experts: A Scalable MoE Framework with Anisotropy-Preserving FusionMixture-of-experts → MoE large language models
Reinforcement Learning from Implicit Neural Feedback for Human-Aligned Robot ControlLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
Socially aware navigation for mobile robots: a survey on deep reinforcement learning approachesDeep reinforcement learning methods → RL for robot navigation and UAV control
From Competition to Coordination: Market Making as a Scalable Framework for Safe and Aligned Multi-Agent LLM SystemsLLM agents and tool use → Financial trading and forecasting agents
Generalizable and Efficient Automated Scoring with a Knowledge-Distilled Multi-Task Mixture-of-ExpertsMixture-of-experts → MoE large language models
Extending Test-Time Scaling: A 3D Perspective with Context, Batch, and TurnLLM reasoning → Test-time compute and efficient thinking +1
Just Asking Questions: Doing Our Own Research on Conspiratorial Ideation by Generative AI ChatbotsLLM safety and security → Prompt injection and agent security
Logit-Based Losses Limit the Effectiveness of Feature Knowledge DistillationKnowledge distillation → General knowledge distillation methods
SVBRD-LLM: Self-Verifying Behavioral Rule Discovery for Autonomous Vehicle IdentificationEnd-to-end autonomous driving → Large language models for driving
Quality-Controlled Multimodal Emotion Recognition in Conversations with Identity-Based Transfer Learning and MAMBA FusionState-space sequence models (S4, Mamba) → Speech and audio SSMs
How to Train Private Clinical Language Models: A Comparative Study of Privacy-Preserving Pipelines for ICD-9 CodingKnowledge distillation → Federated, robust and private distillation
It's LIT! Reliability-Optimized LLMs with Inspectable ToolsLLM agents and tool use → Tool learning and function calling
Skin-R1: Clinical Knowledge-Guided Dermatological Diagnosis Using Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
InstructMix2Mix: Consistent Sparse-View Editing Through Multi-View Model PersonalizationDiffusion models → 3D, point cloud and scene generation
Transformer-Guided Deep Reinforcement Learning for Optimal Takeoff Trajectory Design of an eVTOL DroneDeep reinforcement learning methods → RL for robot navigation and UAV control
GeoSceneGraph: Geometric Scene Graph Diffusion Model for Text-guided 3D Indoor Scene SynthesisDiffusion models → 3D, point cloud and scene generation
When CNNs Outperform Transformers and Mambas: Revisiting Deep Architectures for Dental Caries SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Empowering Multi-Turn Tool-Integrated Agentic Reasoning with Group Turn Policy OptimizationLLM reasoning → Reinforcement learning for reasoning +1
DEVAL: A Framework for Evaluating and Improving the Derivation Capability of Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
UniGen-1.5: Enhancing Image Generation and Editing through Reward Unification in Reinforcement LearningMultimodal LLMs / vision-language models → Image editing and quality assessment +2
π*0.6: a VLA That Learns From ExperienceVision-language-action models → Reinforcement learning and post-training
Co-Me: Confidence-Guided Token Merging for Visual Geometric TransformersEfficient LLM inference and serving → Token pruning and merging for ViTs
Cloud-Native Vector Search: A Comprehensive Performance AnalysisRetrieval-augmented generation → Vector search and ANN indexes
Diffusion As Self-Distillation: End-to-End Latent Diffusion In One ModelDiffusion models → Diffusion foundations and sampling
SMRC: Aligning Large Language Models with Student Reasoning for Mathematical Error CorrectionLLM reasoning → Mathematical reasoning +1
NERD: Network-Regularized Diffusion Sampling For 3D Computed TomographyDiffusion models → Medical imaging and segmentation
Streamlining Industrial Contract Management with Retrieval-Augmented LLMsRetrieval-augmented generation → Legal and financial RAG
NORA-1.5: A Vision-Language-Action Model Trained using World Model- and Action-based Preference RewardsWorld models → Embodied and robot world models +2
AutoTool: Efficient Tool Selection for Large Language Model AgentsLLM agents and tool use → Tool learning and function calling
A Specialized Large Language Model for Clinical Reasoning and Diagnosis in Rare DiseasesLLM reasoning → Clinical and biomedical applications
Enhancing Agentic Autonomous Scientific Discovery with Vision-Language Model CapabilitiesMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
3D-Guided Scalable Flow Matching for Generating Volumetric Tissue Spatial Transcriptomics from Serial HistologyDiffusion models → Medical imaging and segmentation
CCSD: Cross-Modal Compositional Self-Distillation for Robust Brain Tumor Segmentation with Missing ModalitiesKnowledge distillation → Distillation for medical imaging
