研究人员开发了MoRFI（单调稀疏自编码器特征识别）来更好地理解大型语言模型（LLM）如何产生幻觉。通过在Llama 3.1 8B和Gemma 2 9B等模型上使用新知识进行微调，他们观察到长时间的训练会加剧幻觉。MoRFI分析模型的内部状态，以识别残差流中与这些事实不准确性有因果关系的特定方向，从而能够进行有针对性的干预以恢复正确知识。 AI
影响 通过识别特定的内部知识检索路径，提供了一种诊断和潜在缓解LLM幻觉的方法。
排序理由 介绍分析LLM行为新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2604.26866v1 Announce Type: new Abstract: Large language models (LLMs) acquire most of their factual knowledge during the pre-training stage, through next token prediction. Subsequent stages of post-training often introduce new facts outwith the parametric knowledge, giving…
Large language models (LLMs) acquire most of their factual knowledge during the pre-training stage, through next token prediction. Subsequent stages of post-training often introduce new facts outwith the parametric knowledge, giving rise to hallucinations. While it has been demon…