# Image Classification on ImageNet-1K (test) (Confidence Threshold Accuracy) Source: [Image Classification on ImageNet-1K (test) (Confidence Threshold Accuracy)]() Collection index: [llms.txt]() Data updated: 2026-05-25T08:04:54.586Z Evaluation of Image Classification performance on the test split of the ImageNet-1K dataset, reporting accuracy measured at various confidence thresholds (30%, 50%, 70%, 80%, 90%). 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) | ImageNet-1K (test) | | Metric | Accuracy (30% Threshold) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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 (30% Threshold)](): Higher is better; unit: PERCENT. - [Accuracy (50% Threshold)](): Higher is better; unit: PERCENT. - [Accuracy (70% Threshold)](): Higher is better; unit: PERCENT. - [Accuracy (80% Threshold)](): Higher is better; unit: PERCENT. - [Accuracy (90% Threshold)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (30% Threshold) | Result date | Report paper | | --- | --- | --- | --- | | ELFS (DINO) | 73.5 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Fully Supervised | 73.3 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Semi-AUM+Cutoff | 73.3 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | ELFS (Self-Encoder) | 73.2 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Semi-DUAL+Beta | 72.5 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | ZCore | 72.3 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Random | 72.2 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Prototypicality | 70.9 | Not recorded | 6a13d7926f71c5ae7dc9e94e | | Score Extrapolation | 68.9 | Not recorded | 6a13d7926f71c5ae7dc9e94e | ### ELFS (DINO) Accuracy (30% Threshold): 73.5. Result ID: 6a140290fbc373592c0aff6a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### Fully Supervised Accuracy (30% Threshold): 73.3. Result ID: 6a140290fbc373592c0aff67. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With Labels"}. This record does not assert independent reproduction. ### Semi-AUM+Cutoff Accuracy (30% Threshold): 73.3. Result ID: 6a140290fbc373592c0aff6e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With 10% Labels"}. This record does not assert independent reproduction. ### ELFS (Self-Encoder) Accuracy (30% Threshold): 73.2. Result ID: 6a140290fbc373592c0aff6b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### Semi-DUAL+Beta Accuracy (30% Threshold): 72.5. Result ID: 6a140290fbc373592c0aff6f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With 10% Labels"}. This record does not assert independent reproduction. ### ZCore Accuracy (30% Threshold): 72.3. Result ID: 6a140290fbc373592c0aff6c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### Random Accuracy (30% Threshold): 72.2. Result ID: 6a140290fbc373592c0aff68. [Reporting paper]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### Prototypicality Accuracy (30% Threshold): 70.9. Result ID: 6a140290fbc373592c0aff69. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Label Usage":"Without Labels"}. This record does not assert independent reproduction. ### Score Extrapolation Accuracy (30% Threshold): 68.9. Result ID: 6a140290fbc373592c0aff6d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Label Usage":"With 10% Labels"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.