# Text-to-Image Generation Evaluation on GenAi-Bench

Source: [Text-to-Image Generation Evaluation on GenAi-Bench](<https://www.sota2.com/research/sota/text-to-image-generation-evaluation-on-genai-bench>)

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

Data updated: 2026-04-21T10:45:48.534Z

Evaluation of Text-to-Image generation models on the GenAi-Bench dataset, measured using Kendall Tau and Pairwise Accuracy 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 | Text-to-Image Generation Evaluation |
| Dataset (including any stated split) | GenAi-Bench |
| Metric | Kendall Tau B (Basic) |
| 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

- [Kendall Tau B (Basic)](<https://www.sota2.com/research/sota/text-to-image-generation-evaluation-on-genai-bench.md?metric=Kendall%20Tau%20B%20(Basic)>): Higher is better; unit: SCALAR.

- [Kendall Tau B (Advanced)](<https://www.sota2.com/research/sota/text-to-image-generation-evaluation-on-genai-bench.md?metric=Kendall%20Tau%20B%20(Advanced)>): Higher is better; unit: SCALAR.

- [Kendall Tau B (Overall)](<https://www.sota2.com/research/sota/text-to-image-generation-evaluation-on-genai-bench.md?metric=Kendall%20Tau%20B%20(Overall)>): Higher is better; unit: SCALAR.

- [Pairwise Accuracy (Basic)](<https://www.sota2.com/research/sota/text-to-image-generation-evaluation-on-genai-bench.md?metric=Pairwise%20Accuracy%20(Basic)>): Higher is better; unit: PERCENT.

- [Pairwise Accuracy (Advanced)](<https://www.sota2.com/research/sota/text-to-image-generation-evaluation-on-genai-bench.md?metric=Pairwise%20Accuracy%20(Advanced)>): Higher is better; unit: PERCENT.

- [Pairwise Accuracy (Overall)](<https://www.sota2.com/research/sota/text-to-image-generation-evaluation-on-genai-bench.md?metric=Pairwise%20Accuracy%20(Overall)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Kendall Tau B (Basic) | Result date | Report paper |
| --- | --- | --- | --- |
| CROCScore | 0.446 | Not recorded | 68fa789f1e209131cea796a5 |
| CROCScore (plain) | 0.443 | Not recorded | 68fa789f1e209131cea796a5 |
| VQAScore | 0.403 | Not recorded | 68fa789f1e209131cea796a5 |
| BLIP2-ITM | 0.081 | Not recorded | 68fa789f1e209131cea796a5 |
| AlignScore | 0.072 | Not recorded | 68fa789f1e209131cea796a5 |

### CROCScore

Kendall Tau B (Basic): 0.446. Result ID: 69e754d647f34ce924ce5b2a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa789f1e209131cea796a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa789f1e209131cea796a5.md>)

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

### CROCScore (plain)

Kendall Tau B (Basic): 0.443. Result ID: 69e754d647f34ce924ce5b29.

[Reporting paper](<https://www.sota2.com/research/paper/68fa789f1e209131cea796a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa789f1e209131cea796a5.md>)

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

### VQAScore

Kendall Tau B (Basic): 0.403. Result ID: 69e754d647f34ce924ce5b28.

[Reporting paper](<https://www.sota2.com/research/paper/68fa789f1e209131cea796a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a81e209131cea5dbb3.md>)

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

### BLIP2-ITM

Kendall Tau B (Basic): 0.081. Result ID: 69e754d647f34ce924ce5b27.

[Reporting paper](<https://www.sota2.com/research/paper/68fa789f1e209131cea796a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea46fe6.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/blip2>)

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

### AlignScore

Kendall Tau B (Basic): 0.072. Result ID: 69e754d647f34ce924ce5b26.

[Reporting paper](<https://www.sota2.com/research/paper/68fa789f1e209131cea796a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa757a1e209131cea683c9.md>)

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

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