AI Paper Atlas

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

Rethinking Data Augmentation for Robust Visual Question Answering

Long Chen, Yuhang Zheng, Jun Xiao

ECCV 2022 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 66 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 KDDAug augmentationUpDn · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 48.91 -> 61.52Improved
add KDDAug augmentationUpDn · VQA-CP v2 testaccuracy: 39.74 -> 60.24Improved
add KDDAug augmentationUpDn · VQA v2 valaccuracy: 63.48 -> 62.86Hurt
add KDDAug augmentationLMH · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 57.24 -> 60.79Improved
add KDDAug augmentationRUBi · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 49.66 -> 59.75Improved
add KDDAug augmentationCSS+ · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 59.05 -> 61.65Improved
assign answers by KDDAug instead of SimpleAug on the same pairs (ablation)UpDn+augmentation · VQA-CP v2 / VQA v2 val (HM)harmonic mean: over 8% on HMImproved
use dynamic multi-teacher weights instead of OOD teacher only (ablation)KDDAug · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 59.91 -> 61.52Improved
use dynamic multi-teacher weights instead of ID teacher only (ablation)KDDAug · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 51.15 -> 61.52Improved
use dynamic multi-teacher weights instead of simple average (ablation)KDDAug · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 57.79 -> 61.52Improved
add rule-based initial answers (ablation)KDDAug · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 58.26 -> 61.52Improved
use more CLIP-filtered augmented samples (10% to 100%) (ablation)KDDAug · VQA-CP v2 / VQA v2 val (HM)harmonic mean: 60.85 -> 61.52Improved

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