Papers

New on arXiv

Wed 12 Jun 2024

366 papers

Self-Consuming Generative Models with Curated Data Provably Optimize Human PreferencesLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
Optimizing Deep Reinforcement Learning for Adaptive Robotic Arm ControlDeep reinforcement learning methods → RL for robot locomotion and manipulation
MobileAIBench: Benchmarking LLMs and LMMs for On-Device Use CasesMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Watermarking Language Models with Error Correcting CodesLLM safety and security → Watermarking generated content and models
Ad Auctions for LLMs via Retrieval Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
Pandora: Towards General World Model with Natural Language Actions and Video StatesWorld models → Video-generation world models and simulators
Advancing High Resolution Vision-Language Models in BiomedicineMultimodal LLMs / vision-language models → Medical vision-language models
Decision-Centered Abstractions via Orthogonal Estimation of Difference-of-Q FunctionsOffline RL and imitation learning from data → Offline RL algorithms +1
UnO: Unsupervised Occupancy Fields for Perception and ForecastingWorld models → World models for autonomous driving
Security of AI AgentsLLM agents and tool use → Agent threats, guardrails and governance
HelpSteer2: Open-source dataset for training top-performing reward modelsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Vivid-ZOO: Multi-View Video Generation with Diffusion ModelDiffusion models → Video generation
Mistral-C2F: Coarse to Fine Actor for Analytical and Reasoning Enhancement in RLHF and Effective-Merged LLMsLLM reasoning → Reasoning benchmarks and evaluation +1
How to Distinguish AI-Generated Images from Authentic PhotographsDiffusion models → Image generation, text-to-image and editing
Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit DifferenceLLM safety and security → Machine unlearning
FakeInversion: Learning to Detect Images from Unseen Text-to-Image Models by Inverting Stable DiffusionDiffusion models → Image generation, text-to-image and editing
DiTFastAttn: Attention Compression for Diffusion Transformer ModelsDiffusion models → Video generation
Adaptive Teaching with Shared Classifier for Knowledge DistillationKnowledge distillation → General knowledge distillation methods
Optimized Feature Generation for Tabular Data via LLMs with Decision Tree ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning
Predicting Cascading Failures with a Hyperparametric Diffusion ModelDiffusion models → Flow matching, time series and science
Beyond LLaVA-HD: Diving into High-Resolution Large Multimodal ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency
On Evaluating Adversarial Robustness of Volumetric Medical Segmentation ModelsState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Words Worth a Thousand Pictures: Measuring and Understanding Perceptual Variability in Text-to-Image GenerationDiffusion models → Image generation, text-to-image and editing
Enhancing End-to-End Autonomous Driving with Latent World ModelWorld models → World models for autonomous driving +1
Human-3Diffusion: Realistic Avatar Creation via Explicit 3D Consistent Diffusion ModelsDiffusion models → 3D, point cloud and scene generation
RILe: Reinforced Imitation LearningOffline RL and imitation learning from data → Imitation learning methods +1
PAL: Pluralistic Alignment Framework for Learning from Heterogeneous PreferencesLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with NothingLLM reasoning → Reinforcement learning for reasoning +2
Adaptive Swarm Mesh Refinement using Deep Reinforcement Learning with Local RewardsDeep reinforcement learning methods → RL theory and sample complexity
Diffusion Soup: Model Merging for Text-to-Image Diffusion ModelsDiffusion models → Image generation, text-to-image and editing +1
Discovering Preference Optimization Algorithms with and for Large Language ModelsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
MMWorld: Towards Multi-discipline Multi-faceted World Model Evaluation in VideosWorld models → World models for language agents
RRLS : Robust Reinforcement Learning SuiteDeep reinforcement learning methods → RL theory and sample complexity
