Research map / Label-efficient and robust learning / OOD and anomaly detection
CODEs: Chamfer Out-of-Distribution Examples Against Overconfidence Issue
Keke Tang, Dingruibo Miao, Weilong Peng, Jianpeng Wu, Yawen Shi, Zhaoquan Gu, Zhihong Tian, Wenping Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| transform seed examples with Chamfer GAN | seed OOD examples · CIFAR-10 | FID to original data: 50.49 -> 36.53 | Improved |
| transform seed examples with Chamfer GAN | seed OOD examples · CIFAR-100 | FID to original data: 48.83 -> 36.53 | Improved |
| transform seed examples with Chamfer GAN | seed OOD examples · SVHN | FID to original data: 83.51 -> 77.94 | Improved |
| transform seed examples with Chamfer GAN | seed OOD examples · CIFAR-10 | ResNet-18 MMC: 72.53 -> 74.84 | Improved |
| train with suppression on CODEs | baseline classifier · CIFAR-10 (train), OOD SVHN | OOD MMC (%): 73.38 -> 62.37 | Improved |
| train with suppression on CODEs | baseline classifier · CIFAR-10 (train), OOD CIFAR-100 | OOD MMC (%): 79.47 -> 70.14 | Improved |
| train with suppression on CODEs | baseline classifier · CIFAR-10 (train), noise | OOD MMC (%): 69.24 -> 10.27 | Improved |
| use Ours++ variant instead of Ours | CODEs training · CIFAR-10 (train), OOD SVHN | OOD MMC (%): 62.37 -> 47.18 | Improved |
| train with suppression on CODEs | baseline classifier · CIFAR-10 (train), CIFAR-10 test | test error (%): 5.38 -> 5.58 | No clear effect |
| train with suppression on CODEs | baseline classifier · CIFAR-100 (train), CIFAR-100 test | test error (%): 23.34 -> 23.22 | No clear effect |
| use auxiliary dataset (OE) instead of CODEs | CIFAR-10 classifier · CIFAR-10 (train), CIFAR-10 ID | ID MMC (%): 97.15 -> 88.87 | Hurt |
| train with suppression on CODEs | baseline classifier · CIFAR-10 (train), uniform noise | OOD MMC (%): 99.49 -> 65.30 | Mixed |
Closest papers by meaning
- Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests (NeurIPS 2026 · similarity 0.81)
- Simplification Is All You Need against Out-of-Distribution Overconfidence (CVPR 2025 · similarity 0.80)
- ConjNorm: Tractable Density Estimation for Out-of-Distribution Detection (ICLR 2024 · similarity 0.80)
- Tolerant Algorithms for Learning with Arbitrary Covariate Shift (NeurIPS 2024 · similarity 0.79)
- On the Learnability of Distribution Classes with Adaptive Adversaries (ICML 2025 · similarity 0.79)
- A Online Statistical Framework for Out-of-Distribution Detection (ICML 2025 · similarity 0.79)
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.
