Author name: Bilgehan Sel, Vaishakh Keshava, Phillip Wallis, Lukas Rutishauser, Ming Jin, Dingcheng Li

arXiv:2602.08377v2 Announce Type: replace-cross
Abstract: Addressing the critical need for robust safety in Large Language Models (LLMs), particularly against adversarial attacks and in-distribution errors, we introduce Reinforcement Learning with Bac…