# Class-Incremental Learning on Stanford Cars (T=10, test) Source: [Class-Incremental Learning on Stanford Cars (T=10, test)]() Collection index: [llms.txt]() Data updated: 2026-05-12T07:48:26.534Z Evaluation of Class-Incremental Learning performance on the Stanford Cars dataset, configured for 10 sequential tasks, evaluated on the test split. 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) | Stanford Cars CIL, T=10 (test) | | Metric | Avg Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 34 of 34 | | 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. - [Rel. Error Rate](): Lower is better; unit: PERCENT. - [Last Accuracy](): Higher is better; unit: PERCENT. - [FF](): Higher is better; unit: SCALAR. ## Recorded results | Method | Avg Accuracy | Result date | Report paper | | --- | --- | --- | --- | | RanPAC | 77.5 | Not recorded | 68fa71cb1e209131cea53052 | | CPrompt | 76.81 | Not recorded | 68fa72851e209131cea56fec | | CODAprompt | 73.28 | Not recorded | 68fa72851e209131cea56fec | | RanPAC (No RPs) | 73 | Not recorded | 68fa71cb1e209131cea53052 | | ESN | 72.82 | Not recorded | 68fa72851e209131cea56fec | | L2P | 71.92 | Not recorded | 68fa72851e209131cea56fec | | TRM | 70.4 | Not recorded | 6a02b3f7dbc836b87d82458d | | DualPrompt | 70.36 | Not recorded | 68fa72851e209131cea56fec | | LwF | 69.84 | Not recorded | 68fa7bb31e209131cea89fe1 | | CMM | 69.7 | Not recorded | 68fa7bb31e209131cea89fe1 | | CLG-CBM | 68.7 | Not recorded | 6a02b3f7dbc836b87d82458d | | CODAPrompt | 67.4 | Not recorded | 6a02b3f7dbc836b87d82458d | | DualPrompt | 67.1 | Not recorded | 6a02b3f7dbc836b87d82458d | | L2P | 66.7 | Not recorded | 6a02b3f7dbc836b87d82458d | | TIES | 66.1 | Not recorded | 6a02b3f7dbc836b87d82458d | | RanPAC (No Phase 1) | 66 | Not recorded | 68fa71cb1e209131cea53052 | | MagMax | 65.9 | Not recorded | 6a02b3f7dbc836b87d82458d | | RandMix | 65.62 | Not recorded | 68fa7bb31e209131cea89fe1 | | Avg | 65.59 | Not recorded | 68fa7bb31e209131cea89fe1 | | TIES | 64.9 | Not recorded | 68fa7bb31e209131cea89fe1 | | Model Stock | 64.5 | Not recorded | 6a02b3f7dbc836b87d82458d | | MaxAbs | 63.55 | Not recorded | 68fa7bb31e209131cea89fe1 | | RanPAC (No RPs or Phase 1) | 62.2 | Not recorded | 68fa71cb1e209131cea53052 | | LwF | 61.7 | Not recorded | 6a02b3f7dbc836b87d82458d | | EWC | 61.3 | Not recorded | 6a02b3f7dbc836b87d82458d | | EWC | 60.78 | Not recorded | 68fa7bb31e209131cea89fe1 | | Joint linear probe | 51.7 | Not recorded | 68fa71cb1e209131cea53052 | | CODA-Prompt | 43.2 | Not recorded | 68fa71cb1e209131cea53052 | | ADaM | 41.4 | Not recorded | 68fa71cb1e209131cea53052 | | RanPAC (NCM with Phase 1) | 40.5 | Not recorded | 68fa71cb1e209131cea53052 | | DualPrompt | 40.1 | Not recorded | 68fa71cb1e209131cea53052 | | L2P | 38.2 | Not recorded | 68fa71cb1e209131cea53052 | | RanPAC (NCM only) | 37.7 | Not recorded | 68fa71cb1e209131cea53052 | ### RanPAC Avg Accuracy: 77.5. Result ID: 6991fc683a6c3148490bdc3b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Algorithm":1}. This record does not assert independent reproduction. ### CPrompt Avg Accuracy: 76.81. Result ID: 69925a633a6c3148490e6192. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### CODAprompt Avg Accuracy: 73.28. Result ID: 69925a633a6c3148490e6191. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RanPAC (No RPs) Avg Accuracy: 73. Result ID: 6991fc683a6c3148490bdc3c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"RP layer":"false"}. This record does not assert independent reproduction. ### ESN Avg Accuracy: 72.82. Result ID: 69925a633a6c3148490e6190. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### L2P Avg Accuracy: 71.92. Result ID: 69925a633a6c3148490e618e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### TRM Avg Accuracy: 70.4. Result ID: 6a02db3b47f34ce924e5619f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"CLIP ViT-B/16","Method Category":"Model Merging"}. This record does not assert independent reproduction. ### DualPrompt Avg Accuracy: 70.36. Result ID: 69925a633a6c3148490e618f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LwF Avg Accuracy: 69.84. Result ID: 699d011a5698665ec4a13788. