Yu-Wing Tai

CoT-RVS: Zero-Shot Chain-of-Thought Reasoning Segmentation for Videos

Trace Anything: Representing Any Video in 4D via Trajectory Fields

Scale-DiT: Ultra-High-Resolution Image Generation with Hierarchical Local Attention

Agentic 3D Scene Generation with Spatially Contextualized VLMs

MangaCrafter: Training-Free Consistent Manga Generation via Staged Diffusion

ReasonNav: Human-Inspired Global Map Reasoning for Zero-Shot Embodied Navigation

GeoReasoning: Structured Semantic Reasoning for Image-to-Map Localization

MA-RAG: Multi-Agent Retrieval-Augmented Generation via Collaborative Chain-of-Thought Reasoning

LLM-Navi: Navigating Motion Agents in Cluttered and Dynamic Environments via LLM reasoning

CoT-Seg: Rethinking Reasoning Segmentation with Chain of Thoughts and Auto Correction

Stable Segment Anything Model

Motion-Agent: A Conversational Framework for Human Motion Generation with LLMs

LayerCraft: Enhancing Text-to-Image Generation with CoT Reasoning and Layered Object Integration

VP-LLM: Text-Driven 3D Volume Completion with Large Language Models through Patchification

Audio-Agent: Leveraging LLMs For Audio Generation, Editing and Composition

Reward-RAG: Enhancing RAG with Reward Driven Supervision

H-QLoRA: Enhancing Quantized LLMs with Hierarchical Residual Learning