# Class-conditional image generation on ImageNet-1k 100 fine-grained class pairs Source: [Class-conditional image generation on ImageNet-1k 100 fine-grained class pairs]() Collection index: [llms.txt]() Data updated: 2026-07-09T06:38:40.651Z Evaluation of Class-Conditional Image Generation models using the specialized 100 fine-grained class pairs subset of the ImageNet-1K dataset, focusing on classification accuracy and generation quality metrics like FD-DINOv2. 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-conditional image generation | | Dataset (including any stated split) | ImageNet-1k 100 fine-grained class pairs | | Metric | Positive Accuracy | | 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 - [Positive Accuracy](): Higher is better; unit: PERCENT. - [Negative Accuracy](): Higher is better; unit: PERCENT. - [Overall Accuracy](): Higher is better; unit: PERCENT. - [Aesthetic Score](): Higher is better; unit: SCALAR. - [FD-DINOv2](): Lower is better; unit: SCALAR. ## Recorded results | Method | Positive Accuracy | Result date | Report paper | | --- | --- | --- | --- | | CCFG | 91.01 | Not recorded | 68fa75e31e209131cea6a7a0 | | Perp-Neg | 85.44 | Not recorded | 68fa75e31e209131cea6a7a0 | | DNG | 83.07 | Not recorded | 68fa75e31e209131cea6a7a0 | | SLD | 77.85 | Not recorded | 68fa75e31e209131cea6a7a0 | | NP | 49.59 | Not recorded | 68fa75e31e209131cea6a7a0 | ### CCFG Positive Accuracy: 91.01. Result ID: 6a4f41e360b88bb339e41a99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Perp-Neg Positive Accuracy: 85.44. Result ID: 6a4f41e360b88bb339e41a97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DNG Positive Accuracy: 83.07. Result ID: 6a4f41e360b88bb339e41a96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SLD Positive Accuracy: 77.85. Result ID: 6a4f41e360b88bb339e41a98. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### NP Positive Accuracy: 49.59. Result ID: 6a4f41e360b88bb339e41a95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.