Papers

New on arXiv

Wed 5 Feb 2025

372 papers

PRISM: Perspective Reasoning for Integrated Synthesis and Mediation as a Multi-Perspective Framework for AI AlignmentLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Personalized Image Generation for Recommendations Beyond CatalogsDiffusion models → Image generation, text-to-image and editing
Cache-Craft: Managing Chunk-Caches for Efficient Retrieval-Augmented GenerationRetrieval-augmented generation → Generation and context integration +1
QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV CacheEfficient LLM inference and serving → Speculative decoding
Reasoning-as-Logic-Units: Scaling Test-Time Reasoning in Large Language Models Through Logic Unit AlignmentLLM reasoning → Reasoning benchmarks and evaluation
Optimizing Robustness and Accuracy in Mixture of Experts: A Dual-Model ApproachMixture-of-experts → Vision mixture-of-experts
RLOMM: An Efficient and Robust Online Map Matching Framework with Reinforcement LearningDeep reinforcement learning methods → RL with diffusion and flow models
CTR-Driven Advertising Image Generation with Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
DiffListener: Discrete Diffusion Model for Listener GenerationDiffusion models → Audio, speech and music
KDA: A Knowledge-Distilled Attacker for Generating Diverse Prompts to Jailbreak LLMsLLM safety and security → Jailbreaking
Overcoming Vision Language Model Challenges in Diagram Understanding: A Proof-of-Concept with XML-Driven Large Language Models SolutionsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Enhancing Reasoning to Adapt Large Language Models for Domain-Specific ApplicationsLLM reasoning → Reasoning benchmarks and evaluation
Sparse Autoencoders for Hypothesis GenerationMechanistic interpretability → SAE features and their applications
Limitations of Large Language Models in Clinical Problem-Solving Arising from Inflexible ReasoningLLM reasoning → Clinical and biomedical applications
DILLEMA: Diffusion and Large Language Models for Multi-Modal AugmentationDiffusion models → Image generation, text-to-image and editing
An Analysis for Reasoning Bias of Language Models with Small InitializationLLM reasoning → Reasoning benchmarks and evaluation
PerPO: Perceptual Preference Optimization via Discriminative RewardingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
DreamDPO: Aligning Text-to-3D Generation with Human Preferences via Direct Preference OptimizationLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
Lost in Edits? A λ-Compass for AIGC ProvenanceDiffusion models → Robustness, privacy and adversarial aspects
On-device Sora: Enabling Training-Free Diffusion-based Text-to-Video Generation for Mobile DevicesDiffusion models → Video generation
MARAGE: Transferable Multi-Model Adversarial Attack for Retrieval-Augmented Generation Data ExtractionRetrieval-augmented generation → Privacy and security of RAG
Advancing Reasoning in Large Language Models: Promising Methods and ApproachesLLM reasoning → Reasoning benchmarks and evaluation
Unrealized Expectations: Comparing AI Methods vs Classical Algorithms for Maximum Independent SetDeep reinforcement learning methods → RL for games and combinatorial optimization
Towards Physical Understanding in Video Generation: A 3D Point Regularization ApproachDiffusion models → 3D, point cloud and scene generation
The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models via Visual Information SteeringMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
The Logical Implication Steering Method for Conditional Interventions on Transformer GenerationMechanistic interpretability → Circuits and mechanistic analysis
(GG) MoE vs. MLP on Tabular DataMixture-of-experts → MoE theory and gating functions
Simultaneous Multi-Robot Motion Planning with Projected Diffusion ModelsDiffusion models → Diffusion and flow policies for robotics
HACK: Homomorphic Acceleration via Compression of the Key-Value Cache for Disaggregated LLM InferenceEfficient LLM inference and serving → KV cache optimization
TD-M(PC)2: Improving Temporal Difference MPC Through Policy ConstraintDeep reinforcement learning methods → RL theory and sample complexity
Path Planning for Masked Diffusion Model SamplingDiffusion models → Discrete and text diffusion
YINYANG-ALIGN: Benchmarking Contradictory Objectives and Proposing Multi-Objective Optimization based DPO for Text-to-Image AlignmentLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
DC-VSR: Spatially and Temporally Consistent Video Super-Resolution with Video Diffusion PriorDiffusion models → Inverse problems and image restoration
