AI Paper Atlas

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

Leveraging Visual Question Answering for Image-Caption Ranking

Xiao Lin, Devi Parikh

ECCV 2016 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 82 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 VQA-grounded score-level fusion to the VQA-agnostic modelKiros et al. model · MSCOCO 1K testcaption retrieval R@1: 43.4 -> 46.9Improved
add VQA-grounded score-level fusion to the VQA-agnostic modelKiros et al. model · MSCOCO 1K testimage retrieval R@1: 31.0 -> 35.8Improved
add VQA representations (N=3000) to deeper representation-level fusion model (ablation)deeper VQA-agnostic model · MSCOCO 1K testcaption retrieval R@1: 45.8 -> 50.5Improved
add VQA representations (N=3000) to deeper representation-level fusion model (ablation)deeper VQA-agnostic model · MSCOCO 1K testimage retrieval R@1: 33.6 -> 37.0Improved
add an additional layer to the VQA-agnostic model (ablation)Kiros et al. model · MSCOCO 1K testcaption retrieval R@1: 43.4 -> 45.8Improved
add VQA representation for the caption only (ablation)deeper VQA-agnostic model · MSCOCO 1K testcaption retrieval R@1: 45.8 -> 47.3Improved
add VQA representation for the image only (ablation)deeper VQA-agnostic model · MSCOCO 1K testimage retrieval R@1: 33.6 -> 36.4Improved
add VQA representation for the image to caption-only fusion (ablation)caption-only fusion model · MSCOCO 1K testimage retrieval R@1: 35.5 -> 37.0Improved
use VQA hidden activations z instead of VQA-grounded activations u (score-level fusion) (ablation)score-level fusion VQA-aware · MSCOCO 1K testcaption retrieval R@1: 46.9 -> 46.8No clear effect
use VQA hidden activations z instead of VQA-grounded activations u (representation-level fusion) (ablation)representation-level fusion VQA-aware · MSCOCO 1K testcaption retrieval R@1: 50.5 -> 49.3No clear effect
increase number of question-answer pairs N from 30 to 300 and beyond (ablation)score-level and representation-level fusion · MSCOCO 1K testretrieval recallMixed
sample 3 QA pairs per image on 1000 images instead of 1 QA pair per image on 3000 images (ablation)N=3000 score-level fusion · MSCOCO 1K testimage/caption R@1: 46.9 -> 47.7No 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