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 2407.14487cs.CL
收藏

评估大型语言模型中自我解释的可靠性

Evaluating the Reliability of Self-Explanations in Large Language Models

Stockholm University(斯德哥尔摩大学);Athens University of Economics and Business(雅典经济与商业大学);Archimedes/Athena RC(阿基米德/雅典娜研究中心)

Korbinian Randl, John Pavlopoulos, Aron Henriksson, Tony Lindgren

AI总结:

本文评估了大型语言模型在分类任务中生成的抽取式和反事实式自我解释的可靠性,发现反事实解释能弥补感知与实际推理间的差距,提供比传统方法更忠实的替代方案。

摘要:

本文研究了大型语言模型(LLMs)在受到提示去解释其先前输出时所生成解释的可靠性。我们使用三个最先进的LLMs(2B到8B参数)在两个不同的分类任务(客观和主观)上评估了两种此类自我解释——抽取式和反事实式。我们的发现表明,虽然这些自我解释能与人类判断相关联,但它们并未完全且准确地遵循模型的决策过程,表明在感知到的模型推理与实际模型推理之间存在差距。我们展示了这一差距是可以被弥补的,因为提示LLMs生成反事实解释能够产生忠实的、信息丰富的且易于验证的结果。这些反事实解释为传统可解释性方法(例如SHAP、LIME)提供了一种有前景的替代方案,前提是提示需针对特定任务进行定制并检查其有效性。

英文摘要:

This paper investigates the reliability of explanations generated by large language models (LLMs) when prompted to explain their previous output. We evaluate two kinds of such self-explanations - extractive and counterfactual - using three state-of-the-art LLMs (2B to 8B parameters) on two different classification tasks (objective and subjective). Our findings reveal, that, while these self-explanations can correlate with human judgement, they do not fully and accurately follow the model's decision process, indicating a gap between perceived and actual model reasoning. We show that this gap can be bridged because prompting LLMs for counterfactual explanations can produce faithful, informative, and easy-to-verify results. These counterfactuals offer a promising alternative to traditional explainability methods (e.g. SHAP, LIME), provided that prompts are tailored to specific tasks and checked for validity.

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。

↑