Author name: John T. Halloran

arXiv:2605.11217v1 Announce Type: new
Abstract: Large language model (LLM) alignment algorithms typically consist of post-training over preference pairs. While such algorithms are widely used to enable safety guardrails and align LLMs with general hum…

arXiv:2604.02574v1 Announce Type: cross
Abstract: Safety alignment has become a critical step to ensure LLMs refuse harmful requests while providing helpful and harmless responses. However, despite the ubiquity of safety alignment for deployed frontie…