Active Learning for Multilingual Semantic Parser

TL;DR

This paper introduces AL-MSP, an active learning approach for multilingual semantic parsing that reduces translation costs by selecting the most informative examples for translation, improving performance with minimal data.

Contribution

It presents the first active learning method for MSP, including a novel selection strategy and hyperparameter tuning that avoids extra annotation costs.

Findings

AL-MSP significantly reduces translation costs.

The proposed selection method improves parsing performance.

Proper hyperparameters enhance the effectiveness of the selection strategy.

Abstract

Current multilingual semantic parsing (MSP) datasets are almost all collected by translating the utterances in the existing datasets from the resource-rich language to the target language. However, manual translation is costly. To reduce the translation effort, this paper proposes the first active learning procedure for MSP (AL-MSP). AL-MSP selects only a subset from the existing datasets to be translated. We also propose a novel selection method that prioritizes the examples diversifying the logical form structures with more lexical choices, and a novel hyperparameter tuning method that needs no extra annotation cost. Our experiments show that AL-MSP significantly reduces translation costs with ideal selection methods. Our selection method with proper hyperparameters yields better parsing performance than the other baselines on two multilingual datasets.