用知识图谱引导大模型，让医学和数学推理更可靠
MedRule-KG: A Knowledge-Graph--Steered Scaffold for Reliable Mathematical and Biomedical Reasoning
- 用结构化知识图谱注入提示，约束生成过程
- 90项任务中违规减少83.2%，准确率提升
- 轻量验证器零延迟，适合交互式药物设计
我们研究如何在用于科学推理和早期药物发现的大语言模型中引入领域一致的结构。提出MedRule-KG，一个紧凑的知识图谱框架与轻量级验证器相结合，引导生成符合数学和生物医学规律的结果。系统通过在提示中注入精选符号事实，并利用确定性检查器强制规则满足。我们将生成形式化为约束推理，引入适用于解码的软引导代理，并进行了全面的统计分析与不确定性量化。在涵盖反应可行性、代谢相容性和毒性筛查的90项任务中，相比强基线链式思维方法，违规次数降低83.2%，同时提升精确匹配率。结果在分层条件下保持稳定，且随数据集规模增长，验证器带来的延迟可忽略不计，使该方法适用于交互式设计。
原文摘要 · Abstract (English)
We study how to impose domain-consistent structure on large language models (LLMs) used for scientific reasoning and early-stage drug discovery. We present MedRule-KG, a compact knowledge-graph scaffold paired with a lightweight verifier that steers generation toward mathematically and biomedically valid outputs. The system injects curated symbolic facts into prompts and then enforces rule satisfaction with a deterministic checker. We formalize generation as constrained inference, introduce a soft guidance surrogate suitable for decoding, and perform a thorough statistical analysis with uncertainty quantification. Across 90 tasks spanning reaction feasibility, metabolic compatibility, and toxicity screening, MedRule-KG reduces violation counts by 83.2\% relative to a strong chain-of-thought baseline while improving exact match. Results remain stable under stratification and scale with dataset size, and the verifier adds negligible latency, making the approach practical for interactive design.
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。