# Counterfactual Visual Explanation on CelebA (test)

Source: [Counterfactual Visual Explanation on CelebA (test)](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:44:21.089Z

Evaluation of counterfactual visual explanation methods applied to the CelebA test set, specifically focusing on the Age attribute, using metrics related to image quality (FID) and explanation fidelity.

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 | Counterfactual Visual Explanation (Age attribute) |
| Dataset (including any stated split) | CelebA (test) |
| Metric | FID |
| Unit | SCALAR |
| Direction | Lower 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

- [FID](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test.md?metric=FID>): Lower is better; unit: SCALAR.

- [sFID](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test.md?metric=sFID>): Lower is better; unit: SCALAR.

- [FVA](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test.md?metric=FVA>): Higher is better; unit: PERCENT.

- [FS](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test.md?metric=FS>): Higher is better; unit: SCALAR.

- [MNAC](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test.md?metric=MNAC>): Higher is better; unit: SCALAR.

- [CD](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test.md?metric=CD>): Lower is better; unit: SCALAR.

- [COUT](<https://www.sota2.com/research/sota/counterfactual-visual-explanation-on-celeba-test.md?metric=COUT>): Higher is better; unit: SCALAR.

## Recorded results

| Method | FID | Result date | Report paper |
| --- | --- | --- | --- |
| ACE | 1.45 | Not recorded | 68fa6ec71e209131cea4204f |
| ACE | 2.08 | Not recorded | 68fa6ec71e209131cea4204f |
| DIME | 4.15 | Not recorded | 68fa6ec71e209131cea4204f |
| STEEX | 11.8 | Not recorded | 68fa6ec71e209131cea4204f |
| DIVE | 33.8 | Not recorded | 68fa6ec71e209131cea4204f |
| DiVE | 39.9 | Not recorded | 68fa6ec71e209131cea4204f |

### ACE

FID: 1.45. Result ID: 699040c33a6c314849ff3f96.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

Recorded evaluation configuration: {"Distance function":"l1","Sampling steps":5}. This record does not assert independent reproduction.

### ACE

FID: 2.08. Result ID: 699040c33a6c314849ff3f97.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

Recorded evaluation configuration: {"Distance function":"l2","Sampling steps":5}. This record does not assert independent reproduction.

### DIME

FID: 4.15. Result ID: 699040c33a6c314849ff3f95.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c171e209131cea3215c.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### STEEX

FID: 11.8. Result ID: 699040c33a6c314849ff3f94.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d161e209131cea3799c.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DIVE

FID: 33.8. Result ID: 699040c33a6c314849ff3f92.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b301e209131cea2cb71.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DiVE

FID: 39.9. Result ID: 699040c33a6c314849ff3f93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea4204f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b301e209131cea2cb71.md>)

Recorded evaluation configuration: {"Sampling steps":100}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.