DAMP: Doubly Aligned Multilingual Parser for Task-Oriented Dialogue
This addresses a critical issue for bilingual users in global markets by enhancing multilingual and codeswitched semantic parsing, though it is incremental as it builds on existing alignment methods.
The paper tackled the problem of low transfer efficiency in multilingual semantic parsing for task-oriented dialogue, particularly in bilingual contexts like India and Latin America, and improved zero-shot performance by 3x to 81x on benchmarks using a doubly aligned multilingual parser with fewer parameters.
Modern virtual assistants use internal semantic parsing engines to convert user utterances to actionable commands. However, prior work has demonstrated that semantic parsing is a difficult multilingual transfer task with low transfer efficiency compared to other tasks. In global markets such as India and Latin America, this is a critical issue as switching between languages is prevalent for bilingual users. In this work we dramatically improve the zero-shot performance of a multilingual and codeswitched semantic parsing system using two stages of multilingual alignment. First, we show that constrastive alignment pretraining improves both English performance and transfer efficiency. We then introduce a constrained optimization approach for hyperparameter-free adversarial alignment during finetuning. Our Doubly Aligned Multilingual Parser (DAMP) improves mBERT transfer performance by 3x, 6x, and 81x on the Spanglish, Hinglish and Multilingual Task Oriented Parsing benchmarks respectively and outperforms XLM-R and mT5-Large using 3.2x fewer parameters.