# Image Classification on ImageNet (Accuracy and Calibration Metrics) Source: [Image Classification on ImageNet (Accuracy and Calibration Metrics)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:29:22.038Z Evaluation of Image Classification performance on the standard ImageNet dataset, measuring standard accuracy alongside key model calibration and uncertainty metrics (ECE, AUPR, AUC, FPR95). 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 | | Metric | Accuracy | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [ECE](): Lower is better; unit: PERCENT. - [AUPR](): Higher is better; unit: PERCENT. - [AUC](): Higher is better; unit: PERCENT. - [FPR@95](): Lower is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Deep Ensembles | 0.817 | Not recorded | 68fa71ad1e209131cea527f6 | | Laplace | 0.81 | Not recorded | 68fa71ad1e209131cea527f6 | | ABNN | 0.806 | Not recorded | 68fa71ad1e209131cea527f6 | | Laplace | 0.804 | Not recorded | 68fa71ad1e209131cea527f6 | | Single Model | 0.8 | Not recorded | 68fa71ad1e209131cea527f6 | | ABNN | 0.795 | Not recorded | 68fa71ad1e209131cea527f6 | | Deep Ensembles | 0.792 | Not recorded | 68fa71ad1e209131cea527f6 | | Single Model | 0.778 | Not recorded | 68fa71ad1e209131cea527f6 | | MIMO (P = 1) | 0.776 | Not recorded | 68fa71ad1e209131cea527f6 | | BatchEnsemble | 0.759 | Not recorded | 68fa71ad1e209131cea527f6 | ### Deep Ensembles Accuracy: 0.817. Result ID: 6991db4d3a6c3148490b07a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Laplace Accuracy: 0.81. Result ID: 6991db4d3a6c3148490b07a2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### ABNN Accuracy: 0.806. Result ID: 6991db4d3a6c3148490b07a3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Laplace Accuracy: 0.804. Result ID: 6991db4d3a6c3148490b079e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### Single Model Accuracy: 0.8. Result ID: 6991db4d3a6c3148490b07a0. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### ABNN Accuracy: 0.795. Result ID: 6991db4d3a6c3148490b079f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### Deep Ensembles Accuracy: 0.792. Result ID: 6991db4d3a6c3148490b079d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### Single Model Accuracy: 0.778. Result ID: 6991db4d3a6c3148490b079a. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### MIMO (P = 1) Accuracy: 0.776. Result ID: 6991db4d3a6c3148490b079c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### BatchEnsemble Accuracy: 0.759. Result ID: 6991db4d3a6c3148490b079b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.