One Question Answering Model for Many Languages with Cross-lingual Dense Passage Retrieval
This addresses the challenge of providing accurate question answering in many languages, especially low-resource ones, without needing language-specific data or modules, though it builds incrementally on existing retrieval and generation techniques.
The authors tackled the problem of multilingual question answering by introducing CORA, a unified model that answers questions across 26 languages, including 9 unseen during training, and substantially outperforms previous state-of-the-art methods on benchmarks.
We present Cross-lingual Open-Retrieval Answer Generation (CORA), the first unified many-to-many question answering (QA) model that can answer questions across many languages, even for ones without language-specific annotated data or knowledge sources. We introduce a new dense passage retrieval algorithm that is trained to retrieve documents across languages for a question. Combined with a multilingual autoregressive generation model, CORA answers directly in the target language without any translation or in-language retrieval modules as used in prior work. We propose an iterative training method that automatically extends annotated data available only in high-resource languages to low-resource ones. Our results show that CORA substantially outperforms the previous state of the art on multilingual open QA benchmarks across 26 languages, 9 of which are unseen during training. Our analyses show the significance of cross-lingual retrieval and generation in many languages, particularly under low-resource settings.