{"slug": "one-to-more-more-to-one-category-aware-iterative-expert-training-for-software", "title": "One to More, More to One: Category-Aware Iterative Expert Training for Software Engineering Agents", "summary": "A new training method called Category-Aware Iterative Expert Training (One to More, More to One) targets uneven progress in repository-level software engineering agents, where pooled agentic reinforcement learning produces gains in some task categories alongside regressions in others. The approach addresses how aggregate resolution rates obscure these category-level changes in SWE agents.", "body_md": "Repository-level software engineering (SWE) comprises heterogeneous task categories, whose progress under pooled agentic reinforcement learning can be uneven: gains in some categories coincide with regressions in others, while aggregate resolution obscures these changes. Motivated by this category s", "url": "https://wpnews.pro/news/one-to-more-more-to-one-category-aware-iterative-expert-training-for-software", "canonical_source": "https://aiflash.com/news/124107/", "published_at": "2026-09-22 02:30:13+00:00", "updated_at": "2026-09-22 02:53:20.295950+00:00", "lang": "en", "topics": ["ai-agents", "ai-research", "machine-learning", "developer-tools"], "entities": [], "alternates": {"html": "https://wpnews.pro/news/one-to-more-more-to-one-category-aware-iterative-expert-training-for-software", "markdown": "https://wpnews.pro/news/one-to-more-more-to-one-category-aware-iterative-expert-training-for-software.md", "text": "https://wpnews.pro/news/one-to-more-more-to-one-category-aware-iterative-expert-training-for-software.txt", "jsonld": "https://wpnews.pro/news/one-to-more-more-to-one-category-aware-iterative-expert-training-for-software.jsonld"}}