arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2411.10080cs.CLcs.CY
收藏

理解温度对与人类意见对齐的影响

Understanding The Effect Of Temperature On Alignment With Human Opinions

Queen Mary University of London(伦敦玛丽女王大学);University of Utrecht(乌得勒支大学)

Maja Pavlovic, Massimo Poesio

AI总结:

本文实证分析了三种获取LLM意见分布的方法,发现通过简单参数调整的采样和对数概率方法在主观任务中比直接提示更能有效对齐人类意见,同时指出需进一步研究人类主观性对模型不确定性的影响。

摘要:

随着LLMs能力的增强,近期研究关注理解它们代表了谁的意见以及如何有效提取对齐的意见分布。我们对三种获取分布的简单方法进行了实证分析,并在多种指标上评估了结果。研究结果表明,与直接提示相比,通过简单参数调整的采样和对数概率方法能在主观任务中返回对齐度更高的输出。然而,假设模型反映人类意见可能具局限性,凸显了进一步研究人类主观性如何影响模型不确定性的必要性。

英文摘要:

With the increasing capabilities of LLMs, recent studies focus on understanding whose opinions are represented by them and how to effectively extract aligned opinion distributions. We conducted an empirical analysis of three straightforward methods for obtaining distributions and evaluated the results across a variety of metrics. Our findings suggest that sampling and log-probability approaches with simple parameter adjustments can return better aligned outputs in subjective tasks compared to direct prompting. Yet, assuming models reflect human opinions may be limiting, highlighting the need for further research on how human subjectivity affects model uncertainty.

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。

↑