AI Onderzoekspapers Dagelijks

Critical Tokens Matter: Token-Level Contrastive Estimation Enhence LLM's Reasoning Capability

VideoGen-of-Thought: A Collaborative Framework for Multi-Shot Video Generation

MALT: Improving Reasoning with Multi-Agent LLM Training

Free Process Rewards without Process Labels

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

AV-Odyssey Bench: Can Your Multimodal LLMs Really Understand Audio-Visual Information?

OCR Hinders RAG: Evaluating the Cascading Impact of OCR on Retrieval-Augmented Generation

Truth or Mirage? Towards End-to-End Factuality Evaluation with LLM-OASIS

Motion Prompting: Controlling Video Generation with Motion Trajectories

OmniCreator: Self-Supervised Unified Generation with Universal Editing

LSceneLLM: Enhancing Large 3D Scene Understanding Using Adaptive Visual Preferences

Scaling Image Tokenizers with Grouped Spherical Quantization

MaskRIS: Semantic Distortion-aware Data Augmentation for Referring Image Segmentation

Generating a Low-code Complete Workflow via Task Decomposition and RAG

A dynamic parallel method for performance optimization on hybrid CPUs

VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval