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: WebQA: Multihop and Multimodal QA

Abstract: Web search is fundamentally multimodal and multihop. Often, even before asking a question we choose to go directly to image search to find our answers. Further, rarely do we find an answer from a single source but aggregate information and reason through implications. Despite the frequency of this everyday occurrence, at present, there is no unified question answering benchmark that requires a single model to answer long-form natural language questions from text and open-ended visual sources -- akin to a human's experience. We propose to bridge this gap between the natural language and computer vision communities with WebQA. We show that A. our multihop text queries are difficult for a large-scale transformer model, and B. existing multi-modal transformers and visual representations do not perform well on open-domain visual queries. Our challenge for the community is to create a unified multimodal reasoning model that seamlessly transitions and reasons regardless of the source modality.
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Cite as: arXiv:2109.00590 [cs.CL]
  (or arXiv:2109.00590v3 [cs.CL] for this version)

Submission history

From: Yingshan Chang [view email]
[v1] Wed, 1 Sep 2021 19:43:59 GMT (8231kb,D)
[v2] Thu, 16 Sep 2021 20:18:55 GMT (10436kb,D)
[v3] Tue, 21 Sep 2021 20:56:01 GMT (8228kb,D)

Link back to: arXiv, form interface, contact.