发现大模型用外部工具时会与自身知识冲突，且现有方法无效
Investigating Tool-Memory Conflicts in Tool-Augmented LLMs
- 提出工具-记忆冲突新问题：模型内部知识与外部工具知识相悖
- 实测主流模型在理工科任务中冲突严重，且优先级随条件变化
- 验证提示和RAG方法均无法有效解决此类冲突，适用于需高可信推理的场景
工具增强型大语言模型已广泛应用于各类场景，但易出现知识冲突。本文提出一种新型知识冲突——工具-记忆冲突（TMC），即模型内部参数化知识与外部工具知识不一致。研究发现，尽管现有大模型能力强大，但在理工科任务中普遍存在TMC问题。我们进一步揭示，在不同条件下，工具知识与参数化知识的优先级会动态变化。随后评估了基于提示和RAG的冲突缓解技术，结果表明这些方法均无法有效解决工具-记忆冲突。
原文摘要 · Abstract (English)
Tool-augmented large language models (LLMs) have powered many applications. However, they are likely to suffer from knowledge conflict. In this paper, we propose a new type of knowledge conflict -- Tool-Memory Conflict (TMC), where the internal parametric knowledge contradicts with the external tool knowledge for tool-augmented LLMs. We find that existing LLMs, though powerful, suffer from TMC, especially on STEM-related tasks. We also uncover that under different conditions, tool knowledge and parametric knowledge may be prioritized differently. We then evaluate existing conflict resolving techniques, including prompting-based and RAG-based methods. Results show that none of these approaches can effectively resolve tool-memory conflicts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。