# Image Classification on CIFAR-10 (Accuracy Retention Evaluation) Source: [Image Classification on CIFAR-10 (Accuracy Retention Evaluation)]() Collection index: [llms.txt]() Data updated: 2026-06-11T09:26:04.468Z Evaluation of standard Image Classification performance on CIFAR-10, focusing specifically on the 'Accuracy retained (%)' metric, often used to measure performance stability under compression or perturbation. 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 | | Metric | Accuracy Retained | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 Retained](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy Retained | Result date | Report paper | | --- | --- | --- | --- | | Baseline | 100 | Not recorded | 6a2a40efd33dfb23e78391bc | | Energy Penalty | 100 | Not recorded | 6a2a40efd33dfb23e78391bc | | Conservation | 100 | Not recorded | 6a2a40efd33dfb23e78391bc | | Conservation | 93.2 | Not recorded | 6a2a40efd33dfb23e78391bc | | Conservation | 77.4 | Not recorded | 6a2a40efd33dfb23e78391bc | | Baseline | 58.3 | Not recorded | 6a2a40efd33dfb23e78391bc | | Energy Penalty | 51.4 | Not recorded | 6a2a40efd33dfb23e78391bc | | Conservation | 42.8 | Not recorded | 6a2a40efd33dfb23e78391bc | | Baseline | 35.1 | Not recorded | 6a2a40efd33dfb23e78391bc | | Energy Penalty | 30.9 | Not recorded | 6a2a40efd33dfb23e78391bc | | Conservation | 26.1 | Not recorded | 6a2a40efd33dfb23e78391bc | | Baseline | 19.9 | Not recorded | 6a2a40efd33dfb23e78391bc | | Energy Penalty | 18.7 | Not recorded | 6a2a40efd33dfb23e78391bc | | Conservation | 18.6 | Not recorded | 6a2a40efd33dfb23e78391bc | | Baseline | 16.4 | Not recorded | 6a2a40efd33dfb23e78391bc | | Energy Penalty | 15.6 | Not recorded | 6a2a40efd33dfb23e78391bc | | Baseline | 14.6 | Not recorded | 6a2a40efd33dfb23e78391bc | | Energy Penalty | 14.5 | Not recorded | 6a2a40efd33dfb23e78391bc | ### Baseline Accuracy Retained: 100. Result ID: 6a2a7ec7fbc373592c27ef75. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Energy Penalty Accuracy Retained: 100. Result ID: 6a2a7ec7fbc373592c27ef76. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Conservation Accuracy Retained: 100. Result ID: 6a2a7ec7fbc373592c27ef77. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Noise σ":"0.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Conservation Accuracy Retained: 93.2. Result ID: 6a2a7ec7fbc373592c27ef7a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Noise σ":"0.1","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Conservation Accuracy Retained: 77.4. Result ID: 6a2a7ec7fbc373592c27ef7d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Noise σ":"0.2","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Baseline Accuracy Retained: 58.3. Result ID: 6a2a7ec7fbc373592c27ef78. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.1","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Energy Penalty Accuracy Retained: 51.4. Result ID: 6a2a7ec7fbc373592c27ef79. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.1","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Conservation Accuracy Retained: 42.8. Result ID: 6a2a7ec7fbc373592c27ef80. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Noise σ":"0.5","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Baseline Accuracy Retained: 35.1. Result ID: 6a2a7ec7fbc373592c27ef7b. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.2","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Energy Penalty Accuracy Retained: 30.9. Result ID: 6a2a7ec7fbc373592c27ef7c. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.2","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Conservation Accuracy Retained: 26.1. Result ID: 6a2a7ec7fbc373592c27ef83. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Noise σ":"1.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Baseline Accuracy Retained: 19.9. Result ID: 6a2a7ec7fbc373592c27ef7e. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.5","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Energy Penalty Accuracy Retained: 18.7. Result ID: 6a2a7ec7fbc373592c27ef7f. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"0.5","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Conservation Accuracy Retained: 18.6. Result ID: 6a2a7ec7fbc373592c27ef86. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Noise σ":"2.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Baseline Accuracy Retained: 16.4. Result ID: 6a2a7ec7fbc373592c27ef81. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"1.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Energy Penalty Accuracy Retained: 15.6. Result ID: 6a2a7ec7fbc373592c27ef82. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"1.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Baseline Accuracy Retained: 14.6. Result ID: 6a2a7ec7fbc373592c27ef84. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"2.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. ### Energy Penalty Accuracy Retained: 14.5. Result ID: 6a2a7ec7fbc373592c27ef85. [Reporting paper]() Recorded evaluation configuration: {"Noise σ":"2.0","Pipeline depth":"2","Number of random seeds":"5"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.