Papers

New on arXiv

Thu 27 Feb 2025

413 papers

Evaluating System 1 vs. 2 Reasoning Approaches for Zero-Shot Time Series Forecasting: A Benchmark and InsightsLLM reasoning → Time series reasoning
LIVS: A Pluralistic Alignment Dataset for Inclusive Public SpacesLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
MMSciBench: Benchmarking Language Models on Chinese Multimodal Scientific ProblemsLLM reasoning → Multimodal reasoning +1
When Continue Learning Meets Multimodal Large Language Model: A SurveyMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Societal Alignment Frameworks Can Improve LLM AlignmentLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
CRFU: Compressive Representation Forgetting Against Privacy Leakage on Machine UnlearningLLM safety and security → Machine unlearning
Adaptive Attacks Break Defenses Against Indirect Prompt Injection Attacks on LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
Multi2: Multi-Agent Test-Time Scalable Framework for Multi-Document ProcessingLLM reasoning → Reasoning benchmarks and evaluation +1
Cache-of-Thought: Master-Apprentice Framework for Cost-Effective Vision Language Model ReasoningLLM reasoning → Multimodal reasoning +1
Interpreting CLIP with Hierarchical Sparse AutoencodersMechanistic interpretability → Interpreting vision and multimodal models
Visual Reasoning at Urban Intersections: FineTuning GPT-4o for Traffic Conflict DetectionLLM reasoning → Chain-of-thought reasoning for driving
HazardNet: A Small-Scale Vision Language Model for Real-Time Traffic Safety Detection at Edge DevicesMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
Towards Statistical Factuality Guarantee for Large Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Close-Proximity Satellite Operations through Deep Reinforcement Learning and Terrestrial Testing EnvironmentsDeep reinforcement learning methods → RL for robot navigation and UAV control
Qsharp: Provably Optimal Distributional RL for LLM Post-TrainingLLM reasoning → Reinforcement learning for reasoning +1
SoS1: O1 and R1-Like Reasoning LLMs are Sum-of-Square SolversLLM reasoning → Mathematical reasoning
NANOGPT: A Query-Driven Large Language Model Retrieval-Augmented Generation System for Nanotechnology ResearchRetrieval-augmented generation → Agentic RAG, agents and memory
Supervised Fine-Tuning LLMs to Behave as Pedagogical Agents in Programming EducationLLM agents and tool use → Educational agents and math problem solving
Personas Evolved: Designing Ethical LLM-Based Conversational Agent PersonalitiesLLM agents and tool use → Social simulation and multi-agent debate
CoCa-CXR: Contrastive Captioners Learn Strong Temporal Structures for Chest X-Ray Vision-Language UnderstandingMultimodal LLMs / vision-language models → Medical vision-language models
A Thousand Words or An Image: Studying the Influence of Persona Modality in Multimodal LLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Protecting multimodal large language models against misleading visualizationsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Equivariant Reinforcement Learning Frameworks for Quadrotor Low-Level ControlDeep reinforcement learning methods → RL for robot navigation and UAV control
Unifying Model Predictive Path Integral Control, Reinforcement Learning, and Diffusion Models for Optimal Control and PlanningDiffusion models → Diffusion and flow policies for robotics +1
LLM Strategic Reasoning: Agentic Study through Behavioral Game TheoryLLM reasoning → Reasoning for agents, planning and code +1
SEKI: Self-Evolution and Knowledge Inspiration based Neural Architecture Search via Large Language ModelsKnowledge distillation → Knowledge distillation for LLMs
Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on HumanoidsDeep reinforcement learning methods → RL for robot locomotion and manipulation
R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-ExpertsLLM reasoning → Multimodal reasoning +2
Why Are Web AI Agents More Vulnerable Than Standalone LLMs? A Security AnalysisLLM agents and tool use → Agent threats, guardrails and governance
Multi-Turn Code Generation Through Single-Step RewardsLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
Multi-Agent Verification: Scaling Test-Time Compute with Multiple VerifiersLLM reasoning → Test-time compute and efficient thinking +1
PhantomWiki: On-Demand Datasets for Reasoning and Retrieval EvaluationLLM reasoning → Reasoning benchmarks and evaluation
Tight Inversion: Image-Conditioned Inversion for Real Image EditingDiffusion models → Image generation, text-to-image and editing
Constrained Generative Modeling with Manually Bridged Diffusion ModelsDiffusion models → Diffusion foundations and sampling
