Papers

New on arXiv

Tue 17 Dec 2024

389 papers

Entire-Space Variational Information Exploitation for Post-Click Conversion Rate PredictionKnowledge distillation → Distillation for recommender systems
Split Knowledge Distillation for Large Models in IoT: Architecture, Challenges, and SolutionsKnowledge distillation → Federated, robust and private distillation
Coordinated Power Smoothing Control for Wind Storage Integrated System with Physics-informed Deep Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
Evaluating the Capabilities of Large Language Models for Multi-label Emotion UnderstandingLLM reasoning → Audio-language model reasoning
CLIP-RLDrive: Human-Aligned Autonomous Driving via CLIP-Based Reward Shaping in Reinforcement LearningEnd-to-end autonomous driving → Reasoning VLA and VLM driving policies +2
Fooling LLM graders into giving better grades through neural activity guided adversarial promptingLLM safety and security → Prompt injection and agent security
Memory-Augmented Agent Training for Business Document UnderstandingLLM agents and tool use → Agent memory systems
SimGRAG: Leveraging Similar Subgraphs for Knowledge Graphs Driven Retrieval-Augmented GenerationRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
A MapReduce Approach to Effectively Utilize Long Context Information in Retrieval Augmented Language ModelsRetrieval-augmented generation → Medical and biomedical RAG
On the Structural Memory of LLM AgentsLLM agents and tool use → Agent memory systems
Integrating Evidence into the Design of XAI and AI-based Decision Support Systems: A Means-End Framework for End-users in ConstructionLLM reasoning → Reasoning for agents, planning and code
Beyond Accuracy: On the Effects of Fine-tuning Towards Vision-Language Model's Prediction RationalityMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
FastVLM: Efficient Vision Encoding for Vision Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Refining Answer Distributions for Improved Large Language Model ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Posterior Mean Matching: Generative Modeling through Online Bayesian InferenceDiffusion models → Diffusion foundations and sampling
In-Context Learning Distillation for Efficient Few-Shot Fine-TuningKnowledge distillation → Knowledge distillation for LLMs
SafeDrive: Knowledge- and Data-Driven Risk-Sensitive Decision-Making for Autonomous Vehicles with Large Language ModelsEnd-to-end autonomous driving → Adversarial scenarios and safety testing
COSEE: Consistency-Oriented Signal-Based Early Exiting via Calibrated Sample Weighting MechanismEfficient LLM inference and serving → Early exit and edge inference
Creating an LLM-based AI-agent: A high-level methodology towards enhancing LLMs with APIsLLM agents and tool use → General LLM agent frameworks and benchmarks
Physics-model-guided Worst-case Sampling for Safe Reinforcement LearningDeep reinforcement learning methods → Safe and constrained RL
CoMPaSS: Enhancing Spatial Understanding in Text-to-Image Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Proposer-Agent-Evaluator(PAE): Autonomous Skill Discovery For Foundation Model Internet AgentsLLM agents and tool use → Web, search, and research agents
StreetCrafter: Street View Synthesis with Controllable Video Diffusion ModelsWorld models → World models for autonomous driving +1
HandsOnVLM: Vision-Language Models for Hand-Object Interaction PredictionMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Tilted Quantile Gradient Updates for Quantile-Constrained Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model AccelerationMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
Compressed Chain of Thought: Efficient Reasoning Through Dense RepresentationsLLM reasoning → Test-time compute and efficient thinking
C-FedRAG: A Confidential Federated Retrieval-Augmented Generation SystemRetrieval-augmented generation → Privacy and security of RAG
Are Your LLMs Capable of Stable Reasoning?LLM reasoning → Reasoning benchmarks and evaluation
Knowledge-enhanced Pretraining for Vision-language Pathology Foundation Model on Cancer DiagnosisMultimodal LLMs / vision-language models → Medical vision-language models
Active Reinforcement Learning Strategies for Offline Policy ImprovementDeep reinforcement learning methods → Deep RL algorithms and representations
Reservoir Computing for Fast, Simplified Reinforcement Learning on Memory TasksDeep reinforcement learning methods → Deep RL algorithms and representations
3D MedDiffusion: A 3D Medical Latent Diffusion Model for Controllable and High-quality Medical Image GenerationDiffusion models → Medical imaging and segmentation
SMOSE: Sparse Mixture of Shallow Experts for Interpretable Reinforcement Learning in Continuous Control TasksDeep reinforcement learning methods → Deep RL algorithms and representations
