arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.03129cs.CL
收藏

大语言模型的社会偏见评估需要提示变体

Social Bias Evaluation for Large Language Models Requires Prompt Variations

Tokyo Institute of Technology(东京工业大学);MBZUAI(穆罕默德·本·扎耶德人工智能大学);National Institute of Informatics(情报信息学研究所)

Rem Hida, Masahiro Kaneko, Naoaki Okazaki

AI总结:

本文研究大语言模型社会偏见评估对提示变体的敏感性,发现提示变化会导致模型排名波动并揭示性能与偏见之间的权衡,建议使用多样化提示进行评估。

摘要:

警告:本文包含刻板印象和偏见的示例。大语言模型(LLMs)表现出相当大的社会偏见,各种研究试图准确评估和缓解这些偏见。先前的研究使用下游任务作为提示来考察社会偏见的程度,以进行评估和缓解。虽然LLMs的输出高度依赖于提示,但先前评估和缓解偏见的研究往往依赖于有限的提示变体。在本文中,我们通过分析LLMs的任务表现和社会偏见,研究LLMs在改变提示变体(任务指令和提示、少样本示例、去偏提示)时的敏感性。我们的实验结果表明,LLMs对提示高度敏感,以至于在比较模型的任务表现和社会偏见时,LLMs的排名会发生波动。此外,我们表明LLMs在提示导致的表现和社会偏见之间存在权衡。来自提示设置的较少偏见可能导致性能下降。此外,实例的歧义性是先进LLMs对提示敏感的原因之一,导致输出多样化。我们建议使用多样化的提示,如本研究所示,以比较提示对LLMs社会偏见的影响。

英文摘要:

Warning: This paper contains examples of stereotypes and biases. Large Language Models (LLMs) exhibit considerable social biases, and various studies have tried to evaluate and mitigate these biases accurately. Previous studies use downstream tasks as prompts to examine the degree of social biases for evaluation and mitigation. While LLMs' output highly depends on prompts, previous studies evaluating and mitigating bias have often relied on a limited variety of prompts. In this paper, we investigate the sensitivity of LLMs when changing prompt variations (task instruction and prompt, few-shot examples, debias-prompt) by analyzing task performance and social bias of LLMs. Our experimental results reveal that LLMs are highly sensitive to prompts to the extent that the ranking of LLMs fluctuates when comparing models for task performance and social bias. Additionally, we show that LLMs have tradeoffs between performance and social bias caused by the prompts. Less bias from prompt setting may result in reduced performance. Moreover, the ambiguity of instances is one of the reasons for this sensitivity to prompts in advanced LLMs, leading to various outputs. We recommend using diverse prompts, as in this study, to compare the effects of prompts on social bias in LLMs.

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。

↑