Addressee and Response Selection for Multilingual Conversation

TL;DR

This paper introduces multilingual addressee and response selection for conversational systems, utilizing knowledge transfer methods and a new dataset to improve performance across languages.

Contribution

It proposes novel knowledge transfer techniques for multilingual conversational response and addressee selection, supported by a new dataset and experimental validation.

Findings

Knowledge transfer improves multilingual response accuracy

New dataset enables effective evaluation across languages

Methods outperform baseline models in experiments

Abstract

Developing conversational systems that can converse in many languages is an interesting challenge for natural language processing. In this paper, we introduce multilingual addressee and response selection. In this task, a conversational system predicts an appropriate addressee and response for an input message in multiple languages. A key to developing such multilingual responding systems is how to utilize high-resource language data to compensate for low-resource language data. We present several knowledge transfer methods for conversational systems. To evaluate our methods, we create a new multilingual conversation dataset. Experiments on the dataset demonstrate the effectiveness of our methods.