paper-with-me

홈 › Papers

Zero-training Sentence Embedding via Orthogonal Basis

2019-05-01 · ICLR 2019 5 · Ziyi Yang, Chenguang Zhu, Weizhu Chen

We propose a simple and robust training-free 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 its 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 representation. This approach requires zero training and zero parameters, along with efficient inference performance. We evaluate our approach on 11 downstream NLP tasks. Experimental results show that our model outperforms all existing zero-training alternatives in all the tasks and it is competitive to other approaches relying on either large amounts of labelled data or prolonged training time.

📄 PDF Abstract BibTeX

Code (1)

fursovia/geometric_embedding

Tasks

SentenceSentence EmbeddingSentence-EmbeddingWord Embeddings

Similar Papers 제목 키워드 기반

Parameter-free Sentence Embedding via Orthogonal Basis

2018-09-30 · IJCNLP 2019 11 · Ziyi Yang, Chenguang Zhu, Weizhu Chen

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 an…

SentenceSentence EmbeddingSentence-EmbeddingWord Embeddings

Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding

2024-01-17 · Depeng Li, Tianqi Wang, Junwei Chen, Qining Ren 외

Deep neural networks are susceptible to catastrophic forgetting when trained on sequential tasks. Various continual learning (CL) methods often rely on exemplar buffers or/and network expansion for balancing model stabil…

Continual Learning

Mitigating Semantic Leakage in Cross-lingual Embeddings via Orthogonality Constraint

2024-09-24 · Dayeon Ki, Cheonbok Park, Hyunjoong Kim

Accurately aligning contextual representations in cross-lingual sentence embeddings is key for effective parallel data mining. A common strategy for achieving this alignment involves disentangling semantics and language …

DisentanglementRepresentation LearningSemantic Textual SimilaritySentence+1

Inference-InfoGAN: Inference Independence via Embedding Orthogonal Basis Expansion

2021-10-02 · Hongxiang Jiang, Jihao Yin, Xiaoyan Luo, Fuxiang Wang

Disentanglement learning aims to construct independent and interpretable latent variables in which generative models are a popular strategy. InfoGAN is a classic method via maximizing Mutual Information (MI) to obtain in…

AttributeDisentanglement

Linear Cross-Lingual Mapping of Sentence Embeddings

2023-05-23 · Oleg Vasilyev, Fumika Isono, John Bohannon

Semantics of a sentence is defined with much less ambiguity than semantics of a single word, and we assume that it should be better preserved by translation to another language. If multilingual sentence embeddings intend…

SentenceSentence EmbeddingsTranslation