English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We study the new problem of automatic question generation (QG) frommulti-modal sources containing images and texts, significantly expanding thescope of most of the existing work that focuses exclusively on QG from onlytextual sources. We propose a simple solution for our new problem, calledMultiQG-TI, which enables a text-only question generator to process visualinput in addition to textual input. Specifically, we leverage an image-to-textmodel and an optical character recognition model to obtain the textualdescription of the image and extract any texts in the image, respectively, andthen feed them together with the input texts to the question generator. We onlyfine-tune the question generator while keeping the other components fixed. Onthe challenging ScienceQA dataset, we demonstrate that MultiQG-TI significantlyoutperforms ChatGPT with few-shot prompting, despite having hundred-times lesstrainable parameters. Additional analyses empirically confirm the necessity ofboth visual and textual signals for QG and show the impact of various modelingchoices.

MultiQG-TI: Towards Question Generation from Multi-modal Sources