Source-Critical Reinforcement Learning for Transferring Spoken Language Understanding to a New Language

TL;DR

This paper presents a novel reinforcement learning approach to improve language transfer for spoken language understanding systems, especially when only a small parallel corpus is available, significantly enhancing translation quality for SLU tasks.

Contribution

It introduces a source-critical reinforcement learning method to fine-tune translators for SLU, effectively handling semantic labels and cultural differences in low-resource language transfer.

Findings

Achieved over 97% slot F1 score in English SLU corpus

Attained over 84% accuracy in domain classification

Improved transfer performance by 22% in domain accuracy and 71% in slot filling F1 score

Abstract

To deploy a spoken language understanding (SLU) model to a new language, language transferring is desired to avoid the trouble of acquiring and labeling a new big SLU corpus. Translating the original SLU corpus into the target language is an attractive strategy. However, SLU corpora consist of plenty of semantic labels (slots), which general-purpose translators cannot handle well, not to mention additional culture differences. This paper focuses on the language transferring task given a tiny in-domain parallel SLU corpus. The in-domain parallel corpus can be used as the first adaptation on the general translator. But more importantly, we show how to use reinforcement learning (RL) to further finetune the adapted translator, where translated sentences with more proper slot tags receive higher rewards. We evaluate our approach on Chinese to English language transferring for SLU systems. The experimental results show that the generated English SLU corpus via adaptation and reinforcement learning gives us over 97% in the slot F1 score and over 84% accuracy in domain classification. It demonstrates the effectiveness of the proposed language transferring method. Compared with naive translation, our proposed method improves domain classification accuracy by relatively 22%, and the slot filling F1 score by relatively more than 71%.