Reinforcement-Learning on Patrick Konhttps://www.cs-pk.com/tags/reinforcement-learning/Recent content in Reinforcement-Learning on Patrick KonHugo -- 0.128.0enFri, 25 Sep 2026 00:00:00 +0000SWE-Protégé: Learning to Selectively Collaborate With an Expert Unlocks Small Language Models as Software Engineering Agents. NeurIPS 2026.https://www.cs-pk.com/papers/14/Fri, 25 Sep 2026 00:00:00 +0000https://www.cs-pk.com/papers/14/SWE-Protégé post-trains small language models to selectively ask a strong expert model for help on software engineering tasks.