Time-Constrained Robust MDPsDeep reinforcement learning methods → RL theory and sample complexity
VisionLLM v2: An End-to-End Generalist Multimodal Large Language Model for Hundreds of Vision-Language TasksMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
FontStudio: Shape-Adaptive Diffusion Model for Coherent and Consistent Font Effect GenerationDiffusion models → Image generation, text-to-image and editing
Diff-A-Riff: Musical Accompaniment Co-creation via Latent Diffusion ModelsDiffusion models → Audio, speech and music
2.5D Multi-view Averaging Diffusion Model for 3D Medical Image Translation: Application to Low-count PET Reconstruction with CT-less Attenuation CorrectionDiffusion models → Medical imaging and segmentation
Utilizing Navigation Paths to Generate Target Points for Enhanced End-to-End Autonomous Driving PlanningEnd-to-end autonomous driving → End-to-end system architecture
WMAdapter: Adding WaterMark Control to Latent Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects +1
UDON: Universal Dynamic Online distillatioN for generic image representationsKnowledge distillation → General knowledge distillation methods
Decoupling the Class Label and the Target Concept in Machine UnlearningLLM safety and security → Machine unlearning
Dataset Enhancement with Instance-Level AugmentationsDiffusion models → Image generation, text-to-image and editing
Leveraging Large Language Models for Web ScrapingRetrieval-augmented generation → RAG pipeline design and evaluation
Residual Learning and Context Encoding for Adaptive Offline-to-Online Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
MaIL: Improving Imitation Learning with MambaState-space sequence models (S4, Mamba) → Core SSM architectures for language +1
DistilDoc: Knowledge Distillation for Visually-Rich Document ApplicationsKnowledge distillation → Multimodal distillation
Diffusion-Promoted HDR Video ReconstructionDiffusion models → Inverse problems and image restoration
LAFMA: A Latent Flow Matching Model for Text-to-Audio GenerationDiffusion models → Audio, speech and music
MobileAgentBench: An Efficient and User-Friendly Benchmark for Mobile LLM AgentsLLM agents and tool use → GUI and mobile agents
One-Step Effective Diffusion Network for Real-World Image Super-ResolutionDiffusion models → Inverse problems and image restoration
ConMe: Rethinking Evaluation of Compositional Reasoning for Modern VLMsLLM reasoning → Multimodal reasoning
QuantMoE-Bench: Examining Post-Training Quantization for Mixture-of-ExpertsMixture-of-experts → MoE large language models
Probing Implicit Bias in Semi-gradient Q-learning: Visualizing the Effective Loss Landscapes via the Fokker--Planck EquationDeep reinforcement learning methods → RL theory and sample complexity
Legend: Leveraging Representation Engineering to Annotate Safety Margin for Preference DatasetsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Supportiveness-based Knowledge Rewriting for Retrieval-augmented Language ModelingRetrieval-augmented generation → RAG pipeline design and evaluation
A Concept-Based Explainability Framework for Large Multimodal ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
CFG++: Manifold-constrained Classifier Free Guidance for Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Explore-Go: Leveraging Exploration for Generalisation in Deep Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
LVBench: An Extreme Long Video Understanding BenchmarkMultimodal LLMs / vision-language models → Video understanding with MLLMs
Fewer Tokens and Fewer Videos: Extending Video Understanding Abilities in Large Vision-Language ModelsMultimodal LLMs / vision-language models → Video understanding with MLLMs
OpenObj: Open-Vocabulary Object-Level Neural Radiance Fields with Fine-Grained UnderstandingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Efficient Adaptation in Mixed-Motive Environments via Hierarchical Opponent Modeling and PlanningDeep reinforcement learning methods → RL for games and combinatorial optimization
How social reinforcement learning can lead to metastable polarisation and the voter modelDeep reinforcement learning methods → RL training of language and agent models
Reinforcement Learning for High-Level Strategic Control in Tower Defense GamesDeep reinforcement learning methods → RL for games and combinatorial optimization
