arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.13587cs.CLcs.HC
收藏

T3-Vis:一个可视化分析框架,用于在NLP中训练和微调Transformers

T3-Vis: a visual analytic framework for Training and fine-Tuning Transformers in NLP

University of British Columbia(不列颠哥伦比亚大学);Huawei Cananda Technologies Co. Ltd.(华为加拿大技术有限公司)

Raymond Li, Wen Xiao, Lanjun Wang, Hyeju Jang, Giuseppe Carenini

摘要:

Transformers是NLP中的主导架构,但它们的训练和微调仍然非常具有挑战性。在本文中,我们提出了一个视觉分析框架的设计和实现,以帮助研究人员在这样的过程中,通过为他们提供有价值的见解模型的内在属性和行为。我们的框架提供了一个直观的概述,允许用户探索模型的不同方面(例如,隐藏状态、注意力),并允许一套内置算法来计算模型组件和输入序列不同部分的重要性。案例研究和用户焦点小组的反馈表明,该框架是有用的,并提出了几项改进建议。

英文摘要:

Transformers are the dominant architecture in NLP, but their training and fine-tuning is still very challenging. In this paper, we present the design and implementation of a visual analytic framework for assisting researchers in such process, by providing them with valuable insights about the model's intrinsic properties and behaviours. Our framework offers an intuitive overview that allows the user to explore different facets of the model (e.g., hidden states, attention) through interactive visualization, and allows a suite of built-in algorithms that compute the importance of model components and different parts of the input sequence. Case studies and feedback from a user focus group indicate that the framework is useful, and suggest several improvements.

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。

↑