# Mask-based Text-to-Image Generation on Mask-based T2I generation

Source: [Mask-based Text-to-Image Generation on Mask-based T2I generation](<https://www.sota2.com/research/sota/mask-based-text-to-image-generation-on-mask-based-t2i-generation>)

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

Data updated: 2026-02-26T05:47:43.396Z

Evaluation of Text-to-Image models specifically focused on mask-guided generation, utilizing generative quality metrics like FID, KID, and CLIP similarity scores.

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 | Mask-based Text-to-Image Generation |
| Dataset (including any stated split) | Mask-based T2I generation |
| Metric | KID |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 4 of 4 |
| 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

- [KID](<https://www.sota2.com/research/sota/mask-based-text-to-image-generation-on-mask-based-t2i-generation.md?metric=KID>): Lower is better; unit: SCALAR.

- [FID](<https://www.sota2.com/research/sota/mask-based-text-to-image-generation-on-mask-based-t2i-generation.md?metric=FID>): Lower is better; unit: SCALAR.

- [CLIP Similarity](<https://www.sota2.com/research/sota/mask-based-text-to-image-generation-on-mask-based-t2i-generation.md?metric=CLIP%20Similarity>): Higher is better; unit: SCALAR.

## Recorded results

| Method | KID | Result date | Report paper |
| --- | --- | --- | --- |
| SyncSDE | 34.859 | Not recorded | 68fa79151e209131cea7c0ad |
| SyncSDE | 43.774 | Not recorded | 68fa79151e209131cea7c0ad |
| MultiDiffusion | 47.694 | Not recorded | 68fa79151e209131cea7c0ad |
| SyncTweedies | 117.36 | Not recorded | 68fa79151e209131cea7c0ad |

### SyncSDE

KID: 34.859. Result ID: 6995c4a03badfaffe6c7c11d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79151e209131cea7c0ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79151e209131cea7c0ad.md>)

[Implementation](<https://github.com/Hleephilip/CSG>)

Recorded evaluation configuration: {"Backbone":"Stable Diffusion \[36\]","Selection":"best"}. This record does not assert independent reproduction.

### SyncSDE

KID: 43.774. Result ID: 6995c4a03badfaffe6c7c11c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79151e209131cea7c0ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79151e209131cea7c0ad.md>)

[Implementation](<https://github.com/Hleephilip/CSG>)

Recorded evaluation configuration: {"Backbone":"Stable Diffusion \[36\]","1/λ":5}. This record does not assert independent reproduction.

### MultiDiffusion

KID: 47.694. Result ID: 6995c4a03badfaffe6c7c11a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79151e209131cea7c0ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e8d1e209131cea40872.md>)

Recorded evaluation configuration: {"Backbone":"Stable Diffusion \[36\]"}. This record does not assert independent reproduction.

### SyncTweedies

KID: 117.36. Result ID: 6995c4a03badfaffe6c7c11b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79151e209131cea7c0ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa758f1e209131cea68a16.md>)

Recorded evaluation configuration: {"Backbone":"Stable Diffusion \[36\]"}. This record does not assert independent reproduction.

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