Abstract
We propose a simple and robust non-parameterized approach for building sentence representations. Inspired by the Gram-Schmidt Process in geometric theory, we build an orthogonal basis of the subspace spanned by a word and its surrounding context in a sentence. We model the semantic meaning of a word in a sentence based on two aspects. One is its relatedness to the word vector subspace already spanned by its contextual words. The other is the word’s novel semantic meaning which shall be introduced as a new basis vector perpendicular to this existing subspace. Following this motivation, we develop an innovative method based on orthogonal basis to combine pre-trained word embeddings into sentence representations. This approach requires zero parameters, along with efficient inference performance. We evaluate our approach on 11 downstream NLP tasks. Our model shows superior performance compared with non-parameterized alternatives and it is competitive to other approaches relying on either large amounts of labelled data or prolonged training time.- Anthology ID:
- D19-1059
- Volume:
- Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
- Month:
- November
- Year:
- 2019
- Address:
- Hong Kong, China
- Editors:
- Kentaro Inui, Jing Jiang, Vincent Ng, Xiaojun Wan
- Venues:
- EMNLP | IJCNLP
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 638–648
- Language:
- URL:
- https://aclanthology.org/D19-1059
- DOI:
- 10.18653/v1/D19-1059
- Cite (ACL):
- Ziyi Yang, Chenguang Zhu, and Weizhu Chen. 2019. Parameter-free Sentence Embedding via Orthogonal Basis. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 638–648, Hong Kong, China. Association for Computational Linguistics.
- Cite (Informal):
- Parameter-free Sentence Embedding via Orthogonal Basis (Yang et al., EMNLP-IJCNLP 2019)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/D19-1059.pdf
- Code
- ziyi-yang/GEM
- Data
- MPQA Opinion Corpus, SST