Papers

New on arXiv

Thu 24 Oct 2024

339 papers

Implementing Deep Reinforcement Learning-Based Grid Voltage Control in Real-World Power Systems: Challenges and InsightsDeep reinforcement learning methods → RL for energy, networks and operations
Inference time LLM alignment in single and multidomain preference spectrumLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
MAP: Multi-Human-Value Alignment PaletteLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Reinforcement Learning the Chromatic Symmetric FunctionDeep reinforcement learning methods → RL for games and combinatorial optimization
MMAU: A Massive Multi-Task Audio Understanding and Reasoning BenchmarkLLM reasoning → Audio-language model reasoning
Structured Diffusion Models with Mixture of Gaussians as Prior DistributionDiffusion models → Diffusion foundations and sampling
Visual Text Matters: Improving Text-KVQA with Visual Text Entity Knowledge-aware Large Multimodal AssistantMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Versatile Demonstration Interface: Toward More Flexible Robot Demonstration CollectionOffline RL and imitation learning from data → Robot learning from demonstration
AlignCap: Aligning Speech Emotion Captioning to Human PreferencesLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
Hybrid Preferences: Learning to Route Instances for Human vs. AI FeedbackLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Read-ME: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-DesignMixture-of-experts → Efficient MoE training and inference systems
Bio2Token: All-atom tokenization of any biomolecular structure with MambaState-space sequence models (S4, Mamba) → Core SSM architectures for language
Watermarking Large Language Models and the Generated Content: Opportunities and ChallengesLLM safety and security → Watermarking generated content and models
GCoder: Improving Large Language Model for Generalized Graph Problem SolvingLLM reasoning → Knowledge- and tool-augmented reasoning
BIFR"OST: 3D-Aware Image compositing with Language InstructionsDiffusion models → 3D, point cloud and scene generation
The Stepwise Deception: Simulating the Evolution from True News to Fake News with LLM AgentsLLM agents and tool use → Social simulation and multi-agent debate
ReasonAgain: Using Extractable Symbolic Programs to Evaluate Mathematical ReasoningLLM reasoning → Mathematical reasoning
Infogent: An Agent-Based Framework for Web Information AggregationLLM agents and tool use → Web, search, and research agents
Mixture of Parrots: Experts improve memorization more than reasoningMixture-of-experts → MoE large language models
Pay Attention and Move Better: Harnessing Attention for Interactive Motion Generation and Training-free EditingDiffusion models → Video generation
CAMEL-Bench: A Comprehensive Arabic LMM BenchmarkLLM reasoning → Multimodal reasoning +1
Unbounded: A Generative Infinite Game of Character Life SimulationWorld models → Video-generation world models and simulators
3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D GenerationDiffusion models → 3D, point cloud and scene generation
OSCAR: Operating System Control via State-Aware Reasoning and Re-PlanningLLM reasoning → Multimodal reasoning
Stable Consistency Tuning: Understanding and Improving Consistency ModelsDiffusion models → Diffusion foundations and sampling
Dynamic Vocabulary Pruning in Early-Exit LLMsEfficient LLM inference and serving → Early exit and edge inference
SafeBench: A Safety Evaluation Framework for Multimodal Large Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
SegLLM: Multi-round Reasoning SegmentationLLM reasoning → Multimodal reasoning
From Blind Solvers to Logical Thinkers: Benchmarking LLMs' Logical Integrity on Faulty Mathematical ProblemsLLM reasoning → Mathematical reasoning
SkillMimicGen: Automated Demonstration Generation for Efficient Skill Learning and DeploymentOffline RL and imitation learning from data → Robot manipulation imitation learning
Improving Small-Scale Large Language Models Function Calling for Reasoning TasksLLM reasoning → Reasoning benchmarks and evaluation +2
Diff-Instruct++: Training One-step Text-to-image Generator Model to Align with Human PreferencesDiffusion models → Diffusion and flow policies for robotics +1
