# Image Classification on CIFAR-100 (test) (k-Accuracy Profile) Source: [Image Classification on CIFAR-100 (test) (k-Accuracy Profile)]() Collection index: [llms.txt]() Data updated: 2026-06-05T08:13:09.640Z Evaluation of Image Classification performance on the CIFAR-100 test set, reporting accuracy across multiple k-nearest neighbor thresholds (k=20 to k=90) and a reference accuracy. 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-100 (test) | | Metric | Accuracy (k=80) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 20 of 20 | | 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 - [Accuracy (k=80)](): Higher is better; unit: PERCENT. - [Accuracy (k=20)](): Higher is better; unit: PERCENT. - [Accuracy (k=30)](): Higher is better; unit: PERCENT. - [Accuracy (k=40)](): Higher is better; unit: PERCENT. - [Accuracy (k=50)](): Higher is better; unit: PERCENT. - [Accuracy (k=70)](): Higher is better; unit: PERCENT. - [Accuracy (k=90)](): Higher is better; unit: PERCENT. - [Reference Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (k=80) | Result date | Report paper | | --- | --- | --- | --- | | Fully Supervised | 64.8 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Semi-AUM+Cutoff | 63.1 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Semi-DUAL+Beta | 62.9 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | ELFS (DINO) | 61.9 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | ZCore | 61.9 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | ELFS (Self-Encoder) | 59.5 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Random | 59.2 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Score Extrapolation | 53.8 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Jo-SNC-SEM | 44.26 | Not recorded | 68fa7bc01e209131cea8a30a | | Jo-SNC | 43.45 | Not recorded | 68fa7bc01e209131cea8a30a | | Prototypicality | 39.1 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | CA2C-SEM | 36.03 | Not recorded | 68fa7bc01e209131cea8a30a | | CA2C | 33.56 | Not recorded | 68fa7bc01e209131cea8a30a | | UNICON | 31.49 | Not recorded | 68fa7bc01e209131cea8a30a | | Jo-SRC | 23.8 | Not recorded | 68fa7bc01e209131cea8a30a | | JoCoR | 15.49 | Not recorded | 68fa7bc01e209131cea8a30a | | Co-teaching | 15.15 | Not recorded | 68fa7bc01e209131cea8a30a | ### Fully Supervised Accuracy (k=80): 64.8. Result ID: 6a140290fbc373592c0aff5e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With Labels"}. This record does not assert independent reproduction. ### Semi-AUM+Cutoff Accuracy (k=80): 63.1. Result ID: 6a140290fbc373592c0aff65. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With 10% Labels"}. This record does not assert independent reproduction. ### Semi-DUAL+Beta Accuracy (k=80): 62.9. Result ID: 6a140290fbc373592c0aff66. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With 10% Labels"}. This record does not assert independent reproduction. ### ELFS (DINO) Accuracy (k=80): 61.9. Result ID: 6a140290fbc373592c0aff61. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### ZCore Accuracy (k=80): 61.9. Result ID: 6a140290fbc373592c0aff63. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### ELFS (Self-Encoder) Accuracy (k=80): 59.5. Result ID: 6a140290fbc373592c0aff62. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### Random Accuracy (k=80): 59.2. Result ID: 6a140290fbc373592c0aff5f. [Reporting paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### Score Extrapolation Accuracy (k=80): 53.8. Result ID: 6a140290fbc373592c0aff64. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With 10% Labels"}. This record does not assert independent reproduction. ### Jo-SNC-SEM Accuracy (k=80): 44.26. Result ID: 6a228515fbc373592c207055. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Publication":"Ours"}. This record does not assert independent reproduction. ### Jo-SNC Accuracy (k=80): 43.45. Result ID: 6a228515fbc373592c207053. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Publication":"IEEE TPAMI 2025"}. This record does not assert independent reproduction. ### Prototypicality Accuracy (k=80): 39.1. Result ID: 6a140290fbc373592c0aff60. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### CA2C-SEM Accuracy (k=80): 36.03. Result ID: 6a228515fbc373592c207054. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Publication":"Ours"}. This record does not assert independent reproduction. ### CA2C Accuracy (k=80): 33.56. Result ID: 6a228515fbc373592c207052. [Reporting paper]() Recorded evaluation configuration: {"Publication":"ICCV 2025"}. This record does not assert independent reproduction. ### UNICON Accuracy (k=80): 31.49. Result ID: 6a228515fbc373592c207051. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Publication":"CVPR 2022"}. This record does not assert independent reproduction. ### Jo-SRC Accuracy (k=80): 23.8. Result ID: 6a228515fbc373592c207050. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Publication":"CVPR 2021"}. This record does not assert independent reproduction. ### JoCoR Accuracy (k=80): 15.49. Result ID: 6a228515fbc373592c20704f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Publication":"CVPR 2020"}. This record does not assert independent reproduction. ### Co-teaching Accuracy (k=80): 15.15. Result ID: 6a228515fbc373592c20704e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Publication":"NeurIPS 2018"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.