Efficient Image Restoration via Latent Consistency Flow MatchingDiffusion models → Inverse problems and image restoration
FreqPrior: Improving Video Diffusion Models with Frequency Filtering Gaussian NoiseDiffusion models → Video generation
Teaching Language Models to Critique via Reinforcement LearningLLM reasoning → Reasoning for agents, planning and code
LanPaint: Training-Free Diffusion Inpainting with Asymptotically Exact and Fast Conditional SamplingDiffusion models → 3D, point cloud and scene generation
SKI Models: Skeleton Induced Vision-Language Embeddings for Understanding Activities of Daily LivingMultimodal LLMs / vision-language models → Video understanding with MLLMs
Schema-Guided Scene-Graph Reasoning based on Multi-Agent Large Language Model SystemLLM reasoning → Spatial and embodied reasoning +1
Dress-1-to-3: Single Image to Simulation-Ready 3D Outfit with Diffusion Prior and Differentiable PhysicsDiffusion models → 3D, point cloud and scene generation
Masked Autoencoders Are Effective Tokenizers for Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Taking a Big Step: Large Learning Rates in Denoising Score Matching Prevent MemorizationDiffusion models → Diffusion foundations and sampling
On Fairness of Unified Multimodal Large Language Model for Image GenerationMultimodal LLMs / vision-language models → Social bias and fairness in VLMs
Deep Reinforcement Learning-Based Optimization of Second-Life Battery Utilization in Electric Vehicles Charging StationsDeep reinforcement learning methods → RL for energy, networks and operations
Detecting Strategic Deception Using Linear ProbesMechanistic interpretability → Probing deception, bias and beliefs
SPRI: Aligning Large Language Models with Context-Situated PrinciplesLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
LIMO: Less is More for ReasoningLLM reasoning → Mathematical reasoning
Demystifying Long Chain-of-Thought Reasoning in LLMsLLM reasoning → Test-time compute and efficient thinking
RadVLM: A Multitask Conversational Vision-Language Model for RadiologyMultimodal LLMs / vision-language models → Medical vision-language models
A Mixture-Based Framework for Guiding Diffusion ModelsDiffusion models → Inverse problems and image restoration
Electronic Circuit Principles of Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation +1
Out-of-Distribution Detection using Synthetic Data GenerationSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Conditional Prediction by Simulation for Automated DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners
SymAgent: A Neural-Symbolic Self-Learning Agent Framework for Complex Reasoning over Knowledge GraphsLLM reasoning → Knowledge- and tool-augmented reasoning +1
Token Assorted: Mixing Latent and Text Tokens for Improved Language Model ReasoningLLM reasoning → Test-time compute and efficient thinking
A Scalable Approach to Probabilistic Neuro-Symbolic Robustness VerificationLLM reasoning → Reasoning benchmarks and evaluation
How do Humans and Language Models Reason About Creativity? A Comparative AnalysisLLM reasoning → Reasoning benchmarks and evaluation
Calibrated Unsupervised Anomaly Detection in Multivariate Time-series using Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
Exploring the Security Threats of Knowledge Base Poisoning in Retrieval-Augmented Code GenerationRetrieval-augmented generation → Privacy and security of RAG
Efficient Vision Language Model Fine-tuning for Text-based Person Anomaly SearchMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
iVISPAR -- An Interactive Visual-Spatial Reasoning Benchmark for VLMsLLM reasoning → Spatial and embodied reasoning +1
Scalable In-Context Learning on Tabular Data via Retrieval-Augmented Large Language ModelsRetrieval-augmented generation → Generation and context integration
Underwater Soft Fin Flapping Motion with Deep Neural Network Based Surrogate ModelDeep reinforcement learning methods → RL for robot navigation and UAV control
HiLo: Learning Whole-Body Human-like Locomotion with Motion Tracking ControllerDeep reinforcement learning methods → RL for robot locomotion and manipulation
Reveal the Mystery of DPO: The Connection between DPO and RL AlgorithmsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
IAO Prompting: Making Knowledge Flow Explicit in LLMs through Structured Reasoning TemplatesLLM reasoning → Prompting and chain-of-thought reasoning
RoboGrasp: A Universal Grasping Policy for Robust Robotic ControlWorld models → Embodied and robot world models +1
Understanding and Enhancing the Transferability of Jailbreaking AttacksLLM safety and security → Jailbreaking
