AI Paper Atlas

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

What You See is What You Read? Improving Text-Image Alignment Evaluation

Michal Yarom, Yonatan Bitton, Soravit Changpinyo, Roee Aharoni, Jonathan Herzig, Oran Lang, Eran Ofek, Idan Szpektor

NeurIPS 2023 · Vision-language and multimodal › Visual question answering · approach: Visual reasoning diagnostics · 14 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 synthetic training data (COCO t2i, COCO-Con, PickaPic-Con) to SNLI-VE fine-tuning (ablation)PaLI · SeeTRUE avgROC AUC: 79.7 -> 82.9Improved
add synthetic training data (ablation)PaLI · EditBenchROC AUC: 65.5 -> 77.2Improved
add synthetic training data (ablation)PaLI · COCO-ConROC AUC: 91.2 -> 91No clear effect
add synthetic training data (ablation)PaLI · SNLI-VEROC AUC: 95.1 -> 94.2Hurt
average VQ2 and fine-tuned PaLI scores (ablation)PaLI+Syn · SeeTRUE avgROC AUC: 82.9 -> 84.1Improved
yes-no predicate question (type C) instead of NLI comparison (type A) (ablation)VQ2 · EditBench meanROC AUC: 70.9 -> 75.8Improved
yes-no predicate (type C) instead of VQA answer comparison (type B) (ablation)VQ2 · EditBench meanROC AUC: 75.4 -> 75.8Improved
add multi-word answer candidates (ablation)VQ2 · EditBench meanROC AUC: 73.9 -> 75.8Improved
VQ2 reranking instead of CLIP rerankingVQ2 · DrawBench SD 1.4human-labeled quality: 77.4 -> 87.1Improved
VQ2 vs best prior on Winoground group scoreVQ2 · Winogroundgroup score: 16 -> 30.5Improved
VQ2 vs fine-tuned PaLI on COCO t2iVQ2 · COCO t2iROC AUC: 83.4 vs 83.2No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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