# Fine-grained Classification Suite (FLWR, DTD, PETS, CARS, UCF, CAL, FOOD, SUN, AIR, ESAT) (test) Source: [Fine-grained Classification Suite (FLWR, DTD, PETS, CARS, UCF, CAL, FOOD, SUN, AIR, ESAT) (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:15.818Z Evaluation of Fine-grained Classification performance across a suite of ten specific datasets (FLWR, DTD, PETS, CARS, UCF, CAL, FOOD, SUN, AIR, ESAT), typically using models pretrained on LAION2B. 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 | Fine-grained Classification | | Dataset (including any stated split) | Fine-grained Classification Suite (FLWR, DTD, PETS, CARS, UCF, CAL, FOOD, SUN, AIR, ESAT) LAION2B-pretrained backbone (test) | | Metric | Accuracy (FLWR) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | 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 (FLWR)](): Higher is better; unit: PERCENT. - [Accuracy (DTD)](): Higher is better; unit: PERCENT. - [Accuracy (PETS)](): Higher is better; unit: PERCENT. - [Accuracy (CARS)](): Higher is better; unit: PERCENT. - [Accuracy (UCF)](): Higher is better; unit: PERCENT. - [Accuracy (CAL)](): Higher is better; unit: PERCENT. - [Accuracy (FOOD)](): Higher is better; unit: PERCENT. - [Accuracy (SUN)](): Higher is better; unit: PERCENT. - [Accuracy (AIR)](): Higher is better; unit: PERCENT. - [Accuracy (ESAT)](): Higher is better; unit: PERCENT. - [Mean Accuracy](): Higher is better; unit: PERCENT. - [Median Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (FLWR) | Result date | Report paper | | --- | --- | --- | --- | | ZERO | 70.82 | Not recorded | 68fa75901e209131cea68a6d | | Zero-Shot | 69.71 | Not recorded | 68fa75901e209131cea68a6d | | TPT | 69.47 | Not recorded | 68fa75901e209131cea68a6d | | Ensemble | 68.7 | Not recorded | 68fa75901e209131cea68a6d | | ZERO+Ensemble | 68.01 | Not recorded | 68fa75901e209131cea68a6d | ### ZERO Accuracy (FLWR): 70.82. Result ID: 6993ee663a6c314849198f74. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"CLIP-ViT-B-16","Pre-training":"LAION-2B"}. This record does not assert independent reproduction. ### Zero-Shot Accuracy (FLWR): 69.71. Result ID: 6993ee663a6c314849198f71. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"CLIP-ViT-B-16","Pre-training":"LAION-2B"}. This record does not assert independent reproduction. ### TPT Accuracy (FLWR): 69.47. Result ID: 6993ee663a6c314849198f73. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"CLIP-ViT-B-16","Pre-training":"LAION-2B"}. This record does not assert independent reproduction. ### Ensemble Accuracy (FLWR): 68.7. Result ID: 6993ee663a6c314849198f72. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"CLIP-ViT-B-16","Pre-training":"LAION-2B"}. This record does not assert independent reproduction. ### ZERO+Ensemble Accuracy (FLWR): 68.01. Result ID: 6993ee663a6c314849198f75. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"CLIP-ViT-B-16","Pre-training":"LAION-2B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.