AI & ML
SWE-growth: Teacher-Guided Adaptive Self-Play for Software Engineering Agents
arXiv cs.SE0 views
arXiv:2611.23876v1 Announce Type: new Abstract: Self-play offers a scalable way to train software engineering agents by generating repository-level repair tasks. However, with binary repair rewards and finite rollout budgets, all-fail groups provide no reward-based group-relative advantage. Such tasks may still offer learning opportunities when a Teacher can provide verified repairs absent from the Student's own attempts. We introduce SWE-Growth, a teacher-guided adaptive self-play framework that uses verified Teacher outcomes to identify learning opportunities for the current Student. Tasks with all students failure but reliable Teacher success provide demonstrations for supervised fine-tuning (SFT), and post-SFT Student rollouts identify mixed-outcome tasks with a positive Teacher--Student gap for reinforcement learning. Task eligibility also guides Injector updates, connecting Student training with subsequent task generation. Starting from 101 seed issues, \textsc{SWE-Growth} improves Qwen3-Coder-30B-A3B-Instruct from 27.7% to 31.7% on SWE-bench Pro and from 11.0% to 17.2% on held-out TerminalBench 2.1 by Round~2. It exceeds Student-only self-play by 2.6 and 4.1 percentage points, respectively, and Teacher-augmented self-play by 1.8 and 2.5 points under approximately matched budgets. These results support adapting both supervision and task selection to the Student's evolving repair capability.
Source:https://arxiv.org/abs/2611.23876PreviousElephantAgent: Contextual State Continuity in Agentic SystemsNextQuantifying the Intrusiveness of Retrofitting Retrieval-Augmented Generation onto Legacy Software Systems
Related posts

Roblox money is pushing a controversial anorexia treatment
0The Verge

Canva adds an AI ‘off’ button
0The Verge
I Need Your Help Choosing a Final Year Project
0dev.to
GitHub Copilot Review 2026: Is It Still Worth Paying For?
0dev.to
AI Notetakers Now Record Your Client Meetings in Person. Who Asked for Consent?
0dev.to

How to Interpret Logistic Regression and Odds Ratios
0dev.to