← Monday, October 5, 2026AgentsPaper

Recursive self-rewrite turns harness-specific solutions into training trajectories

huggingface.co by Zongxia Li et al. Editor score 8/10

RSR uses a planner, critic and executor to convert successful solutions from diverse harnesses into trajectories for a general harness; fine-tuning raises pass@3 on Terminal-Bench 2 from 57.0% to 74.2%. The results show how harness-generated experience can improve a base model across terminal tasks.

What: Recursive Self-Rewrite reconstructs successful solutions found under specialized harnesses as training trajectories for a general harness.
Method: A planner extracts runbooks, a critic screens and revises them, and an executor follows qualified runbooks in fresh sandboxes.
Why it matters: The approach turns harness-assisted task solutions into reusable supervised training data for models.

RSR uses runbooks refined by a critic before an executor follows them in fresh sandboxes. Sketched from the article's own text.

Read on huggingface.co →

Discussion

Discussion is not switched on yet. Use Note above to keep your own thoughts on this story.

Headline, summary, brief and diagram are written by a model from the source's own text; the link, byline and numbers come from the source. Read the original before relying on a claim.