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

The predominant approach to visual question answering (VQA) relies onencoding the image and question with a "black-box" neural encoder and decodinga single token as the answer like "yes" or "no". Despite this approach's strongquantitative results, it struggles to come up with intuitive, human-readableforms of justification for the prediction process. To address thisinsufficiency, we reformulate VQA as a full answer generation task, whichrequires the model to justify its predictions in natural language. We proposeLRTA [Look, Read, Think, Answer], a transparent neural-symbolic reasoningframework for visual question answering that solves the problem step-by-steplike humans and provides human-readable form of justification at each step.Specifically, LRTA learns to first convert an image into a scene graph andparse a question into multiple reasoning instructions. It then executes thereasoning instructions one at a time by traversing the scene graph using arecurrent neural-symbolic execution module. Finally, it generates a full answerto the given question with natural language justifications. Our experiments onGQA dataset show that LRTA outperforms the state-of-the-art model by a largemargin (43.1% v.s. 28.0%) on the full answer generation task. We also create aperturbed GQA test set by removing linguistic cues (attributes and relations)in the questions for analyzing whether a model is having a smart guess withsuperficial data correlations. We show that LRTA makes a step towards trulyunderstanding the question while the state-of-the-art model tends to learnsuperficial correlations from the training data.

LRTA: A Transparent Neural-Symbolic Reasoning Framework with Modular Supervision for Visual Question Answering