---
license: apache-2.0
---

# FAPO: Flawed-Aware Policy Optimization for Efficient and Reliable Reasoning

This Model is trained on the [FAPO-Reasoning-Dataset](https://huggingface.co/datasets/dyyyyyyyy/FAPO-Reasoning-Dataset) with generative rewards by [FAPO-GenRM-4B](https://huggingface.co/dyyyyyyyy/FAPO-GenRM-4B).

---

Project Homepage: https://fapo-rl.github.io/

Code Implementation: https://github.com/volcengine/verl/tree/main/recipe/fapo

Welcome to follow and cite our works!

BibTeX citation:
```bibtex
@article{ding2025fapo,
  title={FAPO: Flawed-Aware Policy Optimization for Efficient and Reliable Reasoning},
  author={Ding, Yuyang and Zhang, Chi and Li, Juntao and Lin, Haibin and Liu, Xin and Zhang, Min},
  journal={arXiv preprint arXiv:2510.22543},
  year={2025}
}
```