一篇新的研究论文探讨了大型语言模型（LLM）为何在多跳推理方面存在困难，即使它们拥有所需的单个事实。研究发现，模型在组合来自不同事实的信息以回答新问题时会失败，例如从两个相关信息推断出生日期。这种失败归因于预训练阶段缺乏对组合式上下文的暴露，而不是知识的缺失。 AI
影响 强调了LLM推理的一个基本限制，表明改进需要改变预训练数据的构成。
排序理由 关于LLM推理局限性的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
Large Language Models fail at implicit multi-hop reasoning: a model answers "When was $X$ born?" and "Who is $Y$'s closest friend?" correctly but fails on "When was $Y$'s closest friend born?" in a single forward pass, even when both facts are perfectly memorized and individually…
Large Language Models fail at implicit multi-hop reasoning: a model answers "When was $X$ born?" and "Who is $Y$'s closest friend?" correctly but fails on "When was $Y$'s closest friend born?" in a single forward pass, even when both facts are perfectly memorized and individually…