arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2109.04740cs.CL
收藏

微调如何影响嵌入空间的几何性质:关于各向同性的一个案例研究

How Does Fine-tuning Affect the Geometry of Embedding Space: A Case Study on Isotropy

Iran University of Science and Technology(伊朗科技大学);Tehran Institute for Advanced Studies, Khatam University(德黑兰高级研究所,哈塔姆大学)

Sara Rajaee, Mohammad Taher Pilehvar

摘要:

人们普遍认为,微调预训练的语言模型通常会带来下游任务的性能改进。然而,有有限的研究背后的原因,这种有效性,特别是从嵌入空间的结构变化的角度。为了填补这一空白,本文分析了微调后嵌入空间各向同性的变化程度。我们证明,即使各向同性是一个理想的几何属性,微调并不一定会导致各向同性增强。此外,预先训练的上下文单词表示(CWRs)中的局部结构,例如编码令牌类型或频率的结构,在微调期间会发生巨大的变化。我们的实验表明,嵌入空间中的细长方向的数量急剧增长,与预先训练的CWRs相比,这些方向在微调的嵌入空间中携带基本的语言知识,使得现有的各向同性增强方法无效。

英文摘要:

It is widely accepted that fine-tuning pre-trained language models usually brings about performance improvements in downstream tasks. However, there are limited studies on the reasons behind this effectiveness, particularly from the viewpoint of structural changes in the embedding space. Trying to fill this gap, in this paper, we analyze the extent to which the isotropy of the embedding space changes after fine-tuning. We demonstrate that, even though isotropy is a desirable geometrical property, fine-tuning does not necessarily result in isotropy enhancements. Moreover, local structures in pre-trained contextual word representations (CWRs), such as those encoding token types or frequency, undergo a massive change during fine-tuning. Our experiments show dramatic growth in the number of elongated directions in the embedding space, which, in contrast to pre-trained CWRs, carry the essential linguistic knowledge in the fine-tuned embedding space, making existing isotropy enhancement methods ineffective.

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。

↑