arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.19630cs.AI
收藏

大语言模型对自然语言的理解能力揭示

LLMs' Understanding of Natural Language Revealed

Institute for Experiential AI, Northeastern University(东北大学体验式人工智能研究所)

Walid S. Saba

AI总结:

本文通过反向操作测试大语言模型的语言理解能力,揭示其理解仅停留在基于海量文本记忆的表面推断,并未真正理解语言。

摘要:

大语言模型(LLMs)是一场大规模自下而上、数据驱动的语言逆向工程实验的产物。尽管它们在众多下游自然语言处理任务中具有实用价值,但大量研究表明,LLMs 无法在需要量化推理和符号变量操作的任务(例如规划与问题求解)中执行推理;参见文献[25][26]。然而,在本文中,我们将重点测试 LLMs 的语言理解能力——这被认为是它们的强项。正如我们将在文中展示的那样,LLMs 的语言理解能力被广泛夸大了。虽然 LLMs 已被证明能够生成类似人类的连贯语言(因为其设计目标即如此),但它们的语言理解能力尚未得到恰当测试。特别是,我们认为应当通过与“文本生成”相反的操作来测试 LLMs 的语言理解能力,具体做法是将文本片段作为输入提供给 LLM,然后询问该 LLM“理解”了什么。正如我们在此所展示的,这样做之后将变得显而易见:LLMs 并未真正理解语言,其理解仅限于非常表面的推断,而这些推断本质上是对所摄入海量文本进行记忆的副产品。

英文摘要:

Large language models (LLMs) are the result of a massive experiment in bottom-up, data-driven reverse engineering of language at scale. Despite their utility in a number of downstream NLP tasks, ample research has shown that LLMs are incapable of performing reasoning in tasks that require quantification over and the manipulation of symbolic variables (e.g., planning and problem solving); see for example [25][26]. In this document, however, we will focus on testing LLMs for their language understanding capabilities, their supposed forte. As we will show here, the language understanding capabilities of LLMs have been widely exaggerated. While LLMs have proven to generate human-like coherent language (since that's how they were designed), their language understanding capabilities have not been properly tested. In particular, we believe that the language understanding capabilities of LLMs should be tested by performing an operation that is the opposite of 'text generation' and specifically by giving the LLM snippets of text as input and then querying what the LLM "understood". As we show here, when doing so it will become apparent that LLMs do not truly understand language, beyond very superficial inferences that are essentially the byproduct of the memorization of massive amounts of ingested text.

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。

↑