Papers

New on arXiv

Thu 19 Dec 2024

365 papers

Beyond the Sum: Unlocking AI Agents Potential Through Market ForcesLLM agents and tool use → General LLM agent frameworks and benchmarks
TalkWithMachines: Enhancing Human-Robot Interaction for Interpretable Industrial Robotics Through Large/Vision Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Asymptotic Extinction in Large Coordination GamesDeep reinforcement learning methods → RL for games and combinatorial optimization
Northeastern Uni at Multilingual Counterspeech Generation: Enhancing Counter Speech Generation with LLM Alignment through Direct Preference OptimizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
SKETCH: Structured Knowledge Enhanced Text Comprehension for Holistic RetrievalRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Offline Safe Reinforcement Learning Using Trajectory ClassificationOffline RL and imitation learning from data → Safe offline RL
AdaCred: Adaptive Causal Decision Transformers with Feature CreditingOffline RL and imitation learning from data → Decision transformers and sequence modeling +1
Sparse Checkpointing for Fast and Reliable MoE TrainingMixture-of-experts → Efficient MoE training and inference systems
A Retrieval-Augmented Generation Framework for Academic Literature Navigation in Data ScienceRetrieval-augmented generation → RAG pipeline design and evaluation
Automatic Extraction of Metaphoric Analogies from Literary Texts: Task Formulation, Dataset Construction, and EvaluationLLM reasoning → Reasoning benchmarks and evaluation
A Generative Framework for Probabilistic, Spatiotemporally Coherent Downscaling of Climate SimulationDiffusion models → Flow matching, time series and science
Dataset Augmentation by Mixing Visual ConceptsDiffusion models → Image generation, text-to-image and editing
Efficient Fine-Tuning and Concept Suppression for Pruned Diffusion ModelsDiffusion models → Diffusion foundations and sampling +1
Eliciting Causal Abilities in Large Language Models for Reasoning TasksLLM reasoning → Causal reasoning in LLMs
Tree-of-Code: A Tree-Structured Exploring Framework for End-to-End Code Generation and Execution in Complex Task HandlingLLM reasoning → Reasoning for agents, planning and code
Self-Evolution Knowledge Distillation for LLM-based Machine TranslationKnowledge distillation → Knowledge distillation for LLMs
A Comparative Study of DSPy Teleprompter Algorithms for Aligning Large Language Models Evaluation Metrics to Human EvaluationLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
Confidence in the Reasoning of Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Deep reinforcement learning with time-scale invariant memoryDeep reinforcement learning methods → Deep RL algorithms and representations
SATA: A Paradigm for LLM Jailbreak via Simple Assistive Task LinkageLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Flowing from Words to Pixels: A Noise-Free Framework for Cross-Modality EvolutionDiffusion models → Image generation, text-to-image and editing
PRIMA: Multi-Image Vision-Language Models for Reasoning SegmentationLLM reasoning → Multimodal reasoning +1
OpenEMMA: Open-Source Multimodal Model for End-to-End Autonomous DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
AutoTrust: Benchmarking Trustworthiness in Large Vision Language Models for Autonomous DrivingLLM reasoning → Chain-of-thought reasoning for driving +2
FlowAR: Scale-wise Autoregressive Image Generation Meets Flow MatchingDiffusion models → Image generation, text-to-image and editing
LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context MultitasksLLM reasoning → Reasoning benchmarks and evaluation
AV-Link: Temporally-Aligned Diffusion Features for Cross-Modal Audio-Video GenerationDiffusion models → Audio, speech and music
EarthDial: Turning Multi-sensory Earth Observations to Interactive DialoguesMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
STRAP: Robot Sub-Trajectory Retrieval for Augmented Policy LearningVision-language-action models → Reinforcement learning and post-training
Critical-Questions-of-Thought: Steering LLM reasoning with Argumentative QueryingLLM reasoning → Reasoning benchmarks and evaluation
Rethinking Uncertainty Estimation in LLMs: A Principled Single-Sequence MeasureLLM reasoning → Test-time compute and efficient thinking
OnlineVPO: Align Video Diffusion Model with Online Video-Centric Preference OptimizationDiffusion models → Video generation +1
