trends.barnas.net

task · first seen 11 Oct 2026

document visual question answering

Also: DocVQA, document VQA, visual document question answering

Answering questions about document images, requiring the model to read and understand text and layout.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

retrieval-augmented generationadversarial attack on vision-language models