# Subject-driven Generation on DreamBench subject generation

Source: [Subject-driven Generation on DreamBench subject generation](<https://www.sota2.com/research/sota/subject-driven-generation-on-dreambench-subject-generation>)

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

Data updated: 2026-02-26T05:40:17.555Z

Evaluation of models performing subject-driven generation using the DreamBench subject generation context, assessed using DINO, CLIP-I, CLIP-T, and ImageReward 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 | Subject-driven Generation |
| Dataset (including any stated split) | DreamBench subject generation |
| Metric | DINO Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [DINO Score](<https://www.sota2.com/research/sota/subject-driven-generation-on-dreambench-subject-generation.md?metric=DINO%20Score>): Higher is better; unit: SCALAR.

- [CLIP-I Score](<https://www.sota2.com/research/sota/subject-driven-generation-on-dreambench-subject-generation.md?metric=CLIP-I%20Score>): Higher is better; unit: SCALAR.

- [CLIP-T Score](<https://www.sota2.com/research/sota/subject-driven-generation-on-dreambench-subject-generation.md?metric=CLIP-T%20Score>): Higher is better; unit: SCALAR.

- [ImageReward Score](<https://www.sota2.com/research/sota/subject-driven-generation-on-dreambench-subject-generation.md?metric=ImageReward%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | DINO Score | Result date | Report paper |
| --- | --- | --- | --- |
| MCA-Ctrl | 0.6724 | Not recorded | 68fa78591e209131cea77e2d |
| BLIP-Diffusion | 0.67 | Not recorded | 68fa78591e209131cea77e2d |
| DreamBooth | 0.668 | Not recorded | 68fa78591e209131cea77e2d |
| FreeCustom | 0.666 | Not recorded | 68fa78591e209131cea77e2d |
| MCA-Ctrl | 0.661 | Not recorded | 68fa78591e209131cea77e2d |
| IP-Adapter | 0.6504 | Not recorded | 68fa78591e209131cea77e2d |
| Re-Imagen | 0.6 | Not recorded | 68fa78591e209131cea77e2d |
| Textual Inversion | 0.569 | Not recorded | 68fa78591e209131cea77e2d |

### MCA-Ctrl

DINO Score: 0.6724. Result ID: 699586e63badfaffe67fd5e1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Implementation](<https://github.com/yanghan-yh/MCA-Ctrl>)

Recorded evaluation configuration: {"Mode":"zero-shot","Configuration":"Specified"}. This record does not assert independent reproduction.

### BLIP-Diffusion

DINO Score: 0.67. Result ID: 699586e63badfaffe67fd5dd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fe21e209131cea47e53.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/blip-diffusion>)

Recorded evaluation configuration: {"Mode":"zero-shot"}. This record does not assert independent reproduction.

### DreamBooth

DINO Score: 0.668. Result ID: 699586e63badfaffe67fd5da.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d81.md>)

[Implementation](<https://dreambooth.github.io/>)

Recorded evaluation configuration: {"Mode":"fine-tuning"}. This record does not assert independent reproduction.

### FreeCustom

DINO Score: 0.666. Result ID: 699586e63badfaffe67fd5df.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73441e209131cea5b662.md>)

Recorded evaluation configuration: {"Mode":"zero-shot"}. This record does not assert independent reproduction.

### MCA-Ctrl

DINO Score: 0.661. Result ID: 699586e63badfaffe67fd5e0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Implementation](<https://github.com/yanghan-yh/MCA-Ctrl>)

Recorded evaluation configuration: {"Mode":"zero-shot","Configuration":"Uniform"}. This record does not assert independent reproduction.

### IP-Adapter

DINO Score: 0.6504. Result ID: 699586e63badfaffe67fd5de.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70591e209131cea4a798.md>)

[Implementation](<https://ip-adapter.github.io/>)

Recorded evaluation configuration: {"Mode":"zero-shot"}. This record does not assert independent reproduction.

### Re-Imagen

DINO Score: 0.6. Result ID: 699586e63badfaffe67fd5dc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dfb1e209131cea3d26f.md>)

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

### Textual Inversion

DINO Score: 0.569. Result ID: 699586e63badfaffe67fd5db.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78591e209131cea77e2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d2d1e209131cea38190.md>)

Recorded evaluation configuration: {"Mode":"fine-tuning"}. This record does not assert independent reproduction.

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