Modality-Inconsistent Continual Learning of Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
TAME: Temporal Audio-based Mamba for Enhanced Drone Trajectory Estimation and ClassificationState-space sequence models (S4, Mamba) → Speech and audio SSMs
Relational Neurosymbolic Markov ModelsDeep reinforcement learning methods → Interpretable and logic-guided RL
OmniEval: An Omnidirectional and Automatic RAG Evaluation Benchmark in Financial DomainRetrieval-augmented generation → Legal and financial RAG
Stable Diffusion is a Natural Cross-Modal Decoder for Layered AI-generated Image CompressionDiffusion models → Diffusion for compression and wireless
Attentive Eraser: Unleashing Diffusion Model's Object Removal Potential via Self-Attention Redirection GuidanceDiffusion models → 3D, point cloud and scene generation
ArchesWeather & ArchesWeatherGen: a deterministic and generative model for efficient ML weather forecastingDiffusion models → Flow matching, time series and science
Efficient Diffusion Transformer Policies with Mixture of Expert Denoisers for Multitask LearningDiffusion models → Diffusion and flow policies for robotics +2
Improving Fine-grained Visual Understanding in VLMs through Text-Only TrainingMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
CoMT: A Novel Benchmark for Chain of Multi-modal Thought on Large Vision-Language ModelsLLM reasoning → Multimodal reasoning +1
Unsupervised Region-Based Image Editing of Denoising Diffusion ModelsDiffusion models → Diffusion foundations and sampling
An Agentic Approach to Automatic Creation of P&ID Diagrams from Natural Language DescriptionsLLM agents and tool use → Multimodal and embodied agents
ArtAug: Enhancing Text-to-Image Generation through Synthesis-Understanding InteractionDiffusion models → Image generation, text-to-image and editing
RAG-Star: Enhancing Deliberative Reasoning with Retrieval Augmented Verification and RefinementLLM reasoning → Knowledge- and tool-augmented reasoning +1
Physically Interpretable World Models via Weakly Supervised Representation LearningWorld models → Latent predictive world models
Preference-Oriented Supervised Fine-Tuning: Favoring Target Model Over Aligned Large Language ModelsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Efficient Speech Command Recognition Leveraging Spiking Neural Network and Curriculum Learning-based Knowledge DistillationKnowledge distillation → Distillation for spiking neural networks
Benchmarking and Understanding Compositional Relational Reasoning of LLMsLLM reasoning → Reasoning benchmarks and evaluation +1
Exposing the Vulnerability of Decentralized Learning to Membership Inference Attacks Through the Lens of Graph MixingLLM safety and security → Membership inference attacks
FocusChat: Text-guided Long Video Understanding via Spatiotemporal Information FilteringMultimodal LLMs / vision-language models → Video understanding with MLLMs
ComprehendEdit: A Comprehensive Dataset and Evaluation Framework for Multimodal Knowledge EditingMultimodal LLMs / vision-language models → Image editing and quality assessment
Detecting Emotional Incongruity of Sarcasm by Commonsense ReasoningLLM reasoning → Multimodal reasoning
RA-SGG: Retrieval-Augmented Scene Graph Generation Framework via Multi-Prototype LearningRetrieval-augmented generation → Multimodal RAG
Activating Distributed Visual Region within LLMs for Efficient and Effective Vision-Language Training and InferenceMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Rethinking Diffusion-Based Image Generators for Fundus Fluorescein Angiography Synthesis on Limited DataDiffusion models → Medical imaging and segmentation
RemoteRAG: A Privacy-Preserving LLM Cloud RAG ServiceRetrieval-augmented generation → Privacy and security of RAG
Guided and Variance-Corrected Fusion with One-shot Style Alignment for Large-Content Image GenerationDiffusion models → Image generation, text-to-image and editing
Towards a Training Free Approach for 3D Scene EditingDiffusion models → 3D, point cloud and scene generation
CAMEL: Cross-Attention Enhanced Mixture-of-Experts and Language Bias for Code-Switching Speech RecognitionMixture-of-experts → Multimodal MoE models
GIRAFFE: Design Choices for Extending the Context Length of Visual Language ModelsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Defending LVLMs Against Vision Attacks through Partial-Perception SupervisionMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
More Tokens, Lower Precision: Towards the Optimal Token-Precision Trade-off in KV Cache CompressionEfficient LLM inference and serving → KV cache optimization
ParMod: A Parallel and Modular Framework for Learning Non-Markovian TasksDeep reinforcement learning methods → Interpretable and logic-guided RL
