AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

Explanation vs Attention: A Two-Player Game to Obtain Attention for VQA

Badri Patro, Anupriy, Vinay Namboodiri

AAAI 2020 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 28 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add pixel-wise adversarial attention using Grad-CAM supervisionSAN · HATRC: 0.2432 -> 0.3071Improved
Add pixel-wise adversarial attention using Grad-CAM supervisionSAN · VQA-v1 testAll accuracy: 58.7 -> 63.6Improved
Add PAAN attention supervisionMCB · VQA-v1 testAll accuracy: 64.2 -> 67.1Improved
Add PAAN attention supervisionMCB · VQA-v2 testAll accuracy: 64.0 -> 68.4Improved
Use pixel-wise local discrimination rather than global adversarial discrimination (ablation)AAN · HATRC: 0.2896 -> 0.3071Improved
Replace adversarial distribution matching with MMD (ablation)PAAN · HATRC: 0.3071 -> 0.2573Hurt
Replace adversarial distribution matching with CORAL (ablation)PAAN · HATRC: 0.3071 -> 0.2563Hurt
Replace adversarial distribution matching with MSE (ablation)PAAN · HATRC: 0.3071 -> 0.2681Hurt
Use RISE instead of Grad-CAM as explanation supervision (ablation)PAAN · HATRC: 0.3071 -> 0.2754Hurt
Use random explanation masks with 7% overlap instead of Grad-CAM (ablation)PAAN · HATRC: 0.3071 -> 0.1213Hurt
Increase random explanation-mask overlap from 7% to 20% (ablation)PAAN · HATRC: 0.1213 -> 0.1746Improved
Use pixel-wise PAAN rather than global AAN (ablation)SAN · VQA-v1 test-devAll accuracy: 62.3 -> 63.6Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy