# Class-Incremental Learning on ImageNet-100 B50 Inc25 Source: [Class-Incremental Learning on ImageNet-100 B50 Inc25]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:36:04.433Z Evaluation of Class-Incremental Learning performance on the ImageNet-100 subset defined by 50 base classes (B50) across 25 incremental steps (Inc25). 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 | Class-Incremental Learning | | Dataset (including any stated split) | ImageNet-100 B50 Inc25 | | Metric | Avg Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 14 of 14 | | 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 - [Avg Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Avg Accuracy | Result date | Report paper | | --- | --- | --- | --- | | DER w/ C-Flat | 87.96 | Not recorded | 68fa75751e209131cea680db | | FOSTER | 87.81 | Not recorded | 68fa75751e209131cea680db | | FOSTER w/ C-Flat | 87.81 | Not recorded | 68fa75751e209131cea680db | | DER | 87.1 | Not recorded | 68fa75751e209131cea680db | | PODNet w/ C-Flat | 86.85 | Not recorded | 68fa75751e209131cea680db | | PODNet | 85.95 | Not recorded | 68fa75751e209131cea680db | | WA w/ C-Flat | 83.84 | Not recorded | 68fa75751e209131cea680db | | MEMO w/ C-Flat | 83.41 | Not recorded | 68fa75751e209131cea680db | | MEMO | 83.09 | Not recorded | 68fa75751e209131cea680db | | WA | 80.81 | Not recorded | 68fa75751e209131cea680db | | iCaRL | 77.25 | Not recorded | 68fa75751e209131cea680db | | iCaRL w/ C-Flat | 76.22 | Not recorded | 68fa75751e209131cea680db | | Replay w/ C-Flat | 73.37 | Not recorded | 68fa75751e209131cea680db | | Replay | 72.18 | Not recorded | 68fa75751e209131cea680db | ### DER w/ C-Flat Avg Accuracy: 87.96. Result ID: 6993c2bf3a6c3148491855ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Expansion":true,"C-Flat":true}. This record does not assert independent reproduction. ### FOSTER Avg Accuracy: 87.81. Result ID: 6993c2bf3a6c3148491855eb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Expansion":true,"C-Flat":false}. This record does not assert independent reproduction. ### FOSTER w/ C-Flat Avg Accuracy: 87.81. Result ID: 6993c2bf3a6c3148491855ec. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Expansion":true,"C-Flat":true}. This record does not assert independent reproduction. ### DER Avg Accuracy: 87.1. Result ID: 6993c2bf3a6c3148491855e9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Expansion":true,"C-Flat":false}. This record does not assert independent reproduction. ### PODNet w/ C-Flat Avg Accuracy: 86.85. Result ID: 6993c2bf3a6c3148491855e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Regularization":true,"Memory":true,"C-Flat":true}. This record does not assert independent reproduction. ### PODNet Avg Accuracy: 85.95. Result ID: 6993c2bf3a6c3148491855e7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Regularization":true,"Memory":true,"C-Flat":false}. This record does not assert independent reproduction. ### WA w/ C-Flat Avg Accuracy: 83.84. Result ID: 6993c2bf3a6c3148491855e6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Regularization":true,"C-Flat":true}. This record does not assert independent reproduction. ### MEMO w/ C-Flat Avg Accuracy: 83.41. Result ID: 6993c2bf3a6c3148491855ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Expansion":true,"C-Flat":true}. This record does not assert independent reproduction. ### MEMO Avg Accuracy: 83.09. Result ID: 6993c2bf3a6c3148491855ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Expansion":true,"C-Flat":false}. This record does not assert independent reproduction. ### WA Avg Accuracy: 80.81. Result ID: 6993c2bf3a6c3148491855e5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Regularization":true,"C-Flat":false}. This record does not assert independent reproduction. ### iCaRL Avg Accuracy: 77.25. Result ID: 6993c2bf3a6c3148491855e3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Regularization":true,"Memory":true,"C-Flat":false}. This record does not assert independent reproduction. ### iCaRL w/ C-Flat Avg Accuracy: 76.22. Result ID: 6993c2bf3a6c3148491855e4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Regularization":true,"Memory":true,"C-Flat":true}. This record does not assert independent reproduction. ### Replay w/ C-Flat Avg Accuracy: 73.37. Result ID: 6993c2bf3a6c3148491855e2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Regularization":true,"Memory":true,"C-Flat":true}. This record does not assert independent reproduction. ### Replay Avg Accuracy: 72.18. Result ID: 6993c2bf3a6c3148491855e1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Regularization":true,"Memory":true,"C-Flat":false}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.