פיענוח מונע-סירוב: שימור התנהגות איפוס תחת דגימה בטמפרטורה גבוהה

תקציר מקורי באנגליתarXiv:2607.20791v2 Announce Type: replace-cross Abstract: Recent advances in truncation-based sampling have helped mitigate drawbacks of high-temperature sampling such as neural text degeneration, thereby enabling greater diversity without sacrificing coherence. However, increasing the entropy of the token probability distribution via high temperatures has also been shown to weaken the model's refusal response. Existing solutions for maintaining the refusal behavior of LLMs either replace the model's own refusal decision with a separate safety classifier or alter its output distribution for every prompt. To address this gap, we propose refusal-gated decoding (RGD): an efficient sequential decoding approach which preserves a model's greedy decoding refusal response at high temperatures and