# Concept Erasure on MS-COCO (5 Specific Concepts) Source: [Concept Erasure on MS-COCO (5 Specific Concepts)]() Collection index: [llms.txt]() Data updated: 2026-03-04T08:08:05.669Z Evaluation of Concept Erasure performance across five specific concepts derived from MS-COCO prompts (Snoopy, Mickey, Spongebob, Pikachu, Hello Kitty), reporting individual Concept Score (CS) and FID metrics for each concept and the overall dataset. 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 | Concept Erasure | | Dataset (including any stated split) | MS-COCO Concept prompts (Snoopy, Mickey, Spongebob, Pikachu, Hello Kitty) | | Metric | Snoopy CS | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 7 of 7 | | 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 - [Snoopy CS](): Higher is better; unit: SCALAR. - [Snoopy FID](): Lower is better; unit: SCALAR. - [Mickey CS](): Higher is better; unit: SCALAR. - [Mickey FID](): Lower is better; unit: SCALAR. - [Spongebob CS](): Higher is better; unit: SCALAR. - [Spongebob FID](): Lower is better; unit: SCALAR. - [Pikachu CS](): Higher is better; unit: SCALAR. - [Pikachu FID](): Lower is better; unit: SCALAR. - [Hello Kitty CS](): Higher is better; unit: SCALAR. - [Hello Kitty FID](): Lower is better; unit: SCALAR. - [MS-COCO CS](): Higher is better; unit: SCALAR. - [MS-COCO FID](): Lower is better; unit: SCALAR. ## Recorded results | Method | Snoopy CS | Result date | Report paper | | --- | --- | --- | --- | | ConAbl | 25.26 | Not recorded | 68fa7c231e209131cea8ca66 | | UCE | 23.6 | Not recorded | 68fa7c231e209131cea8ca66 | | SPEED | 23.58 | Not recorded | 68fa7c231e209131cea8ca66 | | SPM | 23.18 | Not recorded | 68fa7c231e209131cea8ca66 | | SPM | 22.45 | Not recorded | 68fa7c231e209131cea8ca66 | | MACE | 20.53 | Not recorded | 68fa7c231e209131cea8ca66 | | RECE | 18.57 | Not recorded | 68fa7c231e209131cea8ca66 | ### ConAbl Snoopy CS: 25.26. Result ID: 69a7e8626dbb7075b94ca127. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### UCE Snoopy CS: 23.6. Result ID: 69a7e8626dbb7075b94ca12a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SPEED Snoopy CS: 23.58. Result ID: 69a7e8626dbb7075b94ca12d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SPM Snoopy CS: 23.18. Result ID: 69a7e8626dbb7075b94ca12b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Facilitated Transport":"True"}. This record does not assert independent reproduction. ### SPM Snoopy CS: 22.45. Result ID: 69a7e8626dbb7075b94ca12c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Facilitated Transport":"False"}. This record does not assert independent reproduction. ### MACE Snoopy CS: 20.53. Result ID: 69a7e8626dbb7075b94ca128. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RECE Snoopy CS: 18.57. Result ID: 69a7e8626dbb7075b94ca129. [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.