# Image Classification on CIFAR-10 (test) (Symmetric/Asymmetric Scores) Source: [Image Classification on CIFAR-10 (test) (Symmetric/Asymmetric Scores)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:05:47.493Z Evaluation of Image Classification performance on the CIFAR-10 test set using a comprehensive set of Symmetric and Asymmetric scoring metrics across various thresholds. 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 (test) | | Metric | Symmetric Acc @ 20% | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Symmetric Acc @ 20%](): Higher is better; unit: PERCENT. - [Symmetric Acc @ 50%](): Higher is better; unit: PERCENT. - [Symmetric Acc @ 80%](): Higher is better; unit: PERCENT. - [Symmetric Acc @ 90%](): Higher is better; unit: PERCENT. - [Asymmetric Acc @ 10%](): Higher is better; unit: PERCENT. - [Asymmetric Acc @ 20%](): Higher is better; unit: PERCENT. - [Asymmetric Acc @ 30%](): Higher is better; unit: PERCENT. - [Asymmetric Acc @ 40%](): Higher is better; unit: PERCENT. ## Recorded results | Method | Symmetric Acc @ 20% | Result date | Report paper | | --- | --- | --- | --- | | Sel-CL+ | 95.5 | Not recorded | 68fa6bf21e209131cea31141 | | DivideMix | 95 | Not recorded | 68fa6bf21e209131cea31141 | | ELR (Uns-CL init.) | 94.4 | Not recorded | 68fa6bf21e209131cea31141 | | MOIT+ | 94.1 | Not recorded | 68fa6bf21e209131cea31141 | | M-correction | 93.8 | Not recorded | 68fa6bf21e209131cea31141 | | ELR | 93.8 | Not recorded | 68fa6bf21e209131cea31141 | | Mixup | 92.3 | Not recorded | 68fa6bf21e209131cea31141 | | P-correction | 92 | Not recorded | 68fa6bf21e209131cea31141 | | GCE (Uns-CL init.) | 90 | Not recorded | 68fa6bf21e209131cea31141 | | GCE | 86.6 | Not recorded | 68fa6bf21e209131cea31141 | | Forward | 83.1 | Not recorded | 68fa6bf21e209131cea31141 | | Cross-Entropy | 82.7 | Not recorded | 68fa6bf21e209131cea31141 | ### Sel-CL+ Symmetric Acc @ 20%: 95.5. Result ID: 698eab8cffc3901de09fa6b5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### DivideMix Symmetric Acc @ 20%: 95. Result ID: 698eab8cffc3901de09fa6b0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### ELR (Uns-CL init.) Symmetric Acc @ 20%: 94.4. Result ID: 698eab8cffc3901de09fa6b3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18","initialization":"Uns-CL"}. This record does not assert independent reproduction. ### MOIT+ Symmetric Acc @ 20%: 94.1. Result ID: 698eab8cffc3901de09fa6b4. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### M-correction Symmetric Acc @ 20%: 93.8. Result ID: 698eab8cffc3901de09fa6af. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### ELR Symmetric Acc @ 20%: 93.8. Result ID: 698eab8cffc3901de09fa6b1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### Mixup Symmetric Acc @ 20%: 92.3. Result ID: 698eab8cffc3901de09fa6ab. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### P-correction Symmetric Acc @ 20%: 92. Result ID: 698eab8cffc3901de09fa6ae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### GCE (Uns-CL init.) Symmetric Acc @ 20%: 90. Result ID: 698eab8cffc3901de09fa6b2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18","initialization":"Uns-CL"}. This record does not assert independent reproduction. ### GCE Symmetric Acc @ 20%: 86.6. Result ID: 698eab8cffc3901de09fa6ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### Forward Symmetric Acc @ 20%: 83.1. Result ID: 698eab8cffc3901de09fa6ac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. ### Cross-Entropy Symmetric Acc @ 20%: 82.7. Result ID: 698eab8cffc3901de09fa6aa. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"PreAct ResNet-18"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.