Papers

New on arXiv

Sun 9 Mar 2025

226 papers

Small Vision-Language Models: A Survey on Compact Architectures and TechniquesMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
Dynamics-Invariant Quadrotor Control using Scale-Aware Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Life-Cycle Routing Vulnerabilities of LLM RouterLLM safety and security → Mixture-of-experts routing safety
Probabilistic Shielding for Safe Reinforcement LearningDeep reinforcement learning methods → Safe and constrained RL
Retrieval Augmented Generation with Multi-Modal LLM Framework for Wireless EnvironmentsRetrieval-augmented generation → RAG pipeline design and evaluation
Privacy Auditing of Large Language ModelsLLM safety and security → Membership inference attacks
Does Acceleration Cause Hidden Instability in Vision Language Models? Uncovering Instance-Level Divergence Through a Large-Scale Empirical StudyMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
AutoMisty: A Multi-Agent LLM Framework for Automated Code Generation in the Misty Social RobotLLM agents and tool use → General LLM agent frameworks and benchmarks
GenDR: Lighten Generative Detail RestorationDiffusion models → Inverse problems and image restoration
Infinite Leagues Under the Sea: Photorealistic 3D Underwater Terrain Generation by Latent Fractal Diffusion ModelsDiffusion models → Inverse problems and image restoration
Dr Genre: Reinforcement Learning from Decoupled LLM Feedback for Generic Text RewritingLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Effectiveness of Zero-shot-CoT in Japanese PromptsLLM reasoning → Prompting and chain-of-thought reasoning
Primal-Dual Sample Complexity Bounds for Constrained Markov Decision Processes with Multiple ConstraintsDeep reinforcement learning methods → RL theory and sample complexity
Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language ModelsLLM reasoning → Multimodal reasoning +1
Fully-Decentralized MADDPG with Networked AgentsDeep reinforcement learning methods → RL for games and combinatorial optimization
Color Alignment in DiffusionDiffusion models → Image generation, text-to-image and editing
Beyond Black-Box Benchmarking: Observability, Analytics, and Optimization of Agentic SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
CoDa-4DGS: Dynamic Gaussian Splatting with Context and Deformation Awareness for Autonomous DrivingEnd-to-end autonomous driving → Generative simulation for driving
Diffusion Models are Secretly Zero-Shot 3DGS HarmonizersDiffusion models → 3D, point cloud and scene generation
Intelligent Spectrum Sharing in Integrated TN-NTNs: A Hierarchical Deep Reinforcement Learning ApproachDeep reinforcement learning methods → RL for energy, networks and operations
Alignment for Efficient Tool Calling of Large Language ModelsLLM agents and tool use → Tool learning and function calling
InftyThink: Breaking the Length Limits of Long-Context Reasoning in Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Censoring-Aware Tree-Based Reinforcement Learning for Estimating Dynamic Treatment Regimes with Censored OutcomesDeep reinforcement learning methods → RL for clinical treatment and surgery
Asymmetric Decision-Making in Online Knowledge Distillation:Unifying Consensus and DivergenceKnowledge distillation → General knowledge distillation methods
PixelPonder: Dynamic Patch Adaptation for Enhanced Multi-Conditional Text-to-Image GenerationDiffusion models → Image generation, text-to-image and editing
Learning Few-Step Diffusion Models by Trajectory Distribution MatchingDiffusion models → Diffusion foundations and sampling
Attention, Please! PixelSHAP Reveals What Vision-Language Models Actually Focus OnMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
Exploring LLM Agents for Cleaning Tabular Machine Learning DatasetsLLM agents and tool use → Web, search, and research agents
AxisPose: Model-Free Matching-Free Single-Shot 6D Object Pose Estimation via Axis GenerationDiffusion models → 3D, point cloud and scene generation
Adding Additional Control to One-Step Diffusion with Joint Distribution MatchingDiffusion models → Diffusion foundations and sampling
Evaluating and Aligning Human Economic Risk Preferences in LLMsLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Reinforcement Learning with Verifiable Rewards: GRPO's Effective Loss, Dynamics, and Success AmplificationLLM reasoning → Reinforcement learning for reasoning
CLAD: Constrained Latent Action Diffusion for Vision-Language Procedure PlanningDiffusion models → Diffusion and flow policies for robotics
DiffCLIP: Differential Attention Meets CLIPMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Inverse Reinforcement Learning for Minimum-Exposure Paths in Spatiotemporally Varying Scalar FieldsOffline RL and imitation learning from data → Imitation and offline RL for driving +1
Beyond Decoder-only: Large Language Models Can be Good Encoders for Machine TranslationEfficient LLM inference and serving → Speculative decoding
Automated Proof of Polynomial Inequalities via Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
Agent models: Internalizing Chain-of-Action Generation into Reasoning modelsLLM reasoning → Reasoning for agents, planning and code +1
Non-Equilibrium MAV-Capture-MAV via Time-Optimal Planning and Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Global-Aware Monocular Semantic Scene Completion with State Space ModelsState-space sequence models (S4, Mamba) → Vision Mamba backbones
Conceptrol: Concept Control of Zero-shot Personalized Image GenerationDiffusion models → Image generation, text-to-image and editing
Human Cognition Inspired RAG with Knowledge Graph for Complex Problem SolvingRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Future-Aware Interaction Network For Motion ForecastingState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
TR-DQ: Time-Rotation Diffusion QuantizationDiffusion models → Diffusion foundations and sampling
Generative modelling with jump-diffusionsDiffusion models → Diffusion foundations and sampling
ProJudge: A Multi-Modal Multi-Discipline Benchmark and Instruction-Tuning Dataset for MLLM-based Process JudgesLLM reasoning → Multimodal reasoning +1
