Papers

New on arXiv

Thu 30 May 2024

409 papers

Temporal Grounding of Activities using Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Video understanding with MLLMs
CLAY: A Controllable Large-scale Generative Model for Creating High-quality 3D AssetsDiffusion models → 3D, point cloud and scene generation
SleeperNets: Universal Backdoor Poisoning Attacks Against Reinforcement Learning AgentsLLM safety and security → Backdoor and data poisoning attacks +1
Q-learning as a monotone schemeDeep reinforcement learning methods → RL theory and sample complexity
Unveiling the Impact of Coding Data Instruction Fine-Tuning on Large Language Models ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Aquatic Navigation: A Challenging Benchmark for Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Diffusion On Syntax Trees For Program SynthesisDiffusion models → Discrete and text diffusion
Hybrid Reinforcement Learning Framework for Mixed-Variable ProblemsDeep reinforcement learning methods → RL theory and sample complexity
Transfer Q Star: Principled Decoding for LLM AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Slight Corruption in Pre-training Data Makes Better Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Phantom: General Backdoor Attacks on Retrieval Augmented Language GenerationRetrieval-augmented generation → Privacy and security of RAG +1
Performance of NPG in Countable State-Space Average-Cost RLDeep reinforcement learning methods → RL theory and sample complexity
DepsRAG: Towards Agentic Reasoning and Planning for Software Dependency ManagementLLM reasoning → Reasoning for agents, planning and code +1
Is My Data in Your Retrieval Database? Membership Inference Attacks Against Retrieval Augmented GenerationRetrieval-augmented generation → Privacy and security of RAG +1
P-MSDiff: Parallel Multi-Scale Diffusion for Remote Sensing Image SegmentationDiffusion models → Inverse problems and image restoration
Worse than Random? An Embarrassingly Simple Probing Evaluation of Large Multimodal Models in Medical VQAMultimodal LLMs / vision-language models → Medical vision-language models
Jailbreaking Large Language Models Against Moderation Guardrails via Cipher CharactersLLM safety and security → Jailbreaking
Gradient Inversion of Federated Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
LLMGeo: Benchmarking Large Language Models on Image Geolocation In-the-wildMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Unique3D: High-Quality and Efficient 3D Mesh Generation from a Single ImageDiffusion models → 3D, point cloud and scene generation
Visual Perception by Large Language Model's WeightsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
OccSora: 4D Occupancy Generation Models as World Simulators for Autonomous DrivingEnd-to-end autonomous driving → Generative simulation for driving
Xwin-LM: Strong and Scalable Alignment Practice for LLMsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
MotionFollower: Editing Video Motion via Lightweight Score-Guided DiffusionDiffusion models → Video generation
Don't drop your samples! Coherence-aware training benefits Conditional diffusionDiffusion models → Diffusion foundations and sampling
textitS3Gaussian: Self-Supervised Street Gaussians for Autonomous DrivingWorld models → World models for autonomous driving
Improving the Training of Rectified FlowsDiffusion models → Flow matching, time series and science
S3D: A Simple and Cost-Effective Self-Speculative Decoding Scheme for Low-Memory GPUsEfficient LLM inference and serving → Speculative decoding
Sequence-Augmented SE(3)-Flow Matching For Conditional Protein Backbone GenerationDiffusion models → Flow matching, time series and science
Large Language Models Can Self-Improve At Web Agent TasksLLM agents and tool use → Web, search, and research agents +1
Group Robust Preference Optimization in Reward-free RLHFLLM reasoning → Reinforcement learning for reasoning +1
DITTO-2: Distilled Diffusion Inference-Time T-Optimization for Music GenerationDiffusion models → Audio, speech and music
SemFlow: Binding Semantic Segmentation and Image Synthesis via Rectified FlowDiffusion models → Medical imaging and segmentation
CV-VAE: A Compatible Video VAE for Latent Generative Video ModelsDiffusion models → Diffusion foundations and sampling
Reconstruction Attacks on Machine Unlearning: Simple Models are VulnerableLLM safety and security → Machine unlearning
