English

Sentence Similarity Based on Contexts

Computation and Language 2022-02-01 v2

Abstract

Existing methods to measure sentence similarity are faced with two challenges: (1) labeled datasets are usually limited in size, making them insufficient to train supervised neural models; (2) there is a training-test gap for unsupervised language modeling (LM) based models to compute semantic scores between sentences, since sentence-level semantics are not explicitly modeled at training. This results in inferior performances in this task. In this work, we propose a new framework to address these two issues. The proposed framework is based on the core idea that the meaning of a sentence should be defined by its contexts, and that sentence similarity can be measured by comparing the probabilities of generating two sentences given the same context. The proposed framework is able to generate high-quality, large-scale dataset with semantic similarity scores between two sentences in an unsupervised manner, with which the train-test gap can be largely bridged. Extensive experiments show that the proposed framework achieves significant performance boosts over existing baselines under both the supervised and unsupervised settings across different datasets.

Keywords

Cite

@article{arxiv.2105.07623,
  title  = {Sentence Similarity Based on Contexts},
  author = {Xiaofei Sun and Yuxian Meng and Xiang Ao and Fei Wu and Tianwei Zhang and Jiwei Li and Chun Fan},
  journal= {arXiv preprint arXiv:2105.07623},
  year   = {2022}
}

Comments

Accepted by TACL; pre-MIT Press publication version

R2 v1 2026-06-24T02:09:59.636Z