It Takes Two: On the Seamlessness between Reward and Policy Model in RLHFLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Large Language Model Unlearning via Embedding-Corrupted PromptsLLM safety and security → Machine unlearning
Graph Transductive Defense: a Two-Stage Defense for Graph Membership Inference AttacksLLM safety and security → Membership inference attacks
Guiding Frame-Level CTC Alignments Using Self-knowledge DistillationKnowledge distillation → Speech and audio distillation
Ablation Based CounterfactualsDiffusion models → Diffusion foundations and sampling
Grounding Multimodal Large Language Models in ActionsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
When Do Skills Help Reinforcement Learning? A Theoretical Analysis of Temporal AbstractionsDeep reinforcement learning methods → Deep RL algorithms and representations
A Finite-Sample Analysis of an Actor-Critic Algorithm for Mean-Variance Optimization in a Discounted MDPDeep reinforcement learning methods → RL theory and sample complexity
An Empirical Study of Mamba-based Language ModelsState-space sequence models (S4, Mamba) → Core SSM architectures for language +1
GENIU: A Restricted Data Access Unlearning for Imbalanced DataLLM safety and security → Machine unlearning
Reinforcement Learning to Disentangle Multiqubit Quantum States from Partial ObservationsDeep reinforcement learning methods → Quantum reinforcement learning
Hierarchical Reinforcement Learning for Swarm Confrontation with High UncertaintyDeep reinforcement learning methods → RL for robot navigation and UAV control
Small Scale Data-Free Knowledge DistillationKnowledge distillation → General knowledge distillation methods
Self-Distillation Learning Based on Temporal-Spatial Consistency for Spiking Neural NetworksKnowledge distillation → Distillation for spiking neural networks
Toward Enhanced Reinforcement Learning-Based Resource Management via Digital Twin: Opportunities, Applications, and ChallengesDeep reinforcement learning methods → RL for energy, networks and operations
DiffPop: Plausibility-Guided Object Placement Diffusion for Image CompositionDiffusion models → Image generation, text-to-image and editing
Improving Compositional Attribute Binding in Text-to-Image Generative Models via Enhanced Text EmbeddingsDiffusion models → Image generation, text-to-image and editing
The Max-Min Formulation of Multi-Objective Reinforcement Learning: From Theory to a Model-Free AlgorithmDeep reinforcement learning methods → RL for games and combinatorial optimization
Collective Constitutional AI: Aligning a Language Model with Public InputLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Battling Botpoop using GenAI for Higher Education: A Study of a Retrieval Augmented Generation Chatbots Impact on LearningRetrieval-augmented generation → Agentic RAG, agents and memory
Hierarchical Patch Diffusion Models for High-Resolution Video GenerationDiffusion models → Video generation
A Critical Look At Tokenwise Reward-Guided Text GenerationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
A Study of Backdoors in Instruction Fine-tuned Language ModelsLLM safety and security → Backdoor and data poisoning attacks
OT-VP: Optimal Transport-guided Visual Prompting for Test-Time Adaptation
YOLO advances to its genesis: a decadal and comprehensive review of the You Only Look Once (YOLO) series
A Multi-Resolution Mutual Learning Network for Multi-Label ECG Classification
Enhancing Wearable based Real-Time Glucose Monitoring via Phasic Image Representation Learning based Deep Learning
Analyzing Multi-Head Attention on Trojan BERT Models
Scalable Training of Trustworthy and Energy-Efficient Predictive Graph Foundation Models for Atomistic Materials Modeling: A Case Study with HydraGNN
Rating Multi-Modal Time-Series Forecasting Models (MM-TSFM) for Robustness Through a Causal Lens
Reconciling Kaplan and Chinchilla Scaling Laws
Entropy-statistical approach to phase-locking detection of pulse oscillations: application for the analysis of biosignal synchronization
Data Petri Nets meet Probabilistic Programming (Extended version)
Applications of Explainable artificial intelligence in Earth system science

All 366