AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Visual Reasoning by Progressive Module Networks

Seung Wook Kim, Makarand Tapaswi, Sanja Fidler

ICLR 2019 · Vision-language and multimodal › Human and embodied modeling · approach: Abstract visual reasoning benchmarks · 4 citations (OpenAlex)

Read the paper

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
Compose VQA with object and attribute modules (ablation)VQA baseline · VQA 2.0 valaccuracy: 62.05 -> 63.38Improved
Add relationship module to the VQA composition (ablation)PMN M_vqa1 · VQA 2.0 valaccuracy: 63.38 -> 63.64Improved
Add counting module to the VQA composition (ablation)PMN M_vqa1 · VQA 2.0 valaccuracy: 63.38 -> 64.06Improved
Add relationship and counting modules (ablation)PMN M_vqa1 · VQA 2.0 valaccuracy: 63.38 -> 64.36Improved
Add captioning module to the full VQA composition (ablation)PMN M_vqa4 · VQA 2.0 valaccuracy: 64.36 -> 64.68Improved
Replace learned soft gating with equal importance scores (ablation)PMN M_vqa5 · VQA 2.0 valaccuracy: 64.68% -> 63.65%Hurt
Increase capacity without composing modulesVQA baseline · VQA 2.0 valaccuracy: 62.05% -> 62.0%No clear effect
Add relationship-derived question-answer training data to M_vqa1VQA M_vqa1 · VQA 2.0 valaccuracy: 63.38% -> 63.05%Hurt
Compose relationship detection with object and attribute modules (ablation)Relationship detection · Visual Genome relationship validationObject accuracy (%): 51.0 -> 53.4Improved
Compose relationship detection with object and attribute modules (ablation)Relationship detection · Visual Genome relationship validationSubject accuracy (%): 55.9 -> 57.8Improved
Add object and attribute modules to counting (ablation)Counting baseline · Object countingaccuracy (%): 45.4 -> 47.4Improved
Add the relationship module to counting with object and attribute modules (ablation)Counting M_cnt1 · Object countingaccuracy (%): 47.4 -> 50.0Improved

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