AI Paper Atlas

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

Unshuffling Data for Improved Generalization in Visual Question Answering

Damien Teney, Ehsan Abbasnejad, Anton van den Hengel

ICCV 2021 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA · 122 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 variance regularizer across environment classifiers (ablation)Multi-environment training · VQA-CP v2 test (Other)accuracy: 39.14 -> 47.60Improved
use question-group environments instead of random environments (ablation)Environment training · VQA-CP v2 test (Other)accuracy: 43.51 -> 47.60Improved
use K-means clustered question environments (ablation)Environment training · VQA-CP v2 test (Other)accuracy: 43.33 -> 46.35Improved
use question-group instead of clustered environments (ablation)Environment training · VQA-CP v2 test (Other)accuracy: 46.35 -> 47.60Improved
use relative variance instead of absolute variance (ablation)Variance regularizer · VQA-CP v2 test (Other)accuracy: 46.17 -> 47.60Improved
add alternating optimization with 4 warm-up epochs (ablation)Multi-environment training · VQA-CP v2 test (Other)accuracy: 47.60 -> 48.06Improved
add alternating optimization with 6 warm-up epochs (ablation)Multi-environment training · VQA-CP v2 test (Other)accuracy: 47.60 -> 47.38No clear effect
replace random with environment trainingEnvironments · VQA-CP v2 val (Other)accuracy: 54.74 -> 53.87Mixed
apply method to full VQA-CPBottom-up top-down · VQA-CP v2 test overallaccuracy: 37.87 -> 42.39Improved
apply method to VQA v2 standard splitBottom-up top-down · VQA v2 valoverall accuracy: 61.09 -> 61.08No clear effect
combine method with 4-model ensembleBottom-up top-down · VQA-CP v2 test overallaccuracy: 39.30 -> 43.37Improved
treat VQA v2 and VG as two environments rather than aggregatingBottom-up top-down · VQA v2 valoverall accuracy: 63.32 -> 63.80Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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