SPECIALIZED PRE-TRAINING OF NEURAL NETWORKS ON SYNTHETIC DATA FOR IMPROVING PARAPHRASE GENERATION

Author:

,Skurzhanskyi O.H.,Marchenko O.O., ,Anisimov A.V.,

Abstract

SPECIALIZED PRE-TRAINING OF NEURAL NETWORKS ON SYNTHETIC DATA FOR IMPROVING PARAPHRASE GENERATION Abstract. Generating paraphrases is a fundamental problem in natural language processing. In light of the significant success of transfer learning technology, the “pre-training fine-tuning” approach has become the standard. However, popular general-purpose pre-training methods typically require large datasets and computational resources, and available pre-trained models are limited by fixed architecture and size. We propose a simple and effective approach for pre-training specifically for paraphrase generation, which significantly improves model quality and matches the quality level of general-purpose models. Both existing public data and new data generated by large language models were used. The impact of this procedure on neural networks of different architectures was investigated, and it was shown to work for all of them. Keywords: artificial intelligence, machine learning, neural networks, paraphrase generation, pre-training, fine-tuning.

Publisher

V.M. Glushkov Institute of Cybernetics

Reference27 articles.

1. 1. Han X., Zhang Z., Ding N., Gu Y., Liu X., Huo Y., Qiu J., Yao Y., Zhang A., Zhang L., et al. Pre-trained models: past, present and future. AI Open. 2021. Vol. 2. P. 225-250. https://doi.org/10.1016/j.aiopen.2021.08.002.

2. 2. Zhao W., Wang L., Shen K., Jia R., Liu J. Improving grammatical error correction via pre-training a copy-augmented architecture with unlabeled data. https://doi.org/10.48550/arXiv.1903.00138.

3. 3. Omelianchuk K., Atrasevych V., Chernodub A., Skurzhanskyi O. GECToR-Grammatical error correction: tag, not rewrite. arXiv:2005.12592v2 [cs.CL] 29 May 2020. https://doi.org/10.48550/arXiv.2005.12592.

4. 4. Kasai J., Pappas N., Peng H., Cross J., Smith N.A. Deep encoder, shallow decoder: reevaluating non-autoregressive machine translation. 2020. arXiv:2006.10369v4 [cs.CL]. 24 Jun 2021. https://doi.org/10.48550/arXiv.2006.10369.

5. 5. Wieting J., Gimpel K. ParaNMT-50M: pushing the limits of paraphrastic sentence embeddings with millions of machine translations. arXiv:1711.05732v2 [cs.CL] 20 Apr 2018. https://doi.org/10.48550/arXiv.1711.05732.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3