Mahesh Pasupuleti
Author directory2026
Large Reasoning Models Learn Better Alignment from Flawed Thinking
ShengYun Peng | Pin-Yu Chen | Eric Michael Smith | Song Jiang | Hongyuan Zhan | Haozhu Wang | Mahesh Pasupuleti | Duen Horng Chau | Jianfeng Chi
Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
ShengYun Peng | Pin-Yu Chen | Eric Michael Smith | Song Jiang | Hongyuan Zhan | Haozhu Wang | Mahesh Pasupuleti | Duen Horng Chau | Jianfeng Chi
Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
Large reasoning models (LRMs) "think" by generating structured chain-of-thought (CoT) before producing a final answer, yet they still lack the ability to reason critically about safety alignment and are easily biased when a flawed premise is injected into their thought process. We propose RECAP (Robust Safety Alignment via Counter-Aligned Prefilling), a principled reinforcement learning (RL) method for post-training that explicitly teaches models to override flawed reasoning trajectories and reroute to safe and helpful responses. RECAP trains on a mixture of synthetically generated counter-aligned CoT prefills and standard prompts, integrates seamlessly into standard reinforcement learning from human feedback (RLHF) pipelines without additional RL optimization steps or architectural modifications, and substantially improves safety and jailbreak robustness, reduces overrefusal, and preserves core reasoning capability — all while maintaining inference token budget. Extensive analysis shows that RECAP-trained models engage in self-reflection more frequently and remain robust under adaptive attacks, preserving safety even after repeated attempts to override their reasoning. Our code is publicly available at https://github.com/poloclub/recap.