AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Inferring and Executing Programs for Visual Reasoning

Justin Johnson, Bharath Hariharan, Laurens van der Maaten, Judy Hoffman, Li Fei-Fei, C. Lawrence Zitnick, Ross Girshick

ICCV 2017 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 503 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
use all 700K ground-truth programs instead of 18K (ablation)program generator + execution engine · CLEVR (overall)accuracy: 95.4 -> 96.9Improved
use 18K ground-truth programs instead of 9K (ablation)semi-supervised model · CLEVR (overall)accuracy: 88.6 -> 95.4Improved
replace the linear transform with an MLP (ablation)CNN+LSTM+SA · CLEVR (overall)accuracy: 69.8 -> 73.2Improved
train the execution engine on predicted rather than ground-truth programs (ablation)execution engine · CLEVR (overall)accuracy: loss of about 3 pointsHurt
finetune jointly on short and long questions without programs (ablation)Ours (25K prog.) · CLEVR long questionsaccuracy: 55.3 -> 77.8Improved
finetune jointly on short and long questions (ablation)CNN+LSTM+SA+MLP · CLEVR long questionsaccuracy: 64.3 -> 67.8Improved
train on short questions only versus baseline (ablation)Ours (25K prog.) · CLEVR long questionsaccuracy: 55.3 vs 64.3 for CNN+LSTM+SA+MLPHurt
finetune the program generator on CLEVR-Humans (ablation)Ours (18K prog.) · CLEVR-Humans testaccuracy: 54.0 -> 66.6Improved
finetune on CLEVR-Humans (ablation)CNN+LSTM+SA+MLP · CLEVR-Humans testaccuracy: 50.4 -> 57.6Improved
remove Batch Normalization from modules (ablation)execution engine modules · CLEVR (overall)accuracyImproved

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