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 2408.10474cs.SEcs.AIcs.CLcs.CRcs.LG
收藏

LeCov:大型语言模型的多级测试准则

LeCov: Multi-level Testing Criteria for Large Language Models

University of Alberta(阿尔伯塔大学);The University of Tokyo(东京大学);New York University(纽约大学)

Xuan Xie, Jiayang Song, Yuheng Huang, Da Song, Fuyuan Zhang, Felix Juefei-Xu, Lei Ma

AI总结:

针对LLMs缺乏系统测试准则的问题,提出多级测试准则LeCov,覆盖注意力、前馈神经元和不确定性,经实验验证有效。

摘要:

大型语言模型(LLMs)被广泛应用于许多不同领域,但由于其可解释性有限,人们对其在真实性、毒性等各个方面的可信度存有疑问。近期研究已开始开发针对LLMs的测试方法,旨在部署前发现不可信问题(即缺陷)。然而,系统化且形式化的测试准则仍然缺乏,这阻碍了对测试探索范围和充分性的全面评估。为缓解这一威胁,我们提出了一套针对LLMs的多级测试准则LeCov。该准则考虑了LLMs的三个关键内部组件,即注意力机制、前馈神经元和不确定性,并总共包含九种类型的测试准则。我们将这些准则应用于两个场景:测试优先级排序和覆盖引导测试。在三个模型和四个数据集上的实验评估证明了LeCov的实用性和有效性。

英文摘要:

Large Language Models (LLMs) are widely used in many different domains, but because of their limited interpretability, there are questions about how trustworthy they are in various perspectives, e.g., truthfulness and toxicity. Recent research has started developing testing methods for LLMs, aiming to uncover untrustworthy issues, i.e., defects, before deployment. However, systematic and formalized testing criteria are lacking, which hinders a comprehensive assessment of the extent and adequacy of testing exploration. To mitigate this threat, we propose a set of multi-level testing criteria, LeCov, for LLMs. The criteria consider three crucial LLM internal components, i.e., the attention mechanism, feed-forward neurons, and uncertainty, and contain nine types of testing criteria in total. We apply the criteria in two scenarios: test prioritization and coverage-guided testing. The experiment evaluation, on three models and four datasets, demonstrates the usefulness and effectiveness of LeCov.

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。

↑