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 2405.14899cs.CLcs.AIcs.LG
收藏

DETAIL:面向可解释上下文学习的任务演示归因方法

DETAIL: Task DEmonsTration Attribution for Interpretable In-context Learning

National University of Singapore(新加坡国立大学);Singapore-MIT Alliance for Research and Technology(新加坡-麻省理工学院研究与技术联盟);Institute for Infocomm Research, A*STAR(新加坡科技研究局信息通信研究所);CSAIL, MIT(麻省理工学院计算机科学与人工智能实验室)

Zijian Zhou, Xiaoqiang Lin, Xinyi Xu, Alok Prakash, Daniela Rus, Bryan Kian Hsiang Low

AI总结:

本文提出基于影响函数的归因方法DETAIL,用于解释上下文学习中的任务演示作用,并通过演示重排序与策展提升模型性能,且归因分数可迁移至黑盒模型。

摘要:

上下文学习(ICL)允许基于Transformer的语言模型(这些模型在通用文本上进行了预训练)通过少量“任务演示”快速学习特定任务,而无需更新其参数,从而显著提升其灵活性和泛化能力。ICL具有许多与传统机器学习不同的显著特征,因此需要新的方法来解释这种学习范式。借鉴近期研究表明Transformer通过构建内部优化器进行上下文学习的观点,我们提出了一种基于影响函数的归因技术DETAIL,以应对ICL的特定特征。我们通过实验验证了该方法在演示归因方面的有效性,同时具有计算效率。利用这些结果,我们进一步展示了DETAIL如何通过演示重排序和策展来帮助改善真实场景中的模型性能。最后,我们通过实验证明DETAIL具有广泛的适用性,表明我们在白盒模型上获得的归因分数可以迁移到黑盒模型上,以提升模型性能。

英文摘要:

In-context learning (ICL) allows transformer-based language models that are pre-trained on general text to quickly learn a specific task with a few "task demonstrations" without updating their parameters, significantly boosting their flexibility and generality. ICL possesses many distinct characteristics from conventional machine learning, thereby requiring new approaches to interpret this learning paradigm. Taking the viewpoint of recent works showing that transformers learn in context by formulating an internal optimizer, we propose an influence function-based attribution technique, DETAIL, that addresses the specific characteristics of ICL. We empirically verify the effectiveness of our approach for demonstration attribution while being computationally efficient. Leveraging the results, we then show how DETAIL can help improve model performance in real-world scenarios through demonstration reordering and curation. Finally, we experimentally prove the wide applicability of DETAIL by showing our attribution scores obtained on white-box models are transferable to black-box models in improving model performance.

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。

↑