arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2305.18153cs.CL
收藏

大型语言模型是否知道自己不知道什么?

Do Large Language Models Know What They Don't Know?

Fudan University(复旦大学);National University of Singapore(新加坡国立大学)

Zhangyue Yin, Qiushi Sun, Qipeng Guo, Jiawen Wu, Xipeng Qiu, Xuanjing Huang

AI总结:

本文通过构建SelfAware数据集并检测模型响应中的不确定性,评估了20个LLMs识别不可回答问题的自我知识能力,发现模型具有内在自我知识且可通过上下文学习和指令微调增强,但与人类水平仍有显著差距。

摘要:

大型语言模型(LLMs)拥有丰富的知识,使它们能够在各种自然语言处理(NLP)任务中表现出色。当前的研究侧重于在其已有知识范围内提升性能。尽管拥有海量知识,LLMs仍然受限于它们能够容纳和理解的信息量。因此,理解自身在未知事物上的局限性的能力,即自我知识,至关重要。本研究旨在通过评估LLMs识别不可回答或不可知问题的能力来评价其自我知识。我们引入了一种自动化方法,用于检测这些模型响应中的不确定性,为衡量其自我知识提供了一种新颖的度量。我们进一步引入了一个独特的数据集SelfAware,该数据集包含来自五个不同类别的不可回答问题及其可回答的对应问题。我们进行了涉及20个LLMs(包括GPT-3、InstructGPT和LLaMA)的广泛分析,发现这些模型具有内在的自我知识能力。此外,我们证明了上下文学习和指令微调可以进一步增强这种自我知识。尽管这一见解令人鼓舞,我们的发现也凸显了这些模型的能力与人类在识别自身知识局限性方面的熟练程度之间存在相当大的差距。

英文摘要:

Large language models (LLMs) have a wealth of knowledge that allows them to excel in various Natural Language Processing (NLP) tasks. Current research focuses on enhancing their performance within their existing knowledge. Despite their vast knowledge, LLMs are still limited by the amount of information they can accommodate and comprehend. Therefore, the ability to understand their own limitations on the unknows, referred to as self-knowledge, is of paramount importance. This study aims to evaluate LLMs' self-knowledge by assessing their ability to identify unanswerable or unknowable questions. We introduce an automated methodology to detect uncertainty in the responses of these models, providing a novel measure of their self-knowledge. We further introduce a unique dataset, SelfAware, consisting of unanswerable questions from five diverse categories and their answerable counterparts. Our extensive analysis, involving 20 LLMs including GPT-3, InstructGPT, and LLaMA, discovering an intrinsic capacity for self-knowledge within these models. Moreover, we demonstrate that in-context learning and instruction tuning can further enhance this self-knowledge. Despite this promising insight, our findings also highlight a considerable gap between the capabilities of these models and human proficiency in recognizing the limits of their knowledge.

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。

↑