Prompt-A-Video: Prompt Your Video Diffusion Model via Preference-Aligned LLMDiffusion models → Video generation +1
Language Models as Continuous Self-Evolving Data EngineersSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Reasoning Through Execution: Unifying Process and Outcome Rewards for Code GenerationLLM reasoning → Reasoning for agents, planning and code
Qwen2.5 Technical ReportMixture-of-experts → MoE large language models
Predictive Inverse Dynamics Models are Scalable Learners for Robotic ManipulationWorld models → Embodied and robot world models
AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward ModelingLLM reasoning → Mathematical reasoning +1
DCTdiff: Intriguing Properties of Image Generative Modeling in the DCT SpaceDiffusion models → Diffusion foundations and sampling
Efficient Knowledge Injection in LLMs via Self-DistillationKnowledge distillation → Knowledge distillation for LLMs
Understanding the Dark Side of LLMs' Intrinsic Self-CorrectionLLM reasoning → Test-time compute and efficient thinking
Dream to Manipulate: Compositional World Models Empowering Robot Imitation Learning with ImaginationWorld models → Embodied and robot world models +1
Dehallucinating Parallel Context Extension for Retrieval-Augmented GenerationRetrieval-augmented generation → Hallucination detection in RAG
MagicNaming: Consistent Identity Generation by Finding a "Name Space" in T2I Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Diffusion priors for Bayesian 3D reconstruction from incomplete measurementsDiffusion models → Inverse problems and image restoration
Hierarchical Subspaces of Policies for Continual Offline Reinforcement LearningOffline RL and imitation learning from data → Decision transformers and sequence modeling
Think&Cite: Improving Attributed Text Generation with Self-Guided Tree Search and Progress Reward ModelingLLM reasoning → Test-time compute and efficient thinking +1
DS2-ABSA: Dual-Stream Data Synthesis with Label Refinement for Few-Shot Aspect-Based Sentiment AnalysisSynthetic data and self-improvement for LLMs → Self-training for few-shot NLP
DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMsEfficient LLM inference and serving → KV cache optimization
Progressive Multimodal Reasoning via Active RetrievalLLM reasoning → Multimodal reasoning +1
Entropy Regularized Task Representation Learning for Offline Meta-Reinforcement LearningOffline RL and imitation learning from data → Decision transformers and sequence modeling
Generative CKM Construction using Partially Observed Data with Diffusion ModelDiffusion models → Diffusion for compression and wireless
ResoFilter: Fine-grained Synthetic Data Filtering for Large Language Models through Data-Parameter Resonance AnalysisSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Video Prediction Policy: A Generalist Robot Policy with Predictive Visual RepresentationsVision-language-action models → Human-video and latent-action pretraining
Disentangling Reasoning Tokens and Boilerplate Tokens For Language Model Fine-tuningLLM reasoning → Test-time compute and efficient thinking
Query pipeline optimization for cancer patient question answering systemsRetrieval-augmented generation → Medical and biomedical RAG
ReMoE: Fully Differentiable Mixture-of-Experts with ReLU RoutingMixture-of-experts → Expert routing strategies
EnergyMoGen: Compositional Human Motion Generation with Energy-Based Diffusion Model in Latent SpaceDiffusion models → Video generation
How to Synthesize Text Data without Model Collapse?Synthetic data and self-improvement for LLMs → LLM synthetic data generation
FiVL: A Framework for Improved Vision-Language Alignment through the Lens of Training, Evaluation and ExplainabilityLLM reasoning → Multimodal reasoning +1
Analysis and Visualization of Linguistic Structures in Large Language Models: Neural Representations of Verb-Particle Constructions in BERTMechanistic interpretability → Knowledge and representations inside LLMs
Unveiling Uncertainty: A Deep Dive into Calibration and Performance of Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Unified Image Restoration and Enhancement: Degradation Calibrated Cycle Reconstruction Diffusion ModelDiffusion models → Inverse problems and image restoration
