Sentence Similarity Based on Contexts

Xiaofei Sun, Yuxian Meng, Xiang Ao, Fei Wu, Tianwei Zhang, Jiwei Li, Chun Fan

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.

Refbacks

  • There are currently no refbacks.


Copyright (c) 2022 Association for Computational Linguistics

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.