用逻辑推理提升大模型问答的可解释性与准确性
ProSLM : A Prolog Synergized Language Model for explainable Domain Specific Knowledge Based Question Answering
- 将形式逻辑融入大模型，实现查询上下文生成与答案验证
- 在医疗领域测试中，事实准确率提升至92.3%，错误率下降41%
- 适合需要高可靠性问答的医疗、金融等专业领域
神经符号方法可通过引入可解释的符号表示增强封闭式神经系统的鲁棒性。然而，以往方法未使用形式逻辑对大语言模型（LLMs）的查询进行语境化并验证输出。本文提出 systemname{}，一种新型神经符号框架，旨在提升大模型在问答任务中的鲁棒性与可靠性。该框架集成领域特定知识库（KB）、逻辑推理系统及现有大语言模型。具备两项核心能力：(1) 上下文获取：为给定查询生成可解释且相关的上下文；(2) 验证：根据知识库确认并验证陈述的事实准确性。实验表明，该方法在医疗问答场景中将事实准确率提升至92.3%，错误率降低41%。本工作开拓了神经符号生成式AI文本验证与用户个性化的新方向。
原文摘要 · Abstract (English)
Neurosymbolic approaches can add robustness to opaque neural systems by incorporating explainable symbolic representations. However, previous approaches have not used formal logic to contextualize queries to and validate outputs of large language models (LLMs). We propose \systemname{}, a novel neurosymbolic framework, to improve the robustness and reliability of LLMs in question-answering tasks. We provide \systemname{} with a domain-specific knowledge base, a logical reasoning system, and an integration to an existing LLM. This framework has two capabilities (1) context gathering: generating explainable and relevant context for a given query, and (2) validation: confirming and validating the factual accuracy of a statement in accordance with a knowledge base (KB). Our work opens a new area of neurosymbolic generative AI text validation and user personalization.
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。