Navigating the Grey Area: How Expressions of Uncertainty and Overconfidence Affect Language Models
Kaitlyn Zhou Stanford University \AndDan Jurafsky Stanford University \AndTatsunori Hashimoto Stanford University
Abstract
The increased deployment of LMs for real-world tasks involving knowledge and facts makes it important to understand model epistemology: what LMs think they know, and how their attitudes toward that knowledge are affected by language use in their inputs. Here, we study an aspect of model epistemology: how epistemic markers of certainty, uncertainty, or evidentiality like "I’m sure it’s", "I think it’s", or “Wikipedia says it’s" affect models, and whether they contribute to model failures. We develop a typology of epistemic markers and inject 50 markers into prompts for question answering. We find that LMs are highly sensitive to epistemic markers in prompts, with accuracies varying more than 80%. Surprisingly, we find that expressions of high certainty result in a 7% decrease in accuracy as compared to low certainty expressions; similarly, factive verbs hurt performance, while evidentials benefit performance. Our analysis of a popular pretraining dataset shows that these markers of uncertainty are associated with answers on question-answering websites, while markers of certainty are associated with questions. These associations may suggest that the behavior of LMs is based on mimick
中文速览
大语言模型（LM）在回答问题时，究竟如何理解提问者话语中隐含的"我确定""我觉得""维基百科说"这类认知标记（epistemic markers），是否会因此出错？研究者构建了一套涵盖弱化词、强化词、事实动词、证据标记等50种认知标记的类型体系，将其注入标准问答提示，系统测试了GPT系列模型的表现，发现模型对这些标记极为敏感——不同措辞下同一批题目的准确率相差超过80%。出乎意料的是，使用"我百分之百确定"这类高确信度表达反而比使用"我觉得可能是"这类低确信度表达准确率低约7%，同样，事实动词（如"我们意识到是……"）也会显著拉低准确率，而注明信息来源的证据标记则有助于提升准确率。通过分析预训练语料，研究者发现：在问答网站上，不确定表达更多出现在答案中，而确定表达更多出现在问题里，这表明模型只是在模仿训练数据中的语言分布规律，而非真正理解认知不确定性——这一发现揭示了当前大模型在事实知识处理上的深层缺陷，对提高模型可靠性具有重要意义。
原文 arXiv:2302.13439;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2302.13439v2