AI Paper Atlas

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

Toward Unsupervised Realistic Visual Question Answering

Yuwei Zhang, Chih-Hui Ho, Nuno Vasconcelos

ICCV 2023 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 4 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
train with pseudo UQs from randomly paired images and questionsBUTD/UNITER/LXMERT · RGQAAUAF: 36.50 -> 47.02 (BUTD, MSP vs RP)Improved
add RoI Mixup to pseudo UQ training (ablation)RP · RGQAAUAF: 47.02 -> 47.60 (BUTD)Improved
ensemble RP and Mix models (ablation)Mix · RGQAAUAF: 47.60 -> 48.28 (BUTD)Improved
use post-hoc ODIN, Maha or Energy instead of MSPVQA classifiers · RGQAAUAF: 36.50 -> 36.50 (BUTD, ODIN)No clear effect
train only on hard pseudo UQs instead of random pseudo UQs (ablation)RP · RGQAAUAF: 47.02 -> 40.2 (BUTD)Hurt
replace random pseudo UQ sampling with Resample adversarial weighting (ablation)RP · RGQAAUAF: 47.02 -> 38.07 (BUTD)Hurt
use larger and more recent VQA modelsMSP RVQA · RGQAAUAF: 36.50 (BUTD) -> 44.89 (VinVL)No clear effect
use integrated architecture instead of separated or branched (ablation)RP with LXMERT · RGQAAUAFImproved
train with pseudo UQsLXMERT · GQA testaccuracy: 77.8 -> 76.3 (RP), 77.1 (Mix)No clear effect

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