Skip to content
QbitAI · WeChat

Zhejiang University and Microsoft use 3,000 text prompts to improve video 3D consistency with World-R1

解决视频生成穿帮问题!浙大&微软3000条纯文本让模型理解3D

Zhejiang University and Microsoft introduced World-R1, training Wan 2.1 with about 3,000 text-only prompts, Flow-GRPO, and a four-part reward; the 1.3B version improves PSNR over the baseline by 10.23 dB.

Why it matters: World-R1 from Zhejiang University and Microsoft aligns Wan 2.1 video generation using about 3,000 text-only prompts and no 3D annotations, a cheap route for teams without labeled data.

Read the original ↗Export Markdown