DSBench: A Comprehensive Benchmark for Evaluating External and In-Cabin RisksMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
ReflexGrad: Within-Episode Failure Recovery in LLM Agents via Progress-Gated Dual-Process RoutingLLM agents and tool use → Software engineering coding agents
OmniZip: Audio-Guided Dynamic Token Compression for Fast Omnimodal Large Language ModelsEfficient LLM inference and serving → Audio and omni-modal token compression
Masked IRL: LLM-Guided Reward Disambiguation from Demonstrations and LanguageLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
MissHDD: Hybrid Deterministic Diffusion for Hetrogeneous Incomplete Data ImputationDiffusion models → Flow matching, time series and science
A Neuro-Symbolic Framework for Reasoning under Perceptual Uncertainty: Bridging Continuous Perception and Discrete Symbolic PlanningLLM reasoning → Spatial and embodied reasoning
LiveRAG: A diverse Q&A dataset with varying difficulty level for RAG evaluationRetrieval-augmented generation → RAG pipeline design and evaluation
Full-Atom Peptide Design via Riemannian-Euclidean Bayesian Flow NetworksDiffusion models → Flow matching, time series and science
LiteCache: A Query Similarity-Driven, GPU-Centric KVCache Subsystem for Efficient LLM InferenceEfficient LLM inference and serving → KV cache optimization
Parameter Aware Mamba Model for Multi-task Dense PredictionState-space sequence models (S4, Mamba) → Vision Mamba backbones
Enhancing End-to-End Autonomous Driving with Risk Semantic Distillaion from VLMMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
Towards Stable and Structured Time Series Generation with Perturbation-Aware Flow MatchingDiffusion models → Flow matching, time series and science
Segmentation-Aware Latent Diffusion for Satellite Image Super-Resolution: Enabling Smallholder Farm Boundary DelineationDiffusion models → Inverse problems and image restoration
Agentic AI Systems in Electrical Power Systems Engineering: Current State-of-the-Art and ChallengesLLM agents and tool use → Financial trading and forecasting agents
Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model BehaviorLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
nnterp: A Standardized Interface for Mechanistic Interpretability of TransformersMechanistic interpretability → Circuits and mechanistic analysis
Agent-R1: A Unified and Modular Framework for Agentic Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents
Agentic Video Intelligence: A Flexible Framework for Advanced Video Exploration and UnderstandingLLM agents and tool use → Multimodal and embodied agents
Tell Me: An LLM-powered Mental Well-being Assistant with RAG, Synthetic Dialogue Generation, and Agentic PlanningLLM agents and tool use → Medical and scientific domain agents
MedBench v4: A Robust and Scalable Benchmark for Evaluating Chinese Medical Language Models, Multimodal Models, and Intelligent AgentsLLM agents and tool use → Medical and scientific domain agents
Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
MiAD: Mirage Atom Diffusion for De Novo Crystal GenerationDiffusion models → Flow matching, time series and science
Unified Defense for Large Language Models against Jailbreak and Fine-Tuning Attacks in EducationLLM safety and security → Jailbreaking
Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral CloningVision-language-action models → Human-video and latent-action pretraining +1
Enhancing LLM-based Autonomous Driving with Modular Traffic Light and Sign RecognitionLLM reasoning → Chain-of-thought reasoning for driving +1
O3SLM: Open Weight, Open Data, and Open Vocabulary Sketch-Language ModelMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
ATLAS: A High-Difficulty, Multidisciplinary Benchmark for Frontier Scientific ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning
ArchMap: Arch-Flattening and Knowledge-Guided Vision Language Model for Tooth Counting and Structured Dental UnderstandingMultimodal LLMs / vision-language models → Medical vision-language models
MA-SLAM: Active SLAM in Large-Scale Unknown Environment using Map Aware Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
H-LDM: Hierarchical Latent Diffusion Models for Controllable and Interpretable PCG Synthesis from Clinical MetadataDiffusion models → Medical imaging and segmentation
Iterative Diffusion-Refined Neural Attenuation Fields for Multi-Source Stationary CT Reconstruction: NAF Meets Diffusion ModelDiffusion models → Medical imaging and segmentation
Audio Question Answering with GRPO-Based Fine-Tuning and Calibrated Segment-Level PredictionsLLM reasoning → Audio-language model reasoning
DataSage: Multi-agent Collaboration for Insight Discovery with External Knowledge Retrieval, Multi-role Debating, and Multi-path ReasoningLLM reasoning → Reasoning for agents, planning and code +1
