# Counterfactual Generation on StableDiffusion 3 Evaluation Set

Source: [Counterfactual Generation on StableDiffusion 3 Evaluation Set](<https://www.sota2.com/research/sota/counterfactual-generation-on-stablediffusion-3-evaluation-set>)

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

Data updated: 2026-03-05T08:29:45.666Z

Evaluation of Counterfactual Generation capabilities using the StableDiffusion 3 Evaluation Set, measured by metrics such as Logittgt, CLIPScore, and LPIPS.

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 Generation |
| Dataset (including any stated split) | StableDiffusion 3 Evaluation Set |
| Metric | LogitTgt |
| 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

- [LogitTgt](<https://www.sota2.com/research/sota/counterfactual-generation-on-stablediffusion-3-evaluation-set.md?metric=LogitTgt>): Higher is better; unit: SCALAR.

- [CLIP Score](<https://www.sota2.com/research/sota/counterfactual-generation-on-stablediffusion-3-evaluation-set.md?metric=CLIP%20Score>): Higher is better; unit: PERCENT.

- [LPIPS](<https://www.sota2.com/research/sota/counterfactual-generation-on-stablediffusion-3-evaluation-set.md?metric=LPIPS>): Lower is better; unit: SCALAR.

- [CLIP-Isrc](<https://www.sota2.com/research/sota/counterfactual-generation-on-stablediffusion-3-evaluation-set.md?metric=CLIP-Isrc>): Higher is better; unit: PERCENT.

## Recorded results

| Method | LogitTgt | Result date | Report paper |
| --- | --- | --- | --- |
| Gradient Ascent | 24.875 | Not recorded | 68fa7bc81e209131cea8a752 |
| Trajectory Optimization Control | 24.572 | Not recorded | 68fa7bc81e209131cea8a752 |
| Inversion+FreeDoM | 23.085 | Not recorded | 68fa7bc81e209131cea8a752 |
| Inversion+DPS | 21.628 | Not recorded | 68fa7bc81e209131cea8a752 |
| Inversion+TFG | 21.538 | Not recorded | 68fa7bc81e209131cea8a752 |
| None | 4.8722 | Not recorded | 68fa7bc81e209131cea8a752 |

### Gradient Ascent

LogitTgt: 24.875. Result ID: 69a93ee76dbb7075b952a846.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc81e209131cea8a752.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61c91e209131ce9f5e06.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 3"}. This record does not assert independent reproduction.

### Trajectory Optimization Control

LogitTgt: 24.572. Result ID: 69a93ee76dbb7075b952a84a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc81e209131cea8a752.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc81e209131cea8a752.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 3"}. This record does not assert independent reproduction.

### Inversion+FreeDoM

LogitTgt: 23.085. Result ID: 69a93ee76dbb7075b952a848.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc81e209131cea8a752.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec71e209131cea42035.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 3"}. This record does not assert independent reproduction.

### Inversion+DPS

LogitTgt: 21.628. Result ID: 69a93ee76dbb7075b952a847.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc81e209131cea8a752.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bdd1e209131cea8adbc.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 3"}. This record does not assert independent reproduction.

### Inversion+TFG

LogitTgt: 21.538. Result ID: 69a93ee76dbb7075b952a849.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc81e209131cea8a752.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75bc1e209131cea699d8.md>)

[Implementation](<https://www.sota2.com/Code%20will%20be%20available%20soon.>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 3"}. This record does not assert independent reproduction.

### None

LogitTgt: 4.8722. Result ID: 69a93ee76dbb7075b952a845.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc81e209131cea8a752.md>)

Recorded evaluation configuration: {"Backbone":"StableDiffusion 3"}. This record does not assert independent reproduction.

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