SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.08212 · ACL Findings · 2021

On the Copying Behaviors of Pre-Training for Neural Machine Translation

Shuming Shi, Liang Ding, Longyue Wang, Xuebo Liu, Derek Wong, Zhaopeng Tu, Lidia Chao, Nlp

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. If it links a repository, it is listed below.

Abstract

Previous studies have shown that initializing neural machine translation (NMT) models with the pre-trained language models (LM) can speed up the model training and boost the model performance. In this work, we identify a critical side-effect of pre-training for NMT, which is due to the discrepancy between the training objectives of LM-based pre-training and NMT. Since the LM objective learns to reconstruct a few source tokens and copy most of them, the pre-training initialization would affect the copying behaviors of NMT models. We provide a quantitative analysis of copying behaviors by introducing a metric called copying ratio, which empirically shows that pre-training based NMT models have a larger copying ratio than the standard one. In response to this problem, we propose a simple and effective method named copying penalty to control the copying behaviors in decoding. Extensive experiments on both indomain and out-of-domain benchmarks show that the copying penalty method consistently improves translation performance by controlling copying behaviors for pre-training based NMT models. Source code is freely available at https://github.com/SunbowLiu/ CopyingPenalty.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2107.08212")
get_code_for_paper("2107.08212")
have("2107.08212")

Connect an agent — have() is free.