How Does Fine-tuning Affect the Geometry of Embedding Space: A Case Study on Isotropy
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.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Do LLM Embedding Spaces Recover Expert Structure?
Pretrained text embeddings are increasingly used as representational maps, yet high category separability does not imply that their geometry recovers expert-defined structure. We study this problem in mental-health-relat…
What Happens To BERT Embeddings During Fine-tuning?
While there has been much recent work studying how linguistic information is encoded in pre-trained sentence representations, comparatively little is understood about how these models change when adapted to solve downstr…
Dependency ParsingSentenceA Closer Look at How Fine-tuning Changes BERT
Given the prevalence of pre-trained contextualized representations in today's NLP, there have been many efforts to understand what information they contain, and why they seem to be universally successful. The most common…
Does an LSTM forget more than a CNN? An empirical study of catastrophic forgetting in NLP
Catastrophic forgetting {---} whereby a model trained on one task is fine-tuned on a second, and in doing so, suffers a {``}catastrophic{''} drop in performance over the first task {---} is a hurdle in the development of…
Continual LearningTransfer LearningSoft Prompt Tuning for Cross-Lingual Transfer: When Less is More
Soft Prompt Tuning (SPT) is a parameter-efficient method for adapting pre-trained language models (PLMs) to specific tasks by inserting learnable embeddings, or soft prompts, at the input layer of the PLM, without modify…
Cross-Lingual Transfer