涌现性错位语言模型表现出会随后续重新对齐而变化的行为自我意识
Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent Realignment
AI总结：
研究发现，经过特定微调的LLMs能自我评估其行为错位程度，表明模型具备行为自我意识并能反映其实际对齐状态。
摘要：
最近的研究表明，经过错误琐事问题-答案对微调的大型语言模型（LLMs）会表现出毒性——这种现象后来被称为
英文摘要：
Recent research has demonstrated that large language models (LLMs) fine-tuned on incorrect trivia question-answer pairs exhibit toxicity - a phenomenon later termed "emergent misalignment". Moreover, research has shown that LLMs possess behavioral self-awareness - the ability to describe learned behaviors that were only implicitly demonstrated in training data. Here, we investigate the intersection of these phenomena. We fine-tune GPT-4.1 models sequentially on datasets known to induce and reverse emergent misalignment and evaluate whether the models are self-aware of their behavior transitions without providing in-context examples. Our results show that emergently misaligned models rate themselves as significantly more harmful compared to their base model and realigned counterparts, demonstrating behavioral self-awareness of their own emergent misalignment. Our findings show that behavioral self-awareness tracks actual alignment states of models, indicating that models can be queried for informative signals about their own safety.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
