AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Schema Inference for Interpretable Image Classification

Haofei Zhang, Mengqi Xue, Xiaokang Liu, Kaixuan Chen, Jie Song, Mingli Song

ICLR 2023 · Graph learning › Knowledge and temporal graphs · approach: Graph reasoning for visual recognition · 1 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
Initialize IR-Atlas by averaging instance IR-Graphs and sparsify (remove low-weight vertices/edges) (ablation)SchemaNet with DeiT-Tiny · CIFAR-100 / Caltech-101Top-1 accuracy: +1.34 (CIFAR-100), +4.33 (Caltech-101)Improved
Initialize IR-Atlas by averaging instance IR-Graphs and sparsify (ablation)SchemaNet with DeiT-Tiny · CIFAR-10Top-1 accuracy: +0.04Improved
Use raw attention for ingredient importance instead of CAM (ablation)SchemaNet-Init with DeiT-Tiny · CIFAR-10Top-1 accuracy: +2.38Improved
Use raw attention for ingredient importance instead of ViT-LRP (ablation)SchemaNet-Init with DeiT-Tiny · CIFAR-10Top-1 accuracy: +4.68Improved
Use raw attention for ingredient importance instead of CAM (ablation)SchemaNet-Init with DeiT-Tiny · CIFAR-100Top-1 accuracy: +0.90Improved
Replace SchemaNet-Init with BagNet-style summation (BagNet baseline) on DeiT-TinyVisual word matching on DeiT-Tiny · CIFAR-10Top-1 accuracy: -25.69 vs SchemaNet-InitHurt
SchemaNet-Init (graph matching with interacting ingredients) vs Backbone-FCDeiT-Tiny · CIFAR-10 / CIFAR-100 / Caltech-101Top-1 accuracy: +1.38 (CIFAR-10), +1.37 (CIFAR-100), +10.88 (Caltech-101)Improved
Graph-based matching (SchemaNet) vs bag-of-visual-words with discretized words (BoVW-Deep) on DeiT-TinyVisual word interactions · CIFAR-100 / Caltech-101Top-1 accuracy: +1.66 (CIFAR-100), +22.71 (Caltech-101) using SchemaNet vs BoVW-DeepImproved
Adding SchemaNet-Init graph matching vs frozen-backbone FC on ImageNet-1kDeiT-Small · ImageNet-1kTop-1 accuracy: +4.82Improved
Graph matching (SchemaNet-Init) vs BagNet on adversarial benchmarksDeiT-Small trained on ImageNet-1k · ImageNet-A / ImageNet-RTop-1 accuracy: +7.52 (ImageNet-A), +2.53 (ImageNet-R)Improved
Replace SchemaNet with ViT-LRP attribution cost trade-off (ablation)Ingredient importance computation · CIFAR-10running time: 53.8h vs 0.8hHurt

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