We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Addressing Issues of Cross-Linguality in Open-Retrieval Question Answering Systems For Emergent Domains

Abstract: Open-retrieval question answering systems are generally trained and tested on large datasets in well-established domains. However, low-resource settings such as new and emerging domains would especially benefit from reliable question answering systems. Furthermore, multilingual and cross-lingual resources in emergent domains are scarce, leading to few or no such systems. In this paper, we demonstrate a cross-lingual open-retrieval question answering system for the emergent domain of COVID-19. Our system adopts a corpus of scientific articles to ensure that retrieved documents are reliable. To address the scarcity of cross-lingual training data in emergent domains, we present a method utilizing automatic translation, alignment, and filtering to produce English-to-all datasets. We show that a deep semantic retriever greatly benefits from training on our English-to-all data and significantly outperforms a BM25 baseline in the cross-lingual setting. We illustrate the capabilities of our system with examples and release all code necessary to train and deploy such a system.
Comments: 6 pages, 8 figures
Subjects: Computation and Language (cs.CL); Information Retrieval (cs.IR)
Cite as: arXiv:2201.11153 [cs.CL]
  (or arXiv:2201.11153v1 [cs.CL] for this version)

Submission history

From: Alon Albalak [view email]
[v1] Wed, 26 Jan 2022 19:27:32 GMT (2451kb,D)

Link back to: arXiv, form interface, contact.