Cross-language Sentence Selection via Data Augmentation and Rationale Training

TL;DR

This paper introduces a novel cross-language sentence selection method using data augmentation and rationale training, achieving competitive results in low-resource settings by directly learning cross-lingual relevance models from noisy parallel data.

Contribution

It presents a new approach combining data augmentation, negative sampling, and rationale training to improve cross-language sentence selection in low-resource scenarios.

Findings

Outperforms state-of-the-art translation + monolingual retrieval systems.

Consistent improvements across English-Somali, English-Swahili, and English-Tagalog.

Effective in low-resource language pairs.

Abstract

This paper proposes an approach to cross-language sentence selection in a low-resource setting. It uses data augmentation and negative sampling techniques on noisy parallel sentence data to directly learn a cross-lingual embedding-based query relevance model. Results show that this approach performs as well as or better than multiple state-of-the-art machine translation + monolingual retrieval systems trained on the same parallel data. Moreover, when a rationale training secondary objective is applied to encourage the model to match word alignment hints from a phrase-based statistical machine translation model, consistent improvements are seen across three language pairs (English-Somali, English-Swahili and English-Tagalog) over a variety of state-of-the-art baselines.