The Role of Tactile Sensing for Learning Reach and GraspDeep reinforcement learning methods → RL for robot locomotion and manipulation
Bridging Legal Knowledge and AI: Retrieval-Augmented Generation with Vector Stores, Knowledge Graphs, and Hierarchical Non-negative Matrix FactorizationRetrieval-augmented generation → Legal and financial RAG +1
Bridging the Creativity Understanding Gap: Small-Scale Human Alignment Enables Expert-Level Humor Ranking in LLMsLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Improving the Efficiency of a Deep Reinforcement Learning-Based Power Management System for HPC Clusters Using Curriculum LearningDeep reinforcement learning methods → RL for energy, networks and operations
Safety Representations for Safer Policy LearningDeep reinforcement learning methods → Safe and constrained RL
Thinking Slow, Fast: Scaling Inference Compute with Distilled ReasonersLLM reasoning → Test-time compute and efficient thinking +1
Emergent Symbolic Mechanisms Support Abstract Reasoning in Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
RAPID: Long-Context Inference with Retrieval-Augmented Speculative DecodingRetrieval-augmented generation → Generation and context integration +1
Deep Reinforcement Learning based Autonomous Decision-Making for Cooperative UAVs: A Search and Rescue Real World ApplicationDeep reinforcement learning methods → RL for robot navigation and UAV control
Mobius: Text to Seamless Looping Video Generation via Latent ShiftDiffusion models → Video generation
M^3Builder: A Multi-Agent System for Automated Machine Learning in Medical ImagingLLM agents and tool use → Medical and scientific domain agents
Judge a Book by its Cover: Investigating Multi-Modal LLMs for Multi-Page Handwritten Document TranscriptionMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
Visual Adaptive Prompting for Compositional Zero-Shot LearningMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Conformal Tail Risk Control for Large Language Model AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
On the Importance of Reward Design in Reinforcement Learning-based Dynamic Algorithm Configuration: A Case Study on OneMax with (1+(λ,λ))-GADeep reinforcement learning methods → Deep RL algorithms and representations
From Retrieval to Generation: Comparing Different ApproachesRetrieval-augmented generation → RAG robustness and adaptive retrieval
FINEREASON: Evaluating and Improving LLMs' Deliberate Reasoning through Reflective Puzzle SolvingLLM reasoning → Mathematical reasoning
Attention Distillation: A Unified Approach to Visual Characteristics TransferDiffusion models → Image generation, text-to-image and editing
Mixture of Experts for Recognizing Depression from Interview and Reading TasksMixture-of-experts → Multimodal MoE models
ChineseEcomQA: A Scalable E-commerce Concept Evaluation Benchmark for Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
Highly Parallelized Reinforcement Learning Training with Relaxed Assignment DependenciesDeep reinforcement learning methods → Deep RL algorithms and representations
Mixture of Experts-augmented Deep Unfolding for Activity Detection in IRS-aided SystemsMixture-of-experts → Expert routing strategies
An Extensive Evaluation of PDDL Capabilities in off-the-shelf LLMsLLM reasoning → Reasoning for agents, planning and code
Multimodal Representation Alignment for Image Generation: Text-Image Interleaved Control Is Easier Than You ThinkDiffusion models → Image generation, text-to-image and editing
Accelerating Model-Based Reinforcement Learning with State-Space World ModelsWorld models → Latent dynamics models and model-based RL +1
Finite State Automata Inside Transformers with Chain-of-Thought: A Mechanistic Study on State TrackingLLM reasoning → Prompting and chain-of-thought reasoning
SoRFT: Issue Resolving with Subtask-oriented Reinforced Fine-TuningLLM reasoning → Reasoning for agents, planning and code
FlexiDiT: Your Diffusion Transformer Can Easily Generate High-Quality Samples with Less ComputeDiffusion models → Diffusion foundations and sampling
Self-Training Elicits Concise Reasoning in Large Language ModelsLLM reasoning → Test-time compute and efficient thinking +1
VDT-Auto: End-to-end Autonomous Driving with VLM-Guided Diffusion TransformersDiffusion models → Diffusion and flow policies for robotics +1
Generative augmentations for improved cardiac ultrasound segmentation using diffusion modelsDiffusion models → Medical imaging and segmentation
RIZE: Adaptive Regularization for Imitation LearningOffline RL and imitation learning from data → Inverse reinforcement learning +1
Collab-Overcooked: Benchmarking and Evaluating Large Language Models as Collaborative AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
RouteRL: Multi-agent reinforcement learning framework for urban route choice with autonomous vehiclesEnd-to-end autonomous driving → Reinforcement learning for driving
AsymLoRA: Harmonizing Data Conflicts and Commonalities in MLLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Do Vision Encoders Truly Explain Object Hallucination?: Mitigating Object Hallucination via Simple Fine-Grained CLIPScoreMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Offline Reinforcement Learning via Inverse OptimizationOffline RL and imitation learning from data → Inverse reinforcement learning
Robust Mean Field Social Control: A Unified Reinforcement Learning FrameworkDeep reinforcement learning methods → RL for games and combinatorial optimization
Erasing Without Remembering: Implicit Knowledge Forgetting in Large Language ModelsLLM safety and security → Machine unlearning
Convergence Guarantees of Model-free Policy Gradient Methods for LQR with Stochastic DataDeep reinforcement learning methods → RL for games and combinatorial optimization
Do Sparse Autoencoders Generalize? A Case Study of AnswerabilityMechanistic interpretability → SAE features and their applications
ChatReID: Open-ended Interactive Person Retrieval via Hierarchical Progressive Tuning for Vision Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Space Rotation with Basis Transformation for Training-free Test-Time AdaptationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Alleviating Distribution Shift in Synthetic Data for Machine Translation Quality EstimationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
DiffCSS: Diverse and Expressive Conversational Speech Synthesis with Diffusion ModelsDiffusion models → Audio, speech and music
Pokemon Red via Reinforcement LearningDeep reinforcement learning methods → RL for games and combinatorial optimization
Meta-Reasoner: Dynamic Guidance for Optimized Inference-time Reasoning in Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Picking the Cream of the Crop: Visual-Centric Data Selection with Collaborative AgentsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
CarPlanner: Consistent Auto-regressive Trajectory Planning for Large-scale Reinforcement Learning in Autonomous DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners +1
Order Doesn't Matter, But Reasoning Does: Training LLMs with Order-Centric AugmentationLLM reasoning → Reasoning benchmarks and evaluation
Optimus-2: Multimodal Minecraft Agent with Goal-Observation-Action Conditioned PolicyLLM agents and tool use → Reinforcement learning for LLM agents
High-Fidelity Relightable Monocular Portrait Animation with Lighting-Controllable Video Diffusion ModelDiffusion models → Audio, speech and music
ColorDynamic: Generalizable, Scalable, Real-time, End-to-end Local Planner for Unstructured and Dynamic EnvironmentsDeep reinforcement learning methods → RL for robot navigation and UAV control
Beyond the Tip of Efficiency: Uncovering the Submerged Threats of Jailbreak Attacks in Small Language ModelsLLM safety and security → Jailbreaking
MIND: Towards Immersive Psychological Healing with Multi-agent Inner DialogueLLM agents and tool use → Medical and scientific domain agents
IL-SOAR : Imitation Learning with Soft Optimistic Actor cRiticOffline RL and imitation learning from data → Imitation learning methods +1
One-for-More: Continual Diffusion Model for Anomaly DetectionDiffusion models → Flow matching, time series and science
ProAPO: Progressively Automatic Prompt Optimization for Visual ClassificationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Revisiting Self-Consistency from Dynamic Distributional Alignment Perspective on Answer AggregationLLM reasoning → Test-time compute and efficient thinking
Foot-In-The-Door: A Multi-turn Jailbreak for LLMsLLM safety and security → Jailbreaking
Comet: Fine-grained Computation-communication Overlapping for Mixture-of-ExpertsMixture-of-experts → Efficient MoE training and inference systems
Implicit Search via Discrete Diffusion: A Study on ChessDiffusion models → Discrete and text diffusion
MFSR: Multi-fractal Feature for Super-resolution Reconstruction with Fine Details RecoveryDiffusion models → Inverse problems and image restoration
SCU: An Efficient Machine Unlearning Scheme for Deep Learning Enabled Semantic CommunicationsLLM safety and security → Machine unlearning
Do Retrieval-Augmented Language Models Adapt to Varying User Needs?Retrieval-augmented generation → Generation and context integration
InPK: Infusing Prior Knowledge into Prompt for Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP

All 413