Papers

New on arXiv

Mon 3 Feb 2025

389 papers

Agent-S: LLM Agentic workflow to automate Standard Operating ProceduresLLM agents and tool use → Software engineering coding agents
Explainable Sentiment Analysis with DeepSeek-R1: Performance, Efficiency, and Few-Shot LearningLLM reasoning → Reasoning benchmarks and evaluation
Competitive Programming with Large Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
Efficient Diffusion Models: A SurveyDiffusion models → Diffusion foundations and sampling
Reinforcement Learning with Segment FeedbackDeep reinforcement learning methods → RL training of language and agent models
UVGS: Reimagining Unstructured 3D Gaussian Splatting using UV MappingDiffusion models → 3D, point cloud and scene generation
Diffusion Model for Multiple Antenna CommunicationsDiffusion models → Diffusion for compression and wireless
Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling VerificationLLM reasoning → Test-time compute and efficient thinking
From Foresight to Forethought: VLM-In-the-Loop Policy Steering via Latent AlignmentLLM reasoning → Spatial and embodied reasoning
Firewalls to Secure Dynamic LLM Agentic NetworksLLM agents and tool use → Agent threats, guardrails and governance
Agentic Bug Reproduction for Effective Automated Program Repair at GoogleLLM agents and tool use → Software engineering coding agents
Score as Action: Fine-Tuning Diffusion Generative Models by Continuous-time Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +2
Flow-based Domain Randomization for Learning and Sequencing Robotic SkillsDeep reinforcement learning methods → Safe and constrained RL
An Agentic AI Workflow for Detecting Cognitive Concerns in Real-world DataLLM agents and tool use → Medical and scientific domain agents
AquaticCLIP: A Vision-Language Foundation Model for Underwater Scene AnalysisMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
VILP: Imitation Learning with Latent Video PlanningOffline RL and imitation learning from data → Robot manipulation imitation learning
GNN-DT: Graph Neural Network Enhanced Decision Transformer for Efficient Optimization in Dynamic EnvironmentsOffline RL and imitation learning from data → Decision transformers and sequence modeling +1
Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal SparsityDiffusion models → Video generation
Coarse-to-Fine 3D Keyframe TransporterOffline RL and imitation learning from data → Robot manipulation imitation learning
MJ-VIDEO: Fine-Grained Benchmarking and Rewarding Video Preferences in Video GenerationMixture-of-experts → Vision mixture-of-experts
ACECODER: Acing Coder RL via Automated Test-Case SynthesisLLM reasoning → Reasoning for agents, planning and code +2
Position: Towards a Responsible LLM-empowered Multi-Agent SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
BARE: Leveraging Base Language Models for Few-Shot Synthetic Data GenerationSynthetic data and self-improvement for LLMs → LLM synthetic data generation
SliderSpace: Decomposing the Visual Capabilities of Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
The AI Agent IndexLLM agents and tool use → Agent threats, guardrails and governance
Online Gradient Boosting Decision Tree: In-Place Updates for Efficient Adding/Deleting DataLLM safety and security → Machine unlearning
Adversarial Reasoning at Jailbreaking TimeLLM reasoning → Reasoning for agents, planning and code +1
TReMu: Towards Neuro-Symbolic Temporal Reasoning for LLM-Agents with Memory in Multi-Session DialoguesLLM reasoning → Knowledge- and tool-augmented reasoning +1
MFP-VTON: Enhancing Mask-Free Person-to-Person Virtual Try-On via Diffusion TransformerDiffusion models → 3D, point cloud and scene generation
Rollout Roulette: A Probabilistic Inference Approach to Inference-Time Scaling of LLMs using Particle-Based Monte Carlo MethodsLLM reasoning → Test-time compute and efficient thinking +1
Preference VLM: Leveraging VLMs for Scalable Preference-Based Reinforcement LearningLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Large Language Models Are Human-Like InternallyMechanistic interpretability → Knowledge and representations inside LLMs
Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization ChallengesSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
Reinforcement Learning for Long-Horizon Interactive LLM AgentsLLM agents and tool use → Reinforcement learning for LLM agents
Improving Transformer World Models for Data-Efficient RLWorld models → Latent dynamics models and model-based RL
