arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.19115cs.CLcs.AIcs.DL
收藏

大语言模型如何编码科学质量?通过稀疏自编码器的单义特征进行实证研究

How Do LLMs Encode Scientific Quality? An Empirical Study Using Monosemantic Features from Sparse Autoencoders

Knowledge Media Institute, The Open University, Walton Hall, Milton Keynes, MK7 6AA, United Kingdom(开放大学知识媒体学院);Department of Business and Law, University of Milano Bicocca, Milan, IT(米兰比可卡大学商学院)

Michael McCoubrey, Angelo Salatino, Francesco Osborne, Enrico Motta

AI总结:

本文通过稀疏自编码器的单义特征研究,揭示了大语言模型如何编码科学质量的多个维度。

摘要:

近年来,生成式人工智能,特别是大语言模型(LLMs)在支持科学工作的评估和生成方面得到了广泛应用。尽管一些研究显示LLMs在一定程度上可以根据感知质量评估研究,但对我们能够实现这一能力的内部机制的理解仍然有限。本文首次研究了通过相关单义特征提取稀疏自编码器来调查LLMs如何编码科学质量的概念。我们根据不同的实验设置推导出这些特征,并评估它们在三个与研究质量相关的任务中的预测能力:预测引用次数、期刊SJR和期刊h指数。结果表明,LLMs编码与科学质量多个维度相关的特征。特别是,我们识别出四种反复出现的特征类型,捕捉了研究质量的各个方面:1)反映研究方法的特征;2)与出版类型相关的特征,文献综述通常具有更高的影响;3)与高影响研究领域和技术相关的特征;4)与特定科学术语相关的特征。这些发现是理解LLMs如何封装与研究质量相关概念的重要一步。

英文摘要:

In recent years, there has been a growing use of generative AI, and large language models (LLMs) in particular, to support both the assessment and generation of scientific work. Although some studies have shown that LLMs can, to a certain extent, evaluate research according to perceived quality, our understanding of the internal mechanisms that enable this capability remains limited. This paper presents the first study that investigates how LLMs encode the concept of scientific quality through relevant monosemantic features extracted using sparse autoencoders. We derive such features under different experimental settings and assess their ability to serve as predictors across three tasks related to research quality: predicting citation count, journal SJR, and journal h-index. The results indicate that LLMs encode features associated with multiple dimensions of scientific quality. In particular, we identify four recurring types of features that capture key aspects of how research quality is represented: 1) features reflecting research methodologies; 2) features related to publication type, with literature reviews typically exhibiting higher impact; 3) features associated with high-impact research fields and technologies; and 4) features corresponding to specific scientific jargons. These findings represent an important step toward understanding how LLMs encapsulate concepts related to research quality.

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。

↑