Yanhong Zeng

Researcher@Ant Group·中国·OpenReview

研究方向

inpainting · image and video synthesis and editing · diffusion model · image inpainting · video inpainting · Generative adversarial network · GAN · image and video synthesis

5.5

LEGO-Puzzles: How Good Are MLLMs at Multi-Step Spatial Reasoning?

ICLR 2026Rejected

三作

5.0

CharacterShot: Controllable and Consistent 4D Character Animation

ICLR 2026Rejected

4.0

Ditto: Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset

ICLR 2026Withdrawn

7.3

Live2Diff: Live Stream Translation via Uni-directional Attention in Video Diffusion Models

合作者 (20)

Yanhong Zeng

LEGO-Puzzles: How Good Are MLLMs at Multi-Step Spatial Reasoning?

CharacterShot: Controllable and Consistent 4D Character Animation

Ditto: Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset

WorldMem: Long-term Consistent World Simulation with Memory

FOLEYCRAFTER: BRING SILENT VIDEOS TO LIFE WITH LIFELIKE AND SYNCHRONIZED SOUNDS

StyleShot: A snapshot on any style

Live2Diff: Live Stream Translation via Uni-directional Attention in Video Diffusion Models