RepLoRA: Reparameterizing Low-Rank Adaptation via the Perspective of Mixture of ExpertsMixture-of-experts → Mixture of LoRA experts
The Cake that is Intelligence and Who Gets to Bake it: An AI Analogy and its Implications for ParticipationDeep reinforcement learning methods → RL training of language and agent models
Knowledge Distillation from Large Language Models for Household Energy ModelingKnowledge distillation → Knowledge distillation for LLMs
Analyze Feature Flow to Enhance Interpretation and Steering in Language ModelsMechanistic interpretability → Knowledge and representations inside LLMs
Scaling Laws for Upcycling Mixture-of-Experts Language ModelsMixture-of-experts → MoE large language models
MedBioLM: Optimizing Medical and Biological QA with Fine-Tuned Large Language Models and Retrieval-Augmented GenerationRetrieval-augmented generation → Medical and biomedical RAG
Training an LLM-as-a-Judge Model: Pipeline, Insights, and Practical LessonsLLM reasoning → Reasoning benchmarks and evaluation +1
MobileA3gent: Training Mobile GUI Agents Using Decentralized Self-Sourced Data from Diverse UsersLLM agents and tool use → GUI and mobile agents
Efficiently Disentangling CLIP for Multi-Object PerceptionMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Distributional Statistics Restore Training Data Auditability in One-step Distilled Diffusion ModelsDiffusion models → Diffusion foundations and sampling
ReachAgent: Enhancing Mobile Agent via Page Reaching and OperationLLM agents and tool use → GUI and mobile agents
Direct Distributional Optimization for Provable Alignment of Diffusion ModelsDiffusion models → Diffusion and flow policies for robotics +1
LLM-KT: Aligning Large Language Models with Knowledge Tracing using a Plug-and-Play InstructionLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Behavioral Homophily in Social Media via Inverse Reinforcement Learning: A Reddit Case StudyOffline RL and imitation learning from data → Inverse reinforcement learning +1
Fast T2T: Optimization Consistency Speeds Up Diffusion-Based Training-to-Testing Solving for Combinatorial OptimizationDiffusion models → Diffusion foundations and sampling
Elucidating the Preconditioning in Consistency DistillationDiffusion models → Diffusion foundations and sampling
Robust Reward Alignment via Hypothesis Space Batch CuttingLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Interactive Symbolic Regression through Offline Reinforcement Learning: A Co-Design FrameworkOffline RL and imitation learning from data → Decision transformers and sequence modeling
Policy Abstraction and Nash Refinement in Tree-Exploiting PSRODeep reinforcement learning methods → RL for games and combinatorial optimization
Position: Multimodal Large Language Models Can Significantly Advance Scientific ReasoningLLM reasoning → Multimodal reasoning +1
Towards Large-Scale In-Context Reinforcement Learning by Meta-Training in Randomized WorldsDeep reinforcement learning methods → Deep RL algorithms and representations
Domain-Invariant Per-Frame Feature Extraction for Cross-Domain Imitation Learning with Visual ObservationsOffline RL and imitation learning from data → Imitation learning methods
OceanChat: The Effect of Virtual Conversational AI Agents on Sustainable Attitude and Behavior ChangeLLM agents and tool use → Social simulation and multi-agent debate
Rethinking Latent Redundancy in Behavior Cloning: An Information Bottleneck Approach for Robot ManipulationOffline RL and imitation learning from data → Robot manipulation imitation learning
Task-Aware Virtual Training: Enhancing Generalization in Meta-Reinforcement Learning for Out-of-Distribution TasksDeep reinforcement learning methods → Deep RL algorithms and representations
CAMI: A Counselor Agent Supporting Motivational Interviewing through State Inference and Topic ExplorationLLM agents and tool use → Medical and scientific domain agents
SimMark: A Robust Sentence-Level Similarity-Based Watermarking Algorithm for Large Language ModelsLLM safety and security → Watermarking generated content and models
Multimodal Transformer Models for Turn-taking Prediction: Effects on Conversational Dynamics of Human-Agent Interaction during Cooperative Gameplay
A Beautiful Mind: Principles and Strategies for AI-Augmented Human Reasoning
Circuit Diagram Retrieval Based on Hierarchical Circuit Graph Representation
OPTIC: Optimizing Patient-Provider Triaging & Improving Communications in Clinical Operations using GPT-4 Data Labeling and Model Distillation
Emerging Practices in Frontier AI Safety Frameworks

All 372