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory":"Theta(1)"}. This record does not assert independent reproduction. ### CMM Avg Accuracy: 69.7. Result ID: 699d011a5698665ec4a1378a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory":"Theta(1)"}. This record does not assert independent reproduction. ### CLG-CBM Avg Accuracy: 68.7. Result ID: 6a02db3b47f34ce924e5619b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2025","Backbone":"CLIP ViT-B/16","Method Category":"PEFT"}. This record does not assert independent reproduction. ### CODAPrompt Avg Accuracy: 67.4. Result ID: 6a02db3b47f34ce924e5619a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2023","Backbone":"CLIP ViT-B/16","Method Category":"PEFT"}. This record does not assert independent reproduction. ### DualPrompt Avg Accuracy: 67.1. Result ID: 6a02db3b47f34ce924e56199. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2023","Backbone":"CLIP ViT-B/16","Method Category":"PEFT"}. This record does not assert independent reproduction. ### L2P Avg Accuracy: 66.7. Result ID: 6a02db3b47f34ce924e56198. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Year":"2022","Backbone":"CLIP ViT-B/16","Method Category":"PEFT"}. This record does not assert independent reproduction. ### TIES Avg Accuracy: 66.1. Result ID: 6a02db3b47f34ce924e5619d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2023","Backbone":"CLIP ViT-B/16","Method Category":"Model Merging"}. This record does not assert independent reproduction. ### RanPAC (No Phase 1) Avg Accuracy: 66. Result ID: 6991fc683a6c3148490bdc3d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Phase 1":"false"}. This record does not assert independent reproduction. ### MagMax Avg Accuracy: 65.9. Result ID: 6a02db3b47f34ce924e5619e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2024","Backbone":"CLIP ViT-B/16","Method Category":"Model Merging"}. This record does not assert independent reproduction. ### RandMix Avg Accuracy: 65.62. Result ID: 699d011a5698665ec4a13784. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory":"Theta(N)"}. This record does not assert independent reproduction. ### Avg Avg Accuracy: 65.59. Result ID: 699d011a5698665ec4a13786. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Memory":"Theta(N)"}. This record does not assert independent reproduction. ### TIES Avg Accuracy: 64.9. Result ID: 699d011a5698665ec4a13787. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory":"Theta(N)"}. This record does not assert independent reproduction. ### Model Stock Avg Accuracy: 64.5. Result ID: 6a02db3b47f34ce924e5619c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2024","Backbone":"CLIP ViT-B/16","Method Category":"Model Merging"}. This record does not assert independent reproduction. ### MaxAbs Avg Accuracy: 63.55. Result ID: 699d011a5698665ec4a13785. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory":"Theta(N)"}. This record does not assert independent reproduction. ### RanPAC (No RPs or Phase 1) Avg Accuracy: 62.2. Result ID: 6991fc683a6c3148490bdc3e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"RP layer":"false","Phase 1":"false"}. This record does not assert independent reproduction. ### LwF Avg Accuracy: 61.7. Result ID: 6a02db3b47f34ce924e56196. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2017","Backbone":"CLIP ViT-B/16","Method Category":"Conventional"}. This record does not assert independent reproduction. ### EWC Avg Accuracy: 61.3. Result ID: 6a02db3b47f34ce924e56197. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Year":"2017","Backbone":"CLIP ViT-B/16","Method Category":"Conventional"}. This record does not assert independent reproduction. ### EWC Avg Accuracy: 60.78. Result ID: 699d011a5698665ec4a13789. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory":"Theta(1)"}. This record does not assert independent reproduction. ### Joint linear probe Avg Accuracy: 51.7. Result ID: 6991fc683a6c3148490bdc36. [Reporting paper]() Recorded evaluation configuration: {"Joint training":"true"}. This record does not assert independent reproduction. ### CODA-Prompt Avg Accuracy: 43.2. Result ID: 6991fc683a6c3148490bdc39. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ADaM Avg Accuracy: 41.4. Result ID: 6991fc683a6c3148490bdc3a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RanPAC (NCM with Phase 1) Avg Accuracy: 40.5. Result ID: 6991fc683a6c3148490bdc3f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation":"NCM","Phase 1":"true"}. This record does not assert independent reproduction. ### DualPrompt Avg Accuracy: 40.1. Result ID: 6991fc683a6c3148490bdc38. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### L2P Avg Accuracy: 38.2. Result ID: 6991fc683a6c3148490bdc37. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RanPAC (NCM only) Avg Accuracy: 37.7. Result ID: 6991fc683a6c3148490bdc40. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation":"NCM","Phase 1":"false"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.