# Image Classification on CIFAR-10 Imbalance Ratio 10, Heteroskedastic Noise (val) Source: [Image Classification on CIFAR-10 Imbalance Ratio 10, Heteroskedastic Noise (val)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:06:54.023Z Evaluation of Image Classification performance on the validation split of CIFAR-10 configured with an Imbalance Ratio of 10 and subjected to Heteroskedastic Noise. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Image Classification | | Dataset (including any stated split) | CIFAR-10 Imbalance Ratio 10, Heteroskedastic Noise (val) | | Metric | Top-1 Acc (Noisy&Rare Cls.) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Top-1 Acc (Noisy&Rare Cls.)](): Higher is better; unit: PERCENT. - [Top-1 Acc (Clean Cls.)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Top-1 Acc (Noisy&Rare Cls.) | Result date | Report paper | | --- | --- | --- | --- | | HAR | 63.5 | Not recorded | 68fa69941e209131cea22d32 | | MW-Net | 55.4 | Not recorded | 68fa69941e209131cea22d32 | | MentorNet | 54.1 | Not recorded | 68fa69941e209131cea22d32 | | Unif-reg (optimal) | 53.9 | Not recorded | 68fa69941e209131cea22d32 | | ERM | 52.9 | Not recorded | 68fa69941e209131cea22d32 | | INCV | 48.9 | Not recorded | 68fa69941e209131cea22d32 | | GCE | 48.2 | Not recorded | 68fa69941e209131cea22d32 | | DMI | 44.7 | Not recorded | 68fa69941e209131cea22d32 | | L2RW | 44.3 | Not recorded | 68fa69941e209131cea22d32 | | Co-teaching | 30.2 | Not recorded | 68fa69941e209131cea22d32 | ### HAR Top-1 Acc (Noisy&Rare Cls.): 63.5. Result ID: 698dbb4e8b517eb433236314. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### MW-Net Top-1 Acc (Noisy&Rare Cls.): 55.4. Result ID: 698dbb4e8b517eb433236310. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### MentorNet Top-1 Acc (Noisy&Rare Cls.): 54.1. Result ID: 698dbb4e8b517eb43323630e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### Unif-reg (optimal) Top-1 Acc (Noisy&Rare Cls.): 53.9. Result ID: 698dbb4e8b517eb433236313. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### ERM Top-1 Acc (Noisy&Rare Cls.): 52.9. Result ID: 698dbb4e8b517eb43323630b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### INCV Top-1 Acc (Noisy&Rare Cls.): 48.9. Result ID: 698dbb4e8b517eb43323630d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### GCE Top-1 Acc (Noisy&Rare Cls.): 48.2. Result ID: 698dbb4e8b517eb433236311. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### DMI Top-1 Acc (Noisy&Rare Cls.): 44.7. Result ID: 698dbb4e8b517eb433236312. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### L2RW Top-1 Acc (Noisy&Rare Cls.): 44.3. Result ID: 698dbb4e8b517eb43323630f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. ### Co-teaching Top-1 Acc (Noisy&Rare Cls.): 30.2. Result ID: 698dbb4e8b517eb43323630c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.