arXiv 2605.295552026-05-28cs.CL

从盲目猜测到知情判断：通过构建知识增强的偏好信号教会LLM评估材料

From Blind Guess to Informed Judgment: Teaching LLMs to Evaluate Materials by Building Knowledge-Augmented Preference Signals

School of Computer Engineering & Science, Shanghai University（上海大学计算机工程与科学学院）；Center of Materials Informatics and Data Science, Materials Genome Institute, Shanghai University（上海大学材料信息与数据科学中心）；Key Laboratory of Silicate Cultural Relics Conservation (Shanghai University), Ministry of Education, China（教育部硅酸盐文化 relics 保护重点实验室（上海大学））；Shanghai Institute for Advanced Communication and Data Science, Shanghai University（上海大学高级通信与数据科学研究院）

Yeyong Yu, Wenya Hu, Xing Wu, Quan Qian

备注：: 33 pages, 5 figures

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出知识增强偏好信号框架MaterEval，通过成对偏好数据引导大语言模型从直觉判断转向基于证据的可靠评估，并引入快慢推理方案平衡吞吐量、成本和可靠性，在高熵合金评估中验证了有效性。

摘要：

随着候选生成和高通量实验的进步，材料发现的主要瓶颈正从性质预测转向在大量候选集中进行可靠评估。我们提出了知识增强偏好信号框架MaterEval，该框架自动为同一候选生成两种评估：一种遵循专家规则并提供支持证据的知情判断，另一种是移除规则的盲目猜测。通过将这两种评估配对作为偏好数据，我们引导原本缺乏材料特定标准的通用大语言模型（LLM）从直觉判断转向由明确证据支持的可靠评估。为了平衡吞吐量、成本和可靠性，我们进一步引入了一种快慢推理方案，将大规模快速筛选与对小子集的深入审查解耦。以高熵合金（HEA）评估为例，我们表明，无需外部检索，仅依赖内化能力，小型开源LLM在准确性、结论一致性和证据区分度上取得了显著提升，接近基于规则的闭源LLM的性能。这些结果表明，专家规则可以系统地转化为可学习的偏好信号，从而为自主材料发现循环提供低成本且可部署的评估模块。

英文摘要：

As candidate generation and high-throughput experimentation advance, the primary bottleneck in materials discovery is shifting from property prediction to making reliable evaluations among massive candidate sets. We propose a Knowledge-Augmented Preference Signals Framework, MaterEval, that automatically produces, for the same candidate, two evaluations: an informed judgment that follows expert rules and provides supporting evidence, and a rule-removed blind guess. By pairing the two evaluations as preference data, we guide general-purpose large language models (LLMs), originally lacking materials-specific criteria, from intuitive judgment toward reliable evaluation supported by explicit evidence. To balance throughput, cost, and reliability, we further introduce a fast-slow reasoning scheme that decouples large-scale rapid screening from in-depth review on a small subset. Using high-entropy alloy (HEA) assessment as a case study, we show that, without external retrieval and relying solely on internalized capabilities, small open-source LLMs achieve substantial gains in accuracy, conclusion consistency, and evidence discrimination, approaching the performance of rule-based closed-source LLMs. These results demonstrate that expert rules can be systematically transformed into learnable preference signals, enabling a low-cost and deployable evaluation module for autonomous materials discovery loops.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。