arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2506.04534cs.CLcs.AI
收藏

这仅仅是语义问题吗?大语言模型中话语小品词理解的案例研究

Is It JUST Semantics? A Case Study of Discourse Particle Understanding in LLMs

Linguistics, The University of Texas at Austin(德克萨斯大学奥斯汀分校语言学系);Toyota Technological Institute at Chicago(芝加哥丰田技术研究所);Allen Institute for AI(人工智能研究院);University of Washington(华盛顿大学)

William Sheffield, Kanishka Misra, Valentina Pyatkin, Ashwini Deo, Kyle Mahowald, Junyi Jessy Li

AI总结:

本文通过专家标注数据研究大语言模型区分英语“just”细粒度语义的能力,发现模型虽能区分宽泛类别但难以捕捉细微差别,揭示了其在话语小品词理解上的不足。

摘要:

话语小品词是微妙地塑造文本意义的关键元素。这些词往往具有多功能性,会产生细微且常常相当不同的语义/话语效果,正如小品词“just”的多种用法(例如,排他性、时间性、强调性)所体现的那样。本研究利用由专家语言学家精心创建和标注的数据,考察了大语言模型区分英语“just”的细粒度语义的能力,该词是形式语义学中一个被充分研究的例子。我们的研究结果表明,尽管大语言模型表现出一定的区分较宽泛类别的能力,但它们难以完全捕捉更细微的差别,这凸显了它们在理解话语小品词方面的不足。

英文摘要:

Discourse particles are crucial elements that subtly shape the meaning of text. These words, often polyfunctional, give rise to nuanced and often quite disparate semantic/discourse effects, as exemplified by the diverse uses of the particle "just" (e.g., exclusive, temporal, emphatic). This work investigates the capacity of LLMs to distinguish the fine-grained senses of English "just", a well-studied example in formal semantics, using data meticulously created and labeled by expert linguists. Our findings reveal that while LLMs exhibit some ability to differentiate between broader categories, they struggle to fully capture more subtle nuances, highlighting a gap in their understanding of discourse particles.

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。

↑