The Cat and Mouse Game: The Ongoing Arms Race Between Diffusion Models and Detection MethodsDiffusion models → Image generation, text-to-image and editing
Provably Robust Watermarks for Open-Source Language ModelsLLM safety and security → Watermarking generated content and models
Probabilistic Language-Image Pre-TrainingMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Entry-level guide to the use of large language models for medical researchLLM reasoning → Clinical and biomedical applications
High-dimensional Analysis of Knowledge Distillation: Weak-to-Strong Generalization and Scaling LawsKnowledge distillation → General knowledge distillation methods
Multi-Scale Diffusion: Enhancing Spatial Layout in High-Resolution Panoramic Image GenerationDiffusion models → 3D, point cloud and scene generation
PSY: Posterior Sampling Based Privacy Enhancer in Large Language ModelsLLM safety and security → Membership inference attacks
Fast constrained sampling in pre-trained diffusion modelsDiffusion models → Diffusion foundations and sampling
PointPatchRL -- Masked Reconstruction Improves Reinforcement Learning on Point CloudsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Distill Visual Chart Reasoning Ability from LLMs to MLLMsLLM reasoning → Multimodal reasoning +1
An LLM Agent for Automatic Geospatial Data AnalysisLLM agents and tool use → Multimodal and embodied agents
Denoising diffusion probabilistic models are optimally adaptive to unknown low dimensionalityDiffusion models → Diffusion foundations and sampling
Robust Watermarking Using Generative Priors Against Image Editing: From Benchmarking to AdvancesLLM safety and security → Watermarking generated content and models
Schedule Your Edit: A Simple yet Effective Diffusion Noise Schedule for Image EditingDiffusion models → Image generation, text-to-image and editing
Rectified Diffusion Guidance for Conditional GenerationDiffusion models → Diffusion foundations and sampling
Retrieval-Augmented Diffusion Models for Time Series ForecastingRetrieval-augmented generation → Time series and agricultural RAG +1
Unleashing LLM Reasoning Capability via Scalable Question Synthesis from ScratchLLM reasoning → Mathematical reasoning
Hierarchical Multimodal LLMs with Semantic Space Alignment for Enhanced Time Series ClassificationMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Ali-AUG: Innovative Approaches to Labeled Data Augmentation using One-Step Diffusion ModelDiffusion models → Image generation, text-to-image and editing
DreamClear: High-Capacity Real-World Image Restoration with Privacy-Safe Dataset CurationDiffusion models → Inverse problems and image restoration
Data Scaling Laws in Imitation Learning for Robotic ManipulationOffline RL and imitation learning from data → Robot manipulation imitation learning
Weak-to-Strong Preference Optimization: Stealing Reward from Weak Aligned ModelLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Diffusion Attribution Score: Evaluating Training Data Influence in Diffusion ModelsDiffusion models → Diffusion foundations and sampling
SAMG: Offline-to-Online Reinforcement Learning via State-Action-Conditional Offline Model GuidanceOffline RL and imitation learning from data → Offline RL algorithms
Learning Transparent Reward Models via Unsupervised Feature SelectionOffline RL and imitation learning from data → Inverse reinforcement learning
Knowledge Distillation Using Frontier Open-source LLMs: Generalizability and the Role of Synthetic DataSynthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping +1
Aligning CodeLLMs with Direct Preference OptimizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
SIKeD: Self-guided Iterative Knowledge Distillation for mathematical reasoningLLM reasoning → Prompting and chain-of-thought reasoning +1
Taipan: Efficient and Expressive State Space Language Models with Selective AttentionState-space sequence models (S4, Mamba) → Core SSM architectures for language
Zero-shot Object Navigation with Vision-Language Models ReasoningLLM reasoning → Spatial and embodied reasoning +1
Infinity-MM: Scaling Multimodal Performance with Large-Scale and High-Quality Instruction DataMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Beyond Color and Lines: Zero-Shot Style-Specific Image Variations with Coordinated SemanticsDiffusion models → Image generation, text-to-image and editing
LOGO -- Long cOntext aliGnment via efficient preference OptimizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Instructional Text Across Disciplines: A Survey of Representations, Downstream Tasks, and Open Challenges Toward Capable AI AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
PRACT: Optimizing Principled Reasoning and Acting of LLM AgentLLM reasoning → Reasoning for agents, planning and code +1
Reinforcement Learning Controllers for Soft Robots using Learned EnvironmentsDeep reinforcement learning methods → RL for robot locomotion and manipulation
KVSharer: Efficient Inference via Layer-Wise Dissimilar KV Cache SharingLLM reasoning → Test-time compute and efficient thinking +1
Scaling up Masked Diffusion Models on TextDiffusion models → Discrete and text diffusion
Multi-UAV Formation Control with Static and Dynamic Obstacle Avoidance via Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Iterative Self-Tuning LLMs for Enhanced Jailbreaking CapabilitiesLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Learn 2 Rage: Experiencing The Emotional Roller Coaster That Is Reinforcement LearningDeep reinforcement learning methods → RL for autonomous driving and traffic
Skywork-Reward: Bag of Tricks for Reward Modeling in LLMsLLM reasoning → Reinforcement learning for reasoning +1
ToolFlow: Boosting LLM Tool-Calling Through Natural and Coherent Dialogue SynthesisLLM agents and tool use → Tool learning and function calling
Decoding on Graphs: Faithful and Sound Reasoning on Knowledge Graphs through Generation of Well-Formed ChainsLLM reasoning → Knowledge- and tool-augmented reasoning
FreCaS: Efficient Higher-Resolution Image Generation via Frequency-aware Cascaded SamplingDiffusion models → Diffusion foundations and sampling
MoMQ: Mixture-of-Experts Enhances Multi-Dialect Query Generation across Relational and Non-Relational DatabasesMixture-of-experts → MoE large language models
Interpretable Bilingual Multimodal Large Language Model for Diverse Biomedical TasksMultimodal LLMs / vision-language models → Medical vision-language models
Gibberish is All You Need for Membership Inference Detection in Contrastive Language-Audio PretrainingLLM safety and security → Membership inference attacks
AdaEDL: Early Draft Stopping for Speculative Decoding of Large Language Models via an Entropy-based Lower Bound on Token Acceptance ProbabilityEfficient LLM inference and serving → Speculative decoding
Aggregated Knowledge Model: Enhancing Domain-Specific QA with Fine-Tuned and Retrieval-Augmented Generation ModelsRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
A Comprehensive Survey of Deep Learning for Time Series Forecasting: Architectural Diversity and Open Challenges
VECTOR: Velocity-Enhanced GRU Neural Network for Real-Time 3D UAV Trajectory Prediction
A Survey of AI-Generated Video Evaluation
Critical biblical studies via word frequency analysis: unveiling text authorship
Recommendations for Comprehensive and Independent Evaluation of Machine Learning-Based Earth System Models
Parameter-Efficient Fine-Tuning in Large Models: A Survey of Methodologies
Predicting potato plant vigor from the seed tuber properties
Paved or unpaved? A Deep Learning derived Road Surface Global Dataset from Mapillary Street-View Imagery
Radar and Camera Fusion for Object Detection and Tracking: A Comprehensive Survey
Causal Order Discovery based on Monotonic SCMs
Comparing YOLOv11 and YOLOv8 for instance segmentation of occluded and non-occluded immature green fruits in complex orchard environment
No Free Lunch: Fundamental Limits of Learning Non-Hallucinating Generative Models
Prototypical Hash Encoding for On-the-Fly Fine-Grained Category Discovery
Predicting Liquidity Coverage Ratio with Gated Recurrent Units: A Deep Learning Model for Risk Management
Equitable Federated Learning with Activation Clustering
An Inverse Modeling Constrained Multi-Objective Evolutionary Algorithm Based on Decomposition
Binary Classification: Is Boosting stronger than Bagging?
Making Social Platforms Accessible: Emotion-Aware Speech Generation with Integrated Text Analysis

All 339