ZusammenQA: Data Augmentation with Specialized Models for Cross-lingual Open-retrieval Question Answering System

TL;DR

This paper presents ZusammenQA, a cross-lingual open-retrieval question answering system that leverages specialized models, data augmentation, and multilingual training to improve performance across multiple languages, especially low-resource ones.

Contribution

It introduces a novel combination of data augmentation, specialized passage retrieval, and answer generation models tailored for cross-lingual QA tasks.

Findings

Data augmentation improves low-resource language performance.

Specialized models enhance retrieval and answer quality.

Language- and domain-specific training benefits overall accuracy.

Abstract

This paper introduces our proposed system for the MIA Shared Task on Cross-lingual Open-retrieval Question Answering (COQA). In this challenging scenario, given an input question the system has to gather evidence documents from a multilingual pool and generate from them an answer in the language of the question. We devised several approaches combining different model variants for three main components: Data Augmentation, Passage Retrieval, and Answer Generation. For passage retrieval, we evaluated the monolingual BM25 ranker against the ensemble of re-rankers based on multilingual pretrained language models (PLMs) and also variants of the shared task baseline, re-training it from scratch using a recently introduced contrastive loss that maintains a strong gradient signal throughout training by means of mixed negative samples. For answer generation, we focused on language- and domain-specialization by means of continued language model (LM) pretraining of existing multilingual encoders. Additionally, for both passage retrieval and answer generation, we augmented the training data provided by the task organizers with automatically generated question-answer pairs created from Wikipedia passages to mitigate the issue of data scarcity, particularly for the low-resource languages for which no training data were provided. Our results show that language- and domain-specialization as well as data augmentation help, especially for low-resource languages.