# PKU-Spirit AI Joint Lab's RoboFFT Accepted to CoRL2026, Uses Reinforcement Learning Post-Training to Generate Robotic Policies

- Source: PsiBot/灵初智能
- Published: 2026-11-03T06:38:56.000Z
- RoboHot: https://robohot.news/en/items/qxyw8oy5d24lwiabok3ef8soo
- Original: https://www.psibot.ai/corl2026-%e8%ae%a9%e6%9c%ba%e5%99%a8%e4%ba%ba%e4%bb%8e%e4%ba%a4%e4%ba%92%e4%b8%ad%e6%8c%81%e7%bb%ad%e5%ad%a6%e4%b9%a0%ef%bc%81robofft%e6%8f%90%e5%87%ba%e9%9d%a2%e5%90%91%e7%94%9f%e6%88%90%e5%bc%8f/

## Summary

RoboFFT, proposed by the PKU-Spirit AI Joint Lab, has been accepted to CoRL2026. It is a reinforcement learning post-training method for generative robotic policies that reuses the forward noising process from diffusion model and flow model pretraining to construct an approximate probability ratio for PPO-style clipped updates.

## Why it matters

The method reuses the forward noising loss of diffusion and flow models to construct policy update signals, offering a reusable approach for post-training generative policies.