# Novel Object Synthesis on ImageNet-200 (test)

Source: [Novel Object Synthesis on ImageNet-200 (test)](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test>)

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

Data updated: 2026-02-26T05:36:24.438Z

Evaluation of models performing Novel Object Synthesis on the test split of the ImageNet-200 dataset, measured using similarity, balance, reward, HPSv2, and VQAScore.

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 | Novel Object Synthesis |
| Dataset (including any stated split) | ImageNet-200 (test) |
| Metric | Avg Sim (I-\>I) |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Avg Sim (I-\>I)](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test.md?metric=Avg%20Sim%20(I-%3EI)>): Higher is better; unit: SCALAR.

- [Avg Sim (I-\>T)](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test.md?metric=Avg%20Sim%20(I-%3ET)>): Higher is better; unit: SCALAR.

- [Balance (I-\>I)](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test.md?metric=Balance%20(I-%3EI)>): Higher is better; unit: SCALAR.

- [Balance (I-\>T)](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test.md?metric=Balance%20(I-%3ET)>): Higher is better; unit: SCALAR.

- [Reward](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test.md?metric=Reward>): Higher is better; unit: SCALAR.

- [HPSv2](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test.md?metric=HPSv2>): Higher is better; unit: SCALAR.

- [VQAScore](<https://www.sota2.com/research/sota/novel-object-synthesis-on-imagenet-200-test.md?metric=VQAScore>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Avg Sim (I-\>I) | Result date | Report paper |
| --- | --- | --- | --- |
| SDXL-Turbo | 0.7647 | Not recorded | 694a22289eebe296e72aa885 |
| RMLer | 0.7324 | Not recorded | 694a22289eebe296e72aa885 |
| GPT-Image-1 | 0.7308 | Not recorded | 694a22289eebe296e72aa885 |
| BASS | 0.7026 | Not recorded | 694a22289eebe296e72aa885 |
| ConceptLab | 0.5991 | Not recorded | 694a22289eebe296e72aa885 |

### SDXL-Turbo

Avg Sim (I-\>I): 0.7647. Result ID: 69981bd45698665ec475bd2d.

[Reporting paper](<https://www.sota2.com/research/paper/694a22289eebe296e72aa885.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70011e209131cea487fb.md>)

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

### RMLer

Avg Sim (I-\>I): 0.7324. Result ID: 69981bd45698665ec475bd2a.

[Reporting paper](<https://www.sota2.com/research/paper/694a22289eebe296e72aa885.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694a22289eebe296e72aa885.md>)

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

### GPT-Image-1

Avg Sim (I-\>I): 0.7308. Result ID: 69981bd45698665ec475bd2e.

[Reporting paper](<https://www.sota2.com/research/paper/694a22289eebe296e72aa885.md>)

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

### BASS

Avg Sim (I-\>I): 0.7026. Result ID: 69981bd45698665ec475bd2b.

[Reporting paper](<https://www.sota2.com/research/paper/694a22289eebe296e72aa885.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74241e209131cea5ff24.md>)

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

### ConceptLab

Avg Sim (I-\>I): 0.5991. Result ID: 69981bd45698665ec475bd2c.

[Reporting paper](<https://www.sota2.com/research/paper/694a22289eebe296e72aa885.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71791e209131cea5175d.md>)

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

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