arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.00499cs.CLcs.AIcs.LG
收藏

ConU:具有正确性覆盖率保证的大型语言模型中的共形不确定性

ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees

Department of Computer Science, Drexel University(德雷塞尔大学计算机科学系);Department of Computer Science, University of Illinois Chicago(伊利诺伊大学芝加哥分校计算机科学系)

Zhiyuan Wang, Jinhao Duan, Lu Cheng, Yue Zhang, Qingni Wang, Xiaoshuang Shi, Kaidi Xu, Hengtao Shen, Xiaofeng Zhu

AI总结:

本研究提出ConU方法,通过基于自洽性理论的不确定性度量和共形预测准则,在黑盒LLMs的开放式NLG任务中实现严格的正确性覆盖率保证,并生成高效的小规模校准预测集。

摘要:

自然语言生成(NLG)任务中的不确定性量化(UQ)仍是一个开放性挑战,而最新大型语言模型(LLMs)的闭源性质加剧了这一问题。本研究探讨了在开放式NLG任务中,将共形预测(CP)应用于黑盒LLMs,CP能将任何启发式不确定性概念转化为严格的预测集。我们基于自洽性理论引入了一种新颖的不确定性度量,随后通过将符合正确性的不确定性条件整合到CP算法中,开发了一种共形不确定性准则。实证评估表明,我们的不确定性度量优于先前的最先进方法。此外,我们在4个自由形式NLG数据集上利用7个流行的LLMs,实现了对正确性覆盖率的严格控制,涵盖通用和医疗场景。此外,小规模的校准预测集进一步突显了我们的方法在为实际开放式NLG应用提供可信保证方面的效率。

英文摘要:

Uncertainty quantification (UQ) in natural language generation (NLG) tasks remains an open challenge, exacerbated by the closed-source nature of the latest large language models (LLMs). This study investigates applying conformal prediction (CP), which can transform any heuristic uncertainty notion into rigorous prediction sets, to black-box LLMs in open-ended NLG tasks. We introduce a novel uncertainty measure based on self-consistency theory, and then develop a conformal uncertainty criterion by integrating the uncertainty condition aligned with correctness into the CP algorithm. Empirical evaluations indicate that our uncertainty measure outperforms prior state-of-the-art methods. Furthermore, we achieve strict control over the correctness coverage rate utilizing 7 popular LLMs on 4 free-form NLG datasets, spanning general-purpose and medical scenarios. Additionally, the calibrated prediction sets with small size further highlights the efficiency of our method in providing trustworthy guarantees for practical open-ended NLG applications.

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。

↑