Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Counterfactual Zero-Shot and Open-Set Visual Recognition
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use sample-specific counterfactual generation with counterfactual-faithful training as the stage-one classifier | Generalized ZSL with TF-VAEGAN stage two · AWA2 | H: 64.6 -> 67.0 | Helped |
| Use sample-specific counterfactual generation with counterfactual-faithful training as the stage-one classifier | Generalized ZSL with TF-VAEGAN stage two · CUB, SUN, aPY | H: 50.0 -> 52.2; 40.8 -> 42.2; 43.7 -> 44.9 | Helped |
| Use GCM-CF for open-set seen/unknown classification | OSR · MNIST, SVHN, CIFAR10, C+10, C+50 | F1: 88.95 -> 91.37; 77.95 -> 79.25; 71.38 -> 72.63; 78.68 -> 79.38; 70.96 -> 74.60 | Helped |
| Add counterfactual-faithful training to disentangle sample attribute Z and class attribute Y (ablation) | GCM-CF · CUB, AWA2, SUN, aPY | H: 25.9 -> 60.3; 40.0 -> 67.0; 27.4 -> 42.2; 21.5 -> 44.9 | Helped |
| Add counterfactual-faithful training to disentangle sample attribute Z and class attribute Y (ablation) | GCM-CF · MNIST, SVHN, CIFAR10, C+10, C+50 | F1: 91.37 -> 94.21; 62.57 -> 79.23; 67.03 -> 73.03; 73.81 -> 80.29; 69.18 -> 74.70 | Helped |
| Use the proposed stage-one GCM-CF instead of TF-VAEGAN | two-stage generalized ZSL with RelationNet stage two · AWA2 | H: 61.3 -> 64.0 | Helped |
| Use the proposed stage-one GCM-CF instead of TF-VAEGAN | two-stage generalized ZSL with CADA-VAE stage two · AWA2 | H: 61.5 -> 65.2 | Helped |
| Use the proposed stage-one GCM-CF instead of TF-VAEGAN | two-stage generalized ZSL with LisGAN stage two · AWA2 | H: 60.7 -> 63.9 | Helped |
Closest papers by meaning
- Few-Shot Open-Set Recognition by Transformation Consistency (CVPR 2021 · similarity 0.87)
- All Beings Are Equal in Open Set Recognition (AAAI 2024 · similarity 0.85)
- Representation-Level Counterfactual Calibration for Debiased Zero-Shot Recognition (NeurIPS 2025 · similarity 0.85)
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 2025 · similarity 0.84)
- Robust Visual Recognition with Class-Imbalanced Open-World Noisy Data (AAAI 2024 · similarity 0.84)
- From Coarse to Fine-Grained Open-Set Recognition (CVPR 2024 · similarity 0.84)
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.