SPHERE: Unveiling Spatial Blind Spots in Vision-Language Models Through Hierarchical EvaluationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Neural-Network-Driven Reward Prediction as a Heuristic: Advancing Q-Learning for Mobile Robot Path PlanningDeep reinforcement learning methods → Deep RL algorithms and representations
LLM-based Discriminative Reasoning for Knowledge Graph Question AnsweringLLM reasoning → Knowledge- and tool-augmented reasoning
RDPI: A Refine Diffusion Probability Generation Method for Spatiotemporal Data ImputationDiffusion models → Flow matching, time series and science
Lagrangian Index Policy for Restless Bandits with Average RewardDeep reinforcement learning methods → RL theory and sample complexity
Falcon: Faster and Parallel Inference of Large Language Models through Enhanced Semi-Autoregressive Drafting and Custom-Designed Decoding TreeEfficient LLM inference and serving → Speculative decoding
Make Imagination Clearer! Stable Diffusion-based Visual Imagination for Multimodal Machine TranslationDiffusion models → Image generation, text-to-image and editing
Neighbor-Aware Reinforcement Learning for Mixed Traffic Optimization in Large-scale NetworksDeep reinforcement learning methods → RL for autonomous driving and traffic
Jailbreaking? One Step Is Enough!LLM safety and security → Jailbreaking
MultiLingPoT: Enhancing Mathematical Reasoning with Multilingual Program Fine-tuningLLM reasoning → Mathematical reasoning
Multi-Dimensional Insights: Benchmarking Real-World Personalization in Large Multimodal ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
An Advantage-based Optimization Method for Reinforcement Learning in Large Action SpaceDeep reinforcement learning methods → Deep RL algorithms and representations
Distribution-Free Uncertainty Quantification in Mechanical Ventilation Treatment: A Conformal Deep Q-Learning FrameworkDeep reinforcement learning methods → RL for clinical treatment and surgery
A Simple and Efficient Baseline for Zero-Shot Generative ClassificationDiffusion models → Image generation, text-to-image and editing
Process-Supervised Reward Models for Verifying Clinical Note Generation: A Scalable Approach Guided by Domain ExpertiseLLM alignment / RLHF / preference optimisation → Medical LLM alignment
ChatDiT: A Training-Free Baseline for Task-Agnostic Free-Form Chatting with Diffusion TransformersDiffusion models → Image generation, text-to-image and editing
FCMR: Robust Evaluation of Financial Cross-Modal Multi-Hop ReasoningLLM reasoning → Reasoning for agents, planning and code
Task-Agnostic Language Model Watermarking via High Entropy Passthrough LayersLLM safety and security → Watermarking generated content and models
EXIT: Context-Aware Extractive Compression for Enhancing Retrieval-Augmented GenerationRetrieval-augmented generation → Generation and context integration +1
Consistent Diffusion: Denoising Diffusion Model with Data-Consistent Training for Image RestorationDiffusion models → Inverse problems and image restoration
Seed-CTS: Unleashing the Power of Tree Search for Superior Performance in Competitive Coding TasksLLM reasoning → Reasoning for agents, planning and code
Stiefel Flow Matching for Moment-Constrained Structure ElucidationDiffusion models → Flow matching, time series and science
Addressing Small and Imbalanced Medical Image Datasets Using Generative Models: A Comparative Study of DDPM and PGGANs with Random and Greedy K SamplingDiffusion models → Medical imaging and segmentation
Can Large Language Models Understand You Better? An MBTI Personality Detection Dataset Aligned with Population TraitsLLM reasoning → Reasoning benchmarks and evaluation
LinguaLIFT: An Effective Two-stage Instruction Tuning Framework for Low-Resource Language ReasoningLLM reasoning → Reasoning benchmarks and evaluation
NLSR: Neuron-Level Safety Realignment of Large Language Models Against Harmful Fine-TuningLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Faster Vision Mamba is Rebuilt in Minutes via Merged Token Re-trainingState-space sequence models (S4, Mamba) → Vision Mamba backbones +1
A Simple and Fast Way to Handle Semantic Errors in TransactionsLLM agents and tool use → Web, search, and research agents
DuSSS: Dual Semantic Similarity-Supervised Vision-Language Model for Semi-Supervised Medical Image SegmentationMultimodal LLMs / vision-language models → Medical vision-language models
A System for Microserving of LLMsEfficient LLM inference and serving → Serving systems and request scheduling
LLM is Knowledge Graph Reasoner: LLM's Intuition-aware Knowledge Graph Reasoning for Cold-start Sequential RecommendationLLM reasoning → LLM reasoning for recommendation
PERC: Plan-As-Query Example Retrieval for Underrepresented Code GenerationRetrieval-augmented generation → RAG for code

All 389