# Prompt Recovery on DiffusionDB Source: [Prompt Recovery on DiffusionDB]() Collection index: [llms.txt]() Data updated: 2026-03-06T07:08:18.012Z Evaluation of the model's ability to recover the original text prompt used to generate images within the DiffusionDB dataset, assessed using metrics such as CLIP similarity, perceptual distance (LPIPS), and text generation quality (PPL, Precision/Recall/F1). 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 | Prompt Recovery | | Dataset (including any stated split) | DiffusionDB | | Metric | CLIP Score | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 6 of 6 | | 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 - [CLIP Score](): Higher is better; unit: SCALAR. - [LPIPS](): Lower is better; unit: SCALAR. - [Precision](): Higher is better; unit: PERCENT. - [Recall](): Higher is better; unit: PERCENT. - [F1 Score](): Higher is better; unit: PERCENT. - [PPL](): Lower is better; unit: SCALAR. ## Recorded results | Method | CLIP Score | Result date | Report paper | | --- | --- | --- | --- | | PRISM | 0.822 | Not recorded | 68fa791f1e209131cea7c381 | | EDITOR | 0.807 | Not recorded | 68fa791f1e209131cea7c381 | | VGD | 0.78 | Not recorded | 68fa791f1e209131cea7c381 | | PEZ | 0.766 | Not recorded | 68fa791f1e209131cea7c381 | | STEPS | 0.749 | Not recorded | 68fa791f1e209131cea7c381 | | PH2P | 0.742 | Not recorded | 68fa791f1e209131cea7c381 | ### PRISM CLIP Score: 0.822. Result ID: 69aa7c5b6dbb7075b953baad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### EDITOR CLIP Score: 0.807. Result ID: 69aa7c5b6dbb7075b953baae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### VGD CLIP Score: 0.78. Result ID: 69aa7c5b6dbb7075b953baab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### PEZ CLIP Score: 0.766. Result ID: 69aa7c5b6dbb7075b953baa9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### STEPS CLIP Score: 0.749. Result ID: 69aa7c5b6dbb7075b953baac. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### PH2P CLIP Score: 0.742. Result ID: 69aa7c5b6dbb7075b953baaa. [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.