evaluation-task · Source-linked discovery

DocVQA: A Dataset for VQA on Document Images

DocVQA is a Visual Question Answering benchmark that consists of 50,000 questions covering 12,000+ document images. This implementation solves and scores the "validation" split.

Open interactive record →
OriginMinesh Mathew, Dimosthenis Karatzas, C. V. JawaharTopicsMultimodalStatusimported

Can support

Not independently assessed by FronteraEval yet.

Cannot support by itself

No inference beyond the upstream source should be made until the protocol is reviewed.

Original sources