arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.00290cs.CLcs.AI
收藏

大规模大语言模型红队测试:应对数学任务中的幻觉

Red Teaming for Large Language Models At Scale: Tackling Hallucinations on Mathematics Tasks

Delft University of Technology(代尔夫特理工大学)

Aleksander Buszydlik, Karol Dobiczek, Michał Teodor Okoń, Konrad Skublicki, Philip Lippmann, Jie Yang

AI总结:

本文提出一个程序化生成数学问题的框架,通过红队测试评估不同提示技术对LLM输出质量的影响,发现结构化推理能减缓质量下降,但gpt-3.5-turbo和gpt-4在基础计算与推理任务上表现依然不佳。

摘要:

我们探讨了在基础计算和代数任务上对LLMs进行红队测试的问题,以评估各种提示技术如何影响输出质量。我们提出了一个程序化生成数字问题和谜题的框架,并比较了应用与不应用几种红队技术的结果。研究结果表明,尽管结构化推理和提供解答示例减缓了答案质量的下降,但gpt-3.5-turbo和gpt-4模型仍不适合基础计算与推理任务,即便在遭受红队测试时亦是如此。

英文摘要:

We consider the problem of red teaming LLMs on elementary calculations and algebraic tasks to evaluate how various prompting techniques affect the quality of outputs. We present a framework to procedurally generate numerical questions and puzzles, and compare the results with and without the application of several red teaming techniques. Our findings suggest that even though structured reasoning and providing worked-out examples slow down the deterioration of the quality of answers, the gpt-3.5-turbo and gpt-4 models are not well suited for elementary calculations and reasoning tasks, also when being red teamed.

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。

↑