Graph-based Filtering of Out-of-Vocabulary Words for Encoder-Decoder Models

TL;DR

This paper introduces a graph-based method using the HITS algorithm to select more effective vocabulary for encoder-decoder models, improving performance in machine translation and grammatical error correction.

Contribution

It presents a novel vocabulary selection technique that incorporates co-occurrence information, enhancing encoder-decoder model learning beyond simple frequency-based methods.

Findings

Improved BLEU score by 0.56 points in Japanese-English translation.

Enhanced F0.5-measure by 1.48 points in grammatical error correction.

Outperforms baseline vocabulary selection methods.

Abstract

Encoder-decoder models typically only employ words that are frequently used in the training corpus to reduce the computational costs and exclude noise. However, this vocabulary set may still include words that interfere with learning in encoder-decoder models. This paper proposes a method for selecting more suitable words for learning encoders by utilizing not only frequency, but also co-occurrence information, which we capture using the HITS algorithm. We apply our proposed method to two tasks: machine translation and grammatical error correction. For Japanese-to-English translation, this method achieves a BLEU score that is 0.56 points more than that of a baseline. It also outperforms the baseline method for English grammatical error correction, with an F0.5-measure that is 1.48 points higher.