AI Paper Atlas

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

Semantic Equivalent Adversarial Data Augmentation for Visual Question Answering

Ruixue Tang, Chao Ma, Wei Emma Zhang, Qi Wu, Xiaokang Yang

ECCV 2020 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 56 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 semantic equivalent adversarial data augmentation (visual + question)BUTD · VQAv2 valaccuracy (%): 63.32 -> 65.16Improved
add question (paraphrase) augmentation (ablation)BUTD + Ours · VQAv2 valaccuracy (%): 64.69 -> 65.16Improved
add visual adversarial augmentation (ablation)BUTD + Ours · VQAv2 valaccuracy (%): 65.05 -> 65.16Improved
add EDA text-editing question augmentationBUTD · VQAv2 valaccuracy (%): 63.32 -> 62.73Hurt
jointly train with random visual noiseBUTD · VQAv2 valaccuracy (%): 63.32 -> 63.28No clear effect
inject adversarial examples only in epochs (10,15) instead of (5,25) (ablation)BUTD + Ours · VQAv2 valaccuracy (%): 63.93 -> 65.16Improved
inject adversarial examples only in epochs (10,15) instead of (15,20) (ablation)BUTD + Ours · VQAv2 valaccuracy (%): 64.18 -> 65.16Improved
use augmentation with 20% of training dataBUTD · VQAv2 val (20% train)accuracy (%): 55.45 -> 57.39Improved
use augmentation with 80% of training dataBUTD · VQAv2 val (80% train)accuracy (%): 62.77 -> 64.27Improved
adversarial training against IFGSM test attackBUTD · VQAv2 val (IFGSM attack)accuracy (%): 30.83 -> 44.57Improved
adversarial training against PGD test attackBUTD · VQAv2 val (PGD attack)accuracy (%): 18.05 -> 22.64Improved
adversarial training against paraphrase attackBUTD · VQAv2 val (paraphrase)accuracy (%): 54.03 -> 63.18Improved

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