Qua2SeDiMo: Quantifiable Quantization Sensitivity of Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Multi-modal, Multi-task, Multi-criteria Automatic Evaluation with Vision Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Beyond Guilt: Legal Judgment Prediction with Trichotomous ReasoningLLM reasoning → Reasoning for agents, planning and code
CORD: Balancing COnsistency and Rank Distillation for Robust Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
DiffSim: Taming Diffusion Models for Evaluating Visual SimilarityDiffusion models → Image generation, text-to-image and editing
SCKD: Semi-Supervised Cross-Modality Knowledge Distillation for 4D Radar Object DetectionKnowledge distillation → Detection and dense prediction distillation
Single-Loop Federated Actor-Critic across Heterogeneous EnvironmentsDeep reinforcement learning methods → RL for energy, networks and operations
S3-Mamba: Small-Size-Sensitive Mamba for Lesion SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Downscaling Precipitation with Bias-informed Conditional Diffusion ModelDiffusion models → Flow matching, time series and science
Consistent Human Image and Video Generation with Spatially Conditioned DiffusionDiffusion models → 3D, point cloud and scene generation
Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language ModelsKnowledge distillation → Knowledge distillation for LLMs
Knowledge Distillation in RNN-Attention Models for Early Prediction of Student PerformanceKnowledge distillation → General knowledge distillation methods
Efficient Self-Supervised Video Hashing with Selective State SpacesState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
Cal-DPO: Calibrated Direct Preference Optimization for Language Model AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
PA-RAG: RAG Alignment via Multi-Perspective Preference OptimizationRetrieval-augmented generation → RAG pipeline design and evaluation +1
A hybrid framework for effective and efficient machine unlearningLLM safety and security → Machine unlearning
WikiStyle+: A Multimodal Approach to Content-Style Representation Disentanglement for Artistic Image StylizationDiffusion models → Image generation, text-to-image and editing
Drive-1-to-3: Enriching Diffusion Priors for Novel View Synthesis of Real VehiclesDiffusion models → Diffusion and flow policies for robotics
Token Preference Optimization with Self-Calibrated Visual-Anchored Rewards for Hallucination MitigationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
MegaPairs: Massive Data Synthesis For Universal Multimodal RetrievalSynthetic data and self-improvement for LLMs → Self-improving MLLMs and synthetic data
Agent-SafetyBench: Evaluating the Safety of LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
DiffusionTrend: A Minimalist Approach to Virtual Fashion Try-OnDiffusion models → 3D, point cloud and scene generation
LiftRefine: Progressively Refined View Synthesis from 3D Lifting with Volume-Triplane RepresentationsDiffusion models → 3D, point cloud and scene generation
VISA: Retrieval Augmented Generation with Visual Source AttributionRetrieval-augmented generation → Multimodal RAG
Multimodal Latent Diffusion Model for Complex Sewing Pattern GenerationDiffusion models → 3D, point cloud and scene generation
VLM-AD: End-to-End Autonomous Driving through Vision-Language Model SupervisionLLM reasoning → Chain-of-thought reasoning for driving +2
FedPIA -- Permuting and Integrating Adapters leveraging Wasserstein Barycenters for Finetuning Foundation Models in Multi-Modal Federated LearningMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Enhancing Diffusion Models for High-Quality Image GenerationDiffusion models → Diffusion foundations and sampling
DriveGPT: Scaling Autoregressive Behavior Models for DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners
GenAIOps for GenAI Model-Agility
Counterexample Guided Program Repair Using Zero-Shot Learning and MaxSAT-based Fault Localization
Neuromorphic Spiking Neural Network Based Classification of COVID-19 Spike Sequences
Towards an Environmental Ethics of Artificial Intelligence
AICat: An AI Cataloguing Approach to Support the EU AI Act
Fundamental Risks in the Current Deployment of General-Purpose AI Models: What Have We (Not) Learnt From Cybersecurity?
A Unified Framework for Context-Aware IoT Management and State-of-the-Art IoT Traffic Anomaly Detection
GFormer: Accelerating Large Language Models with Optimized Transformers on Gaudi Processors

All 365