QuantCache: Adaptive Importance-Guided Quantization with Hierarchical Latent and Layer Caching for Video GenerationDiffusion models → Diffusion foundations and sampling
One-Step Diffusion Model for Image Motion-DeblurringDiffusion models → Inverse problems and image restoration
MetaXCR: Reinforcement-Based Meta-Transfer Learning for Cross-Lingual Commonsense ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive ReinforcementLLM reasoning → Multimodal reasoning
Can Small Language Models Reliably Resist Jailbreak Attacks? A Comprehensive EvaluationLLM safety and security → Jailbreaking
Towards Superior Quantization Accuracy: A Layer-sensitive ApproachLLM reasoning → Test-time compute and efficient thinking
GFlowVLM: Enhancing Multi-step Reasoning in Vision-Language Models with Generative Flow NetworksLLM reasoning → Multimodal reasoning +1
HFedCKD: Toward Robust Heterogeneous Federated Learning via Data-free Knowledge Distillation and Two-way ContrastKnowledge distillation → Federated, robust and private distillation
Evaluation of Safety Cognition Capability in Vision-Language Models for Autonomous DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
VisualSimpleQA: A Benchmark for Decoupled Evaluation of Large Vision-Language Models in Fact-Seeking Question AnsweringMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
MoFE: Mixture of Frozen Experts ArchitectureMixture-of-experts → Mixture of LoRA experts
PerturboLLaVA: Reducing Multimodal Hallucinations with Perturbative Visual TrainingMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
A Mesh Is Worth 512 Numbers: Spectral-domain Diffusion Modeling for High-dimension Shape GenerationDiffusion models → 3D, point cloud and scene generation
CalliReader: Contextualizing Chinese Calligraphy via an Embedding-Aligned Vision-Language ModelMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Long-tailed Adversarial Training with Self-DistillationKnowledge distillation → Federated, robust and private distillation
Efficient Input-level Backdoor Defense on Text-to-Image Synthesis via Neuron Activation VariationDiffusion models → Robustness, privacy and adversarial aspects +1
M3amba: CLIP-driven Mamba Model for Multi-modal Remote Sensing ClassificationState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
Seesaw: High-throughput LLM Inference via Model Re-shardingEfficient LLM inference and serving → Serving systems and request scheduling
Graph Retrieval-Augmented LLM for Conversational Recommendation SystemsRetrieval-augmented generation → RAG for recommendation
Federated Learning for Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
Consistent Image Layout Editing with Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Swift Hydra: Self-Reinforcing Generative Framework for Anomaly Detection with Multiple Mamba ModelsMixture-of-experts → Vision mixture-of-experts +1
Performant LLM Agentic Framework for Conversational AILLM agents and tool use → General LLM agent frameworks and benchmarks
How a Bilingual LM Becomes Bilingual: Tracing Internal Representations with Sparse AutoencodersMechanistic interpretability → SAE features and their applications
EPR-GAIL: An EPR-Enhanced Hierarchical Imitation Learning Framework to Simulate Complex User Consumption BehaviorsOffline RL and imitation learning from data → Imitation learning methods
ProSE: Diffusion Priors for Speech EnhancementDiffusion models → Audio, speech and music
Dynamic Load Balancing for EV Charging Stations Using Reinforcement Learning and Demand PredictionDeep reinforcement learning methods → RL for energy, networks and operations
Adaptive Audio-Visual Speech Recognition via Matryoshka-Based Multimodal LLMsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Conformance Checking for Less: Efficient Conformance Checking for Long Event Sequences
HCT-QA: A Benchmark for Question Answering on Human-Centric Tables
Non-Canonical Crosslinks Confound Evolutionary Protein Structure Models
PythonPal: Enhancing Online Programming Education through Chatbot-Driven Personalized Feedback
Accodemy: AI Powered Code Learning Platform to Assist Novice Programmers in Overcoming the Fear of Coding
Statistical Study of Sensor Data and Investigation of ML-based Calibration Algorithms for Inexpensive Sensor Modules: Experiments from Cape Point
Machine learning for triage of strokes with large vessel occlusion using photoplethysmography biomarkers
Green Prompting: Characterizing Prompt-driven Energy Costs of LLM Inference
Semantic Wave Functions: Exploring Meaning in Large Language Models Through Quantum Formalism
Optimal Transport for Brain-Image Alignment: Unveiling Redundancy and Synergy in Neural Information Processing
A Block-Based Heuristic Algorithm for the Three-Dimensional Nuclear Waste Packing Problem
SDTrack: A Baseline for Event-based Tracking via Spiking Neural Networks
WECAR: An End-Edge Collaborative Inference and Training Framework for WiFi-Based Continuous Human Activity Recognition
CLIMB: Data Foundations for Large Scale Multimodal Clinical Foundation Models
Multimodal Emotion Recognition and Sentiment Analysis in Multi-Party Conversation Contexts
Actionable AI: Enabling Non Experts to Understand and Configure AI Systems
VideoPhy-2: A Challenging Action-Centric Physical Commonsense Evaluation in Video Generation
Characterizing Learning in Spiking Neural Networks with Astrocyte-Like Units
Multimodal AI-driven Biomarker for Early Detection of Cancer Cachexia
Robotic Ultrasound-Guided Femoral Artery Reconstruction of Anatomically-Representative Phantoms
On the Mutual Influence of Gender and Occupation in LLM Representations
Dubito Ergo Sum: Exploring AI Ethics
Large Language Models Are Effective Human Annotation Assistants, But Not Good Independent Annotators
Agile Climate-Sensor Design and Calibration Algorithms Using Machine Learning: Experiments From Cape Point
Investigating Image Manifolds of 3D Objects: Learning, Shape Analysis, and Comparisons
Task-Oriented Connectivity for Networked Robotics with Generative AI and Semantic Communications

All 226