AI Paper Atlas

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

Why Does a Visual Question Have Different Answers?

Nilavra Bhattacharya, Qing Li, Danna Gurari

ICCV 2019 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 61 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 predicted answers (A) as input cue to question and image (ablation)Q+I model · VizWizmean average precision: 45.73 -> 50.02Improved
add predicted answers (A) as input cue to question and image (ablation)Q+I model · VQA_2.0mean average precision: 43.16 -> 44.55Improved
add predicted answers (A) as input cue, spam category excluded (ablation)Q+I model · VQA_2.0 (spam excluded)mean average precision: 47.87 -> 48.05No clear effect
add predicted answers (A) as input cue, spam category excluded (ablation)Q+I model · VizWiz (spam excluded)mean average precision: 50.59 -> 55.18Improved
add image input to question input (ablation)Q model · VizWizmean average precision: 40.5 -> 45.73Improved
add image input to question input (ablation)Q model · VQA_2.0mean average precision: 43.47 -> 43.16No clear effect
fine-tune the VQA model on the answer-difference datasets (ablation)Q+I+A · VizWizmean average precision: 50.02 -> 50.01No clear effect
fine-tune the VQA model on the answer-difference datasets (ablation)Q+I+A · VQA_2.0mean average precision: 44.55 -> 44.46No clear effect
replace predicted answers with ground truth soft-target answers (ablation)Q+I+A · VizWizmean average precision: 50.02 -> 50.68No clear effect
replace predicted answers with ground truth soft-target answers (ablation)Q+I+A · VQA_2.0mean average precision: 44.55 -> 44.09No clear effect
use proposed Q+I+A model instead of QI-relevance baselineQI-Relevance · VizWizmean average precision: 31.71 -> 50.02Improved
use proposed Q+I+A model instead of QI-relevance baselineQI-Relevance · VQA_2.0mean average precision: 32.23 -> 44.55Improved

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