AI Paper Atlas

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

Supervising the Transfer of Reasoning Patterns in VQA

Corentin Kervadec, Christian Wolf, Grigory Antipov, Moez Baccouche, Madiha Nadri

NeurIPS 2021 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 5 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 program supervision to oracle transferVL-Transformer · GQA test-devaccuracy: 57.0{ $\pm0.3$} -> 57.8{ $\pm0.2$}Improved
Add program supervision to oracle transferVL-Transformer · GQA-OOD testacc-tail: 48.2{ $\pm0.3$} -> 48.8}{ $\pm0.1$Improved
Add program supervision to oracle transfer with LXMERT pre-trainingVL-Transformer · GQA-OOD testacc-tail: 47.1 -> 48.0}{ $\pm0.6$Improved
Add full program supervision, especially visual-argument supervision (ablation)Compact VL-Transformer · GQA val.accuracy: 62.2 -> 66.2Improved
Add full program supervision, especially visual-argument supervision (ablation)Compact VL-Transformer · GQA-OOD val.acc-tail: 46.9 -> 49.9Improved
Add visual-argument supervision to the other program losses (ablation)Compact VL-Transformer · GQA val.accuracy: 63.7 -> 66.2Improved
Add visual-argument supervision to the other program losses (ablation)Compact VL-Transformer · GQA-OOD val.acc-tail: 47.3 -> 49.9Improved
Predict programs after cross-modal rather than uni-modal layers (ablation)Oracle transfer VL-Transformer · GQA val.accuracy: 66.5 -> 67.4Improved
Replace ground-truth program labels with programs from other questions (ablation)Compact VL-Transformer · GQA val.accuracy: 66.2 -> 61.4Hurt
Increase object proposals from 36 to 100 per imageOracle transfer + program supervision · GQA test-devaccuracy: 57.0{ $\pm0.4$} -> 58.2}{ $\pm0.1$Improved
Use VinVL instead of RCNN visual featuresOracle transfer + program supervision · GQA test-devaccuracy: 58.2}{ $\pm0.1$ -> 60.9}{ $\pm0.2$Improved

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