Auto-Arena: Automating LLM Evaluations with Agent Peer Battles and Committee DiscussionsLLM agents and tool use → General LLM agent frameworks and benchmarks
Towards Hierarchical Multi-Agent Workflows for Zero-Shot Prompt OptimizationLLM agents and tool use → General LLM agent frameworks and benchmarks
Retrieval Augmented Structured Generation: Business Document Information Extraction As Tool UseRetrieval-augmented generation → Multimodal RAG
MOFA-Video: Controllable Image Animation via Generative Motion Field Adaptions in Frozen Image-to-Video Diffusion ModelDiffusion models → Video generation
Boost Your Human Image Generation Model via Direct Preference OptimizationLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
TS-Align: A Teacher-Student Collaborative Framework for Scalable Iterative Finetuning of Large Language ModelsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods +1
Transformers and Slot Encoding for Sample Efficient Physical World ModellingWorld models → Latent predictive world models
Randomized Exploration for Reinforcement Learning with Multinomial Logistic Function ApproximationDeep reinforcement learning methods → RL theory and sample complexity
Reasoning about concepts with LLMs: Inconsistencies aboundLLM reasoning → Knowledge- and tool-augmented reasoning
MotionDreamer: Exploring Semantic Video Diffusion features for Zero-Shot 3D Mesh AnimationDiffusion models → Video generation
Uncovering Bias in Large Vision-Language Models at Scale with CounterfactualsMultimodal LLMs / vision-language models → Social bias and fairness in VLMs
MSSC-BiMamba: Multimodal Sleep Stage Classification and Early Diagnosis of Sleep Disorders with Bidirectional MambaState-space sequence models (S4, Mamba) → Physiological signal modeling
OpenDAS: Open-Vocabulary Domain Adaptation for 2D and 3D SegmentationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
GNN-RAG: Graph Neural Retrieval for Large Language Model ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
Defensive Prompt Patch: A Robust and Interpretable Defense of LLMs against Jailbreak AttacksLLM safety and security → Jailbreaking
Transfer Attack for Bad and Good: Explain and Boost Adversarial Transferability across Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
NoiseBoost: Alleviating Hallucination with Noise Perturbation for Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Would I Lie To You? Inference Time Alignment of Language Models using Direct Preference HeadsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Safe Multi-agent Reinforcement Learning with Natural Language ConstraintsLLM agents and tool use → Reinforcement learning for LLM agents
Efficient LLM-Jailbreaking via Multimodal-LLM JailbreakLLM safety and security → Jailbreaking
DP-IQA: Utilizing Diffusion Prior for Blind Image Quality Assessment in the WildDiffusion models → Diffusion foundations and sampling
DiffPhysBA: Diffusion-based Physical Backdoor Attack against Person Re-Identification in Real-WorldLLM safety and security → Backdoor and data poisoning attacks
Video-Language Critic: Transferable Reward Functions for Language-Conditioned RoboticsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
PLA4D: Pixel-Level Alignments for Text-to-4D Gaussian SplattingDiffusion models → 3D, point cloud and scene generation
Exploring Diffusion Models' Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural NetworksDiffusion models → Diffusion foundations and sampling
Multimodal Cross-Domain Few-Shot Learning for Egocentric Action RecognitionKnowledge distillation → Multimodal distillation
Adaptive Advantage-Guided Policy Regularization for Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
Similarity is Not All You Need: Endowing Retrieval Augmented Generation with Multi Layered ThoughtsLLM reasoning → Knowledge- and tool-augmented reasoning +1
Fourier Controller Networks for Real-Time Decision-Making in Embodied LearningDeep reinforcement learning methods → Deep RL algorithms and representations
Learning from Random Demonstrations: Offline Reinforcement Learning with Importance-Sampled Diffusion ModelsWorld models → Latent dynamics models and model-based RL +2
Quest: Query-centric Data Synthesis Approach for Long-context Scaling of Large Language ModelSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Improve Student's Reasoning Generalizability through Cascading Decomposed CoTs DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
Efficient Stimuli Generation using Reinforcement Learning in Design VerificationDeep reinforcement learning methods → RL training of language and agent models
MetaCURL: Non-stationary Concave Utility Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
Preference Alignment with Flow MatchingLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
CharacterGPT: A Persona Reconstruction Framework for Role-Playing AgentsLLM agents and tool use → Social simulation and multi-agent debate
VQA Training Sets are Self-play Environments for Generating Few-shot PoolsSynthetic data and self-improvement for LLMs → Self-improving MLLMs and synthetic data
Enhancing Reinforcement Learning with Label-Sensitive Reward for Natural Language UnderstandingLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
HQ-DiT: Efficient Diffusion Transformer with FP4 Hybrid QuantizationDiffusion models → Diffusion foundations and sampling
X-Instruction: Aligning Language Model in Low-resource Languages with Self-curated Cross-lingual InstructionsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Beyond Imitation: Learning Key Reasoning Steps from Dual Chain-of-Thoughts in Reasoning DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
Intrinsic Dynamics-Driven Generalizable Scene Representations for Vision-Oriented Decision-Making ApplicationsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Streaming Video Diffusion: Online Video Editing with Diffusion ModelsDiffusion models → Video generation
Enhancing Large Vision Language Models with Self-Training on Image ComprehensionMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
SpecDec++: Boosting Speculative Decoding via Adaptive Candidate LengthsEfficient LLM inference and serving → Speculative decoding
Text Guided Image Editing with Automatic Concept Locating and ForgettingDiffusion models → Image generation, text-to-image and editing
Bilevel reinforcement learning via the development of hyper-gradient without lower-level convexityDeep reinforcement learning methods → RL theory and sample complexity
Grade Like a Human: Rethinking Automated Assessment with Large Language ModelsLLM reasoning → Mathematical reasoning
Diffusion Policies creating a Trust Region for Offline Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
Autonomous Driving with Spiking Neural NetworksEnd-to-end autonomous driving → End-to-end system architecture
Large Language Model Watermark Stealing With Mixed Integer ProgrammingLLM safety and security → Watermarking generated content and models
Knowledge-grounded Adaptation Strategy for Vision-language Models: Building Unique Case-set for Screening Mammograms for Residents TrainingMultimodal LLMs / vision-language models → Medical vision-language models
Bridging Model-Based Optimization and Generative Modeling via Conservative Fine-Tuning of Diffusion ModelsDiffusion models → Flow matching, time series and science
One Token Can Help! Learning Scalable and Pluggable Virtual Tokens for Retrieval-Augmented Large Language ModelsRetrieval-augmented generation → RAG pipeline design and evaluation
AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided OptimizationLLM safety and security → Jailbreaking
Unlocking the Power of Spatial and Temporal Information in Medical Multimodal Pre-trainingMultimodal LLMs / vision-language models → Medical vision-language models
SparseDrive: End-to-End Autonomous Driving via Sparse Scene RepresentationEnd-to-end autonomous driving → End-to-end system architecture
Easy Problems That LLMs Get WrongLLM reasoning → Reasoning benchmarks and evaluation
Keyword-driven Retrieval-Augmented Large Language Models for Cold-start User RecommendationsRetrieval-augmented generation → RAG for recommendation
Fair Machine Learning for Healthcare Requires Recognizing the Intersectionality of Sociodemographic Factors, a Case Study
Analysing the Public Discourse around OpenAI's Text-To-Video Model 'Sora' using Topic Modeling
Deep Learning Approaches for Detecting Adversarial Cyberbullying and Hate Speech in Social Networks
Applications of interpretable deep learning in neuroimaging: a comprehensive review
Utilizing Weak-to-Strong Consistency for Semi-Supervised Glomeruli Segmentation
Cutting Through the Noise: Boosting LLM Performance on Math Word Problems

All 409