研究大模型用语气词表达不确定性的可靠性，发现跨场景时效果不稳定。
Revisiting Epistemic Markers in Confidence Estimation: Can Markers Accurately Reflect Large Language Models' Uncertainty?
- 用模型使用语气词时的正确率定义标记置信度，评估其一致性。
- 同分布下标记可信，但跨分布时准确率显著下降。
- 提示开发者谨慎依赖语气词判断模型不确定性。
随着大语言模型在高风险领域应用增多，准确评估其置信度至关重要。人类通常通过语气词（如“相当有把握”）表达信心，而非数值。然而，由于难以量化不同语气词对应的实际不确定性，尚不清楚大模型是否能一致地使用这些标记反映内在信心。为此，我们首次将标记置信度定义为模型使用特定语气词时的观测准确率，并在多个问答数据集上，对开源与专有大模型在分布内与分布外场景下的表现进行了评估。结果表明，语气词在同分布下具有较好泛化能力，但在分布外场景中置信度表现不一致。这一发现揭示了基于语气词的置信度估计存在严重可靠性问题，凸显了提升标记置信度与模型真实不确定性之间对齐的必要性。代码已公开于 https://github.com/HKUST-KnowComp/MarConf。
原文摘要 · Abstract (English)
As large language models (LLMs) are increasingly used in high-stakes domains, accurately assessing their confidence is crucial. Humans typically express confidence through epistemic markers (e.g., "fairly confident") instead of numerical values. However, it remains unclear whether LLMs consistently use these markers to reflect their intrinsic confidence due to the difficulty of quantifying uncertainty associated with various markers. To address this gap, we first define marker confidence as the observed accuracy when a model employs an epistemic marker. We evaluate its stability across multiple question-answering datasets in both in-distribution and out-of-distribution settings for open-source and proprietary LLMs. Our results show that while markers generalize well within the same distribution, their confidence is inconsistent in out-of-distribution scenarios. These findings raise significant concerns about the reliability of epistemic markers for confidence estimation, underscoring the need for improved alignment between marker based confidence and actual model uncertainty. Our code is available at https://github.com/HKUST-KnowComp/MarConf.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。