Effect of Post-processing on Contextualized Word Representations
원문 초록
Post-processing of static embedding has beenshown to improve their performance on both lexical and sequence-level tasks. However, post-processing for contextualized embeddings is an under-studied problem. In this work, we question the usefulness of post-processing for contextualized embeddings obtained from different layers of pre-trained language models. More specifically, we standardize individual neuron activations using z-score, min-max normalization, and by removing top principle components using the all-but-the-top method. Additionally, we apply unit length normalization to word representations. On a diverse set of pre-trained models, we show that post-processing unwraps vital information present in the representations for both lexical tasks (such as word similarity and analogy)and sequence classification tasks. Our findings raise interesting points in relation to theresearch studies that use contextualized representations, and suggest z-score normalization as an essential step to consider when using them in an application.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Analyzing Continuous Semantic Shifts with Diachronic Word Similarity Matrices
LGDE: Local Graph-based Dictionary Expansion
A Likelihood Ratio Test of Genetic Relationship among Languages
Are ELECTRA's Sentence Embeddings Beyond Repair? The Case of Semantic Textual Similarity
Visual Grounding Helps Learn Word Meanings in Low-Data Regimes
Improving Heterogeneous Graph Learning with Weighted Mixed-Curvature Product Manifold
토론
이 논문과 연결된 포럼 글이 아직 없습니다.