Research map / Label-efficient and robust learning / Recognition and long-tail
Classification-Reconstruction Learning for Open-Set 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 classification-reconstruction learning and the joint prediction-plus-latent representation for detection | CROSR with Plain CNN · CIFAR-10, ImageNet-crop; ImageNet-resize; LSUN-crop; LSUN-resize | open-set score: 0.660 -> 0.721; 0.684 -> 0.735; 0.657 -> 0.720; 0.668 -> 0.749 | Helped |
| use classification-reconstruction learning with latent representations | CROSR · DBpedia, 4/14; 4/12; 4/8 | F1-score: 0.532 -> 0.582; 0.574 -> 0.627; 0.729 -> 0.765 | Helped |
| use CROSR instead of Counterfactual | CROSR · SVHN | AUC: 0.910 $\pm$ 0.010 -> 0.899 $\pm$ 0.018 | Hurt |
| use CROSR instead of Counterfactual | CROSR · MNIST and TinyImageNet | AUC: 0.988 $\pm$ 0.004 -> 0.991 $\pm$ 0.004; 0.586 -> 0.589 | Helped |
| add reconstruction-loss regularization (ablation) | LadderNet · MNIST-vs-Omniglot and MNIST-vs-MNIST-Noise | F1-score: 8.4%; 10.1% | Helped |
| add only the reconstruction-error term without explicitly using reconstructive factors (ablation) | natural-image open-set classifiers · CIFAR-10 | open-set performance | No clear effect |
| replace the ℓ² distance detector with OCSVM (ablation) | DHRNet · MNIST, Omniglot | F1-score: 0.793 -> 0.702 | Hurt |
| replace the ℓ² distance detector with OCSVM (ablation) | DHRNet · MNIST, Noise and MNIST-noise | F1-score: 0.826 -> 0.979; 0.827 -> 0.976 | Helped |
Closest papers by meaning
- All Beings Are Equal in Open Set Recognition (AAAI 2024 · similarity 0.86)
- Unlocking Better Closed-Set Alignment Based on Neural Collapse for Open-Set Recognition (AAAI 2025 · similarity 0.86)
- Exploring Diverse Representations for Open Set Recognition (AAAI 2024 · similarity 0.85)
- From Coarse to Fine-Grained Open-Set Recognition (CVPR 2024 · similarity 0.85)
- Frequency Shuffling and Enhancement for Open Set Recognition (AAAI 2024 · similarity 0.84)
- Meta Evidential Transformer for Few-Shot Open-Set Recognition (ICML 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.