String Transduction with Target Language Models and Insertion Handling

TL;DR

This paper demonstrates that combining target language models from unannotated corpora with precise data alignment significantly improves performance on various character-level sequence transduction tasks.

Contribution

It introduces a novel approach that leverages target language models and alignment techniques to enhance sequence-to-sequence transduction accuracy.

Findings

Achieved state-of-the-art results on cognate projection

Improved inflection generation performance

Enhanced phoneme-to-grapheme conversion accuracy

Abstract

Many character-level tasks can be framed as sequence-to-sequence transduction, where the target is a word from a natural language. We show that leveraging target language models derived from unannotated target corpora, combined with a precise alignment of the training data, yields state-of-the art results on cognate projection, inflection generation, and phoneme-to-grapheme conversion.