ReasoningWeekly: A General Knowledge and Verbal Reasoning Challenge for Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Robust-LLaVA: On the Effectiveness of Large-Scale Robust Image Encoders for Multi-modal Large Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
MakeAnything: Harnessing Diffusion Transformers for Multi-Domain Procedural Sequence GenerationDiffusion models → Image generation, text-to-image and editing
Latent Thought Models with Variational Bayes Inference-Time ComputationLLM reasoning → Test-time compute and efficient thinking
Memento No More: Coaching AI Agents to Master Multiple Tasks via Hints InternalizationLLM agents and tool use → Agent memory systems
Search-Based Adversarial Estimates for Improving Sample Efficiency in Off-Policy Reinforcement LearningDeep reinforcement learning methods → Adversarial robustness and RL in finance
VideoRAG: Retrieval-Augmented Generation with Extreme Long-Context VideosRetrieval-augmented generation → Multimodal RAG
Dynamic object goal pushing with mobile manipulators through model-free constrained reinforcement learningDeep reinforcement learning methods → RL for robot locomotion and manipulation
VR-Robo: A Real-to-Sim-to-Real Framework for Visual Robot Navigation and LocomotionDeep reinforcement learning methods → RL for robot locomotion and manipulation
The in-context inductive biases of vision-language models differ across modalitiesMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Efficiently Integrate Large Language Models with Visual Perception: A Survey from the Training Paradigm PerspectiveMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Unpaired Deblurring via Decoupled Diffusion ModelDiffusion models → Inverse problems and image restoration
Symmetry-Guided Memory Augmentation for Efficient Locomotion LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
Memorization Inheritance in Sequence-Level Knowledge Distillation for Neural Machine TranslationKnowledge distillation → Knowledge distillation for LLMs
Achieving Time Series Reasoning Requires Rethinking Model Design, Tasks Formulation, and EvaluationLLM reasoning → Time series reasoning
Generalization Error Analysis for Selective State-Space Models Through the Lens of AttentionState-space sequence models (S4, Mamba) → Core SSM architectures for language
FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language ModelLLM alignment / RLHF / preference optimisation → LLM preference optimization methods +1
Embrace Collisions: Humanoid Shadowing for Deployable Contact-Agnostics MotionsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Process Reinforcement through Implicit RewardsLLM reasoning → Reinforcement learning for reasoning +1
Simulating Rumor Spreading in Social Networks using LLM AgentsLLM agents and tool use → Social simulation and multi-agent debate
Improved Training Technique for Latent Consistency ModelsDiffusion models → Diffusion foundations and sampling
Visual Attention Never Fades: Selective Progressive Attention ReCalibration for Detailed Image Captioning in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Assessing the use of Diffusion models for motion artifact correction in brain MRIDiffusion models → Medical imaging and segmentation
Human Body Restoration with One-Step Diffusion Model and A New BenchmarkDiffusion models → Inverse problems and image restoration
Plan-Then-Execute: An Empirical Study of User Trust and Team Performance When Using LLM Agents As A Daily AssistantLLM agents and tool use → General LLM agent frameworks and benchmarks
TeLL-Drive: Enhancing Autonomous Driving with Teacher LLM-Guided Deep Reinforcement LearningEnd-to-end autonomous driving → Large language models for driving
Topic-FlipRAG: Topic-Orientated Adversarial Opinion Manipulation Attacks to Retrieval-Augmented Generation ModelsRetrieval-augmented generation → Privacy and security of RAG
Detecting Backdoor Samples in Contrastive Language Image PretrainingLLM safety and security → Backdoor and data poisoning attacks
Fine-Tuning Discrete Diffusion Models with Policy Gradient MethodsDiffusion models → Diffusion and flow policies for robotics +1
Trajectory World Models for Heterogeneous EnvironmentsWorld models → Latent predictive world models
PSSD: Making Large Language Models Self-denial via Human Psyche StructureLLM reasoning → Reasoning benchmarks and evaluation
AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Document UnderstandingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Learning Fused State Representations for Control from Multi-View ObservationsDeep reinforcement learning methods → RL for robot locomotion and manipulation
CleanPose: Category-Level Object Pose Estimation via Causal Learning and Knowledge DistillationKnowledge distillation → Detection and dense prediction distillation
Heterogeneous Image GNN: Graph-Conditioned Diffusion for Image SynthesisDiffusion models → Diffusion foundations and sampling
Improving the Effectiveness of Potential-Based Reward Shaping in Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
Exploratory Utility Maximization Problem with Tsallis EntropyDeep reinforcement learning methods → Adversarial robustness and RL in finance
Resilient UAV Trajectory Planning via Few-Shot Meta-Offline Reinforcement LearningOffline RL and imitation learning from data → Offline multi-agent RL
Can Reinforcement Learning Solve Asymmetric Combinatorial-Continuous Zero-Sum Games?Deep reinforcement learning methods → RL for games and combinatorial optimization
Peering Behind the Shield: Guardrail Identification in Large Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
The Differences Between Direct Alignment Algorithms are a BlurLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Eliciting Language Model Behaviors with Investigator AgentsLLM agents and tool use → Medical and scientific domain agents
The dark deep side of DeepSeek: Fine-tuning attacks against the safety alignment of CoT-enabled modelsLLM reasoning → Reasoning for agents, planning and code +2
Provable Ordering and Continuity in Vision-Language Pretraining for Generalizable Embodied AgentsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
On Almost Surely Safe Alignment of Large Language Models at Inference-TimeLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
KL-Regularized RLHF with Multiple Reference Models: Exact Solutions and Sample ComplexityLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Compressed Image Generation with Denoising Diffusion Codebook ModelsDiffusion models → Diffusion for compression and wireless
Deep Active Speech Cancellation with Mamba-Masking NetworkState-space sequence models (S4, Mamba) → Speech and audio SSMs
AtmosSci-Bench: Evaluating the Recent Advance of Large Language Model for Atmospheric ScienceLLM reasoning → Time series reasoning
MIND: Modality-Informed Knowledge Distillation Framework for Multimodal Clinical Prediction TasksKnowledge distillation → Distillation for medical imaging
Jailbreaking with Universal Multi-PromptsLLM safety and security → Jailbreaking
Gradient Norm-based Fine-Tuning for Backdoor Defense in Automatic Speech RecognitionLLM safety and security → Audio jailbreaks and speech backdoors
DeepRAG: Thinking to Retrieve Step by Step for Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning +1
Deep Reinforcement Learning for Dynamic Resource Allocation in Wireless NetworksDeep reinforcement learning methods → RL for energy, networks and operations
Picky LLMs and Unreliable RMs: An Empirical Study on Safety Alignment after Instruction TuningLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
GFM-RAG: Graph Foundation Model for Retrieval Augmented GenerationRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
LayerTracer: Cognitive-Aligned Layered SVG Synthesis via Diffusion TransformerDiffusion models → Image generation, text-to-image and editing
ZebraLogic: On the Scaling Limits of LLMs for Logical ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Tool Unlearning for Tool-Augmented LLMsLLM agents and tool use → Tool learning and function calling +1
The Jumping Reasoning Curve? Tracking the Evolution of Reasoning Performance in GPT-[n] and o-[n] Models on Multimodal PuzzlesLLM reasoning → Multimodal reasoning
Omni-Mol: Multitask Molecular Model for Any-to-any ModalitiesMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Scalable, Training-Free Visual Language Robotics: A Modular Multi-Model Framework for Consumer-Grade GPUsVision-language-action models → Generalist VLA models and benchmarks
FastKV: Decoupling of Context Reduction and KV Cache Compression for Prefill-Decoding AccelerationEfficient LLM inference and serving → KV cache optimization
Mitigating Hallucinations in Large Vision-Language Models with Internal Fact-based Contrastive DecodingMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Diffusion Model as a Noise-Aware Latent Reward Model for Step-Level Preference OptimizationDiffusion models → Diffusion and flow policies for robotics +1

All 389