跳到正文
arXiv cs.AI· Liyan Wang, Longze Li, Mohamed Hassan, Isshu Lee, Dajie Sun, Ruimin Feng, Aleksandar Vakanski·· 56 分钟前AI 评分33

NEBench R1:用于评估大语言模型核工程领域能力的推理基准

NEBench R1: a Reasoning Benchmark for Evaluating Nuclear Domain Competence in Large Language Models

AI 导读

研究者推出推理基准 NEBench R1,包含从美国核管理委员会(NRC)反应堆操作员执照考试提取的 7,546 道选择题,覆盖 18 个核设施,并用官方解释作为推理评估参考。对 6 个大语言模型的测试显示整体准确率在 34.0% 到 73.9% 之间;基于 LLM 评判的推理质量评估发现,答对并不等于推理技术可靠,且更长的推理链与更低准确率相关,推理 token 消耗可作为识别不可靠回答的信号。

正文 · 原文

View PDF

Abstract:Large language models (LLMs) are increasingly being explored for knowledge-intensive and safety-critical applications in nuclear engineering. However, existing nuclear-domain benchmarks provide limited content coverage or focus primarily on answer accuracy without systematically evaluating the quality of the underlying reasoning. This study introduces NEBench R1, a reasoning-oriented benchmark comprising 7,546 unique multiple-choice questions extracted from U.S. Nuclear Regulatory Commission (NRC) reactor operator licensing examinations covering 18 nuclear facilities. Each question is classified according to the NRC Knowledge and Abilities catalogs and accompanied by an official explanation serving as a reference for reasoning evaluation. Six LLMs were evaluated, with overall accuracy ranging from 34.0% to 73.9% across three distinct performance tiers. To assess reasoning quality, an LLM-based judge evaluated correctly answered responses using a multidimensional rubric. Although responses generally demonstrated strong problem comprehension, they scored lower in factual and procedural accuracy and distractor analysis, indicating that correct answers were not necessarily supported by technically sound justifications. Response clarity was also weakly indicative of technical quality. Furthermore, longer reasoning traces were associated with lower accuracy in both frontier reasoning models, suggesting that reasoning length may reflect question difficulty rather than improved problem-solving performance. Reasoning-token consumption may therefore provide an additional signal for identifying potentially unreliable responses. NEBench R1 establishes a benchmark for evaluating both answer accuracy and reasoning quality, providing a foundation for assessing the domain competence and reliability of LLMs in nuclear engineering.
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.21682 [cs.AI]
  (or arXiv:2611.21682v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.21682

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Liyan Wang Ms. [view email]
[v1] Sun, 15 Nov 2026 19:02:18 UTC (545 KB)

来源:arXiv cs.AI · arxiv.org