# Style Erasure on SDXL Erasing Van Gogh Source: [Style Erasure on SDXL Erasing Van Gogh]() Collection index: [llms.txt]() Data updated: 2026-05-12T15:02:43.278Z Evaluation of Style Erasure performance on the SDXL Erasing Van Gogh dataset, measured using style consistency (CS), style alignment (CA), and kernel inception distance (KID) across multiple target styles (Van Gogh, Monet, Picasso, Watercolor, Hokusai, Warhol), alongside resource consumption metrics. 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 | Style Erasure | | Dataset (including any stated split) | SDXL Erasing Van Gogh | | Metric | CS (Van Gogh) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 13 of 13 | | 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 - [CS (Van Gogh)](): Higher is better; unit: SCALAR. - [CA (Van Gogh)](): Higher is better; unit: SCALAR. - [KID (Van Gogh)](): Lower is better; unit: SCALAR. - [CS (Monet)](): Higher is better; unit: SCALAR. - [CA (Monet)](): Higher is better; unit: SCALAR. - [KID (Monet)](): Lower is better; unit: SCALAR. - [CS (Picasso)](): Higher is better; unit: SCALAR. - [CA (Picasso)](): Higher is better; unit: SCALAR. - [KID (Picasso)](): Lower is better; unit: SCALAR. - [CS (Watercolor)](): Higher is better; unit: SCALAR. - [CA (Watercolor)](): Higher is better; unit: SCALAR. - [KID (Watercolor)](): Lower is better; unit: SCALAR. - [CS (Hokusai)](): Higher is better; unit: SCALAR. - [CA (Hokusai)](): Higher is better; unit: SCALAR. - [KID (Hokusai)](): Lower is better; unit: SCALAR. - [CS (Warhol)](): Higher is better; unit: SCALAR. - [CA (Warhol)](): Higher is better; unit: SCALAR. - [KID (Warhol)](): Lower is better; unit: SCALAR. - [Memory (GB)](): Lower is better; unit: SIZE. - [Size (MB)](): Lower is better; unit: SIZE. - [Resource Cost (MB)](): Lower is better; unit: SIZE. ## Recorded results | Method | CS (Van Gogh) | Result date | Report paper | | --- | --- | --- | --- | | SDXL | 33.81 | Not recorded | 6a02b49bdbc836b87d824e13 | | SPEED | 33.64 | Not recorded | 6a02b49bdbc836b87d824e13 | | UCE | 33.56 | Not recorded | 6a02b49bdbc836b87d824e13 | | SPEED | 33.49 | Not recorded | 6a02b49bdbc836b87d824e13 | | UCE | 33.39 | Not recorded | 6a02b49bdbc836b87d824e13 | | RECE | 32.8 | Not recorded | 6a02b49bdbc836b87d824e13 | | SAFREE | 32.44 | Not recorded | 6a02b49bdbc836b87d824e13 | | SAFREE | 31.81 | Not recorded | 6a02b49bdbc836b87d824e13 | | CAbl | 31.55 | Not recorded | 6a02b49bdbc836b87d824e13 | | CAbl | 31.29 | Not recorded | 6a02b49bdbc836b87d824e13 | | SPACE | 30 | Not recorded | 6a02b49bdbc836b87d824e13 | | RECE | 29.75 | Not recorded | 6a02b49bdbc836b87d824e13 | | SPACE | 28.76 | Not recorded | 6a02b49bdbc836b87d824e13 | ### SDXL CS (Van Gogh): 33.81. Result ID: 6a03405d47f34ce924e83bc8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"baseline":"true"}. This record does not assert independent reproduction. ### SPEED CS (Van Gogh): 33.64. Result ID: 6a03405d47f34ce924e83bd9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### UCE CS (Van Gogh): 33.56. Result ID: 6a03405d47f34ce924e83bcb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### SPEED CS (Van Gogh): 33.49. Result ID: 6a03405d47f34ce924e83bcd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### UCE CS (Van Gogh): 33.39. Result ID: 6a03405d47f34ce924e83bd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### RECE CS (Van Gogh): 32.8. Result ID: 6a03405d47f34ce924e83bcc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### SAFREE CS (Van Gogh): 32.44. Result ID: 6a03405d47f34ce924e83bd6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"false"}. This record does not assert independent reproduction. ### SAFREE CS (Van Gogh): 31.81. Result ID: 6a03405d47f34ce924e83bca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"false"}. This record does not assert independent reproduction. ### CAbl CS (Van Gogh): 31.55. Result ID: 6a03405d47f34ce924e83bd5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### CAbl CS (Van Gogh): 31.29. Result ID: 6a03405d47f34ce924e83bc9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### SPACE CS (Van Gogh): 30. Result ID: 6a03405d47f34ce924e83bce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### RECE CS (Van Gogh): 29.75. Result ID: 6a03405d47f34ce924e83bd8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. ### SPACE CS (Van Gogh): 28.76. Result ID: 6a03405d47f34ce924e83bda. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"closed-form":"true"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.