GEN3D: Generating Domain-Free 3D Scenes from a Single ImageWorld models → Video-generation world models and simulators
Let Language Constrain Geometry: Vision-Language Models as Semantic and Spatial Critics for 3D GenerationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Object-Centric World Models for Causality-Aware Reinforcement LearningWorld models → Latent predictive world models +1
Entropy-Guided Reasoning CompressionLLM reasoning → Test-time compute and efficient thinking
PathMind: A Retrieve-Prioritize-Reason Framework for Knowledge Graph Reasoning with Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
AfriSpeech-MultiBench: A Verticalized Multidomain Multicountry Benchmark Suite for African Accented English ASRMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
StreamingTalker: Audio-driven 3D Facial Animation with Autoregressive Diffusion ModelDiffusion models → Audio, speech and music
LLM-Aligned Geographic Item Tokenization for Local-Life RecommendationLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
Twice Sequential Monte Carlo for Tree SearchDeep reinforcement learning methods → RL theory and sample complexity
Listen Like a Teacher: Mitigating Whisper Hallucinations using Adaptive Layer Attention and Knowledge DistillationKnowledge distillation → Speech and audio distillation
Orion: A Unified Visual Agent for Multimodal Perception, Advanced Visual Reasoning and ExecutionLLM reasoning → Multimodal reasoning +1
InstantViR: Real-Time Video Inverse Problem Solver with Distilled Diffusion PriorDiffusion models → Inverse problems and image restoration
Online Data Curation for Object Detection via Marginal Contributions to Dataset-level Average PrecisionKnowledge distillation → Detection and dense prediction distillation
N-GLARE: An Non-Generative Latent Representation-Efficient LLM Safety EvaluatorLLM safety and security → Safety alignment, refusal and guardrails
PAVE: An End-to-End Dataset for Production Autonomous Vehicle EvaluationEnd-to-end autonomous driving → End-to-end system architecture
WebRec: Enhancing LLM-based Recommendations with Attention-guided RAG from WebRetrieval-augmented generation → RAG for recommendation
Towards Deploying VLA without Fine-Tuning: Plug-and-Play Inference-Time VLA Policy Steering via Embodied Evolutionary DiffusionVision-language-action models → Reinforcement learning and post-training
AdaTok: Adaptive Token Compression with Object-Aware Representations for Efficient Multimodal LLMsMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
Certified Signed Graph UnlearningLLM safety and security → Graph unlearning
A Receding Horizon Reinforcement Learning Framework for Campus Chiller Energy Management - A case study from an Australian UniversityDeep reinforcement learning methods → RL for energy, networks and operations
MVI-Bench: A Comprehensive Benchmark for Evaluating Robustness to Misleading Visual Inputs in LVLMsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action ModelsVision-language-action models → Efficient VLA inference and adaptation +1
Beyond Fixed and Dynamic Prompts: Embedded Jailbreak Templates for Advancing LLM SecurityLLM safety and security → Jailbreaking
Beyond Accuracy: A Multi-Dimensional Framework for Evaluating Enterprise Agentic AI SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
Run, Ruminate, and Regulate: A Dual-process Thinking System for Vision-and-Language NavigationLLM reasoning → Spatial and embodied reasoning
MalRAG: A Retrieval-Augmented LLM Framework for Open-set Malicious Traffic IdentificationRetrieval-augmented generation → Privacy and security of RAG
Multi-view Phase-aware Pedestrian-Vehicle Incident Reasoning Framework with Vision-Language ModelsLLM reasoning → Chain-of-thought reasoning for driving +1
FailSafe: High-performance Resilient ServingEfficient LLM inference and serving → Serving systems and request scheduling
Coffee: Controllable Diffusion Fine-tuningDiffusion models → Robustness, privacy and adversarial aspects
Stealth Fine-Tuning: Efficiently Breaking Alignment in RVLMs Using Self-Generated CoTLLM reasoning → Reasoning for agents, planning and code +4
MoE-SpeQ: Speculative Quantized Decoding with Proactive Expert Prefetching and Offloading for Mixture-of-ExpertsMixture-of-experts → Efficient MoE training and inference systems
APD-Agents: A Large Language Model-Driven Multi-Agents Collaborative Framework for Automated Page DesignLLM agents and tool use → Multimodal and embodied agents
Text-Driven Reasoning Video Editing via Reinforcement Learning on Digital Twin RepresentationsLLM reasoning → Multimodal reasoning
FAPE-IR: Frequency-Aware Planning and Execution Framework for All-in-One Image RestorationMixture-of-experts → Vision mixture-of-experts

All 391