AI Paper Atlas

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

Revisiting Visual Question Answering Baselines

Allan Jabri, Armand Joulin, Laurens van der Maaten

ECCV 2016 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 238 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 image input to answer+question (ablation)MLP · Visual7W Tellingaccuracy (%): 58.5 -> 67.1Improved
add question input to answer+image (ablation)MLP · Visual7W Tellingaccuracy (%): 63.8 -> 67.1Improved
add image and question to answer-only model (ablation)MLP · Visual7W Tellingaccuracy (%): 52.9 -> 67.1Improved
treat the answer as an input with binary loss instead of softmax over answers (ablation)MLP · Visual7W Tellingaccuracy (%): 52.2 -> 67.1Improved
treat the answer as an input with binary loss instead of softmax (ablation)MLP · VQA test2015-devaccuracy (%): 61.1 -> 64.9Improved
binary loss instead of softmax (ablation)Linear · Visual7W Tellingaccuracy (%): 42.6 -> 44.7Improved
use ResNet-101 features instead of AlexNet (ablation)MLP · Visual7W Tellingaccuracy (%): 63.5 -> 67.1Improved
replace BoW with LSTM text encoder (ablation)MLP · Visual7W Telling (no image)accuracy (%): 58.5 -> 52.9Hurt
replace BoW with LSTM text encoder (ablation)Bilinear · Visual7W Telling (no image)accuracy (%): 52.6 -> 54.3Improved
train on VQA and transfer directly instead of training on Visual7WMLP (A+Q+I) · Visual7W Tellingaccuracy (%): 67.1 -> 53.8Hurt
add finetuning on Visual7W after VQA transferMLP (A+Q+I) · Visual7W Tellingaccuracy (%): 67.1 -> 68.5Improved
use MLP over bilinear model (ablation)answer-as-input classifier · Visual7W Tellingaccuracy (%): 63.6 -> 67.1Improved

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