Banger paper from Tencent on recursive self-improvement for terminal agents.
(bookmark it)
This is a great read if you are tracking research on training terminal or coding agents with RL
Starting from Qwen3.8-27B, they alternate two phases over five stages. One phase fine-tunes on rewritten versions of trajectories that passed the verifier. The other runs PPO with a critic inherited from the previous stage.
The four-suite Terminal-Bench average goes from 20.68% to 50.58%, with 86.52% on Terminal-Bench 2.1 and 79.00% on Terminal-Bench Hard.
The reward audit is just as useful. Under a dense progress reward, failed runs collected 50.09% of the reward, and 48 failures passed every executed check but still missed the final test. Separating strict success from partial credit cut that share to 28.72%.
https://arxiv.org/abs/2610.22192
Chat with Paper: https://academy.dair.ai/papers/terminal-rsi-recursive-self-improving-terminal-agents-via-recursive-self-rewrite-2610.22192