# Image2Description on Synthetic Eval Dataset Flowchart

Source: [Image2Description on Synthetic Eval Dataset Flowchart](<https://www.sota2.com/research/sota/image2description-on-synthetic-eval-dataset-flowchart>)

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

Data updated: 2026-02-26T05:37:01.718Z

Evaluation of model performance on generating natural language descriptions from images using the Synthetic Eval Dataset Flowchart, assessed using sequence generation metrics (BLEU, ROUGE, METEOR, etc.).

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 | Image2Description |
| Dataset (including any stated split) | Synthetic Eval Dataset Flowchart |
| Metric | BLEU |
| 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

- [BLEU](<https://www.sota2.com/research/sota/image2description-on-synthetic-eval-dataset-flowchart.md?metric=BLEU>): Higher is better; unit: SCALAR.

- [SacreBLEU](<https://www.sota2.com/research/sota/image2description-on-synthetic-eval-dataset-flowchart.md?metric=SacreBLEU>): Higher is better; unit: SCALAR.

- [METEOR](<https://www.sota2.com/research/sota/image2description-on-synthetic-eval-dataset-flowchart.md?metric=METEOR>): Higher is better; unit: SCALAR.

- [chrF](<https://www.sota2.com/research/sota/image2description-on-synthetic-eval-dataset-flowchart.md?metric=chrF>): Higher is better; unit: SCALAR.

- [BLEURT](<https://www.sota2.com/research/sota/image2description-on-synthetic-eval-dataset-flowchart.md?metric=BLEURT>): Higher is better; unit: SCALAR.

- [ROUGE-L](<https://www.sota2.com/research/sota/image2description-on-synthetic-eval-dataset-flowchart.md?metric=ROUGE-L>): Higher is better; unit: SCALAR.

## Recorded results

| Method | BLEU | Result date | Report paper |
| --- | --- | --- | --- |
| Llama-VL-TUG | 0.9914 | Not recorded | 697846caa2b28e761d87c70e |
| Qwen2.5-VL-7B-Instruct | 0.6823 | Not recorded | 697846caa2b28e761d87c70e |
| Llama3.2-11B-Vision-Instruct | 0.6598 | Not recorded | 697846caa2b28e761d87c70e |
| Gemma3-12B-Instruction-Tuned | 0.4695 | Not recorded | 697846caa2b28e761d87c70e |
| MiniCPM-V-2-6 | 0.297 | Not recorded | 697846caa2b28e761d87c70e |
| GPT-4o-mini | 0.2728 | Not recorded | 697846caa2b28e761d87c70e |

### Llama-VL-TUG

BLEU: 0.9914. Result ID: 69966d415698665ec4b41213.

[Reporting paper](<https://www.sota2.com/research/paper/697846caa2b28e761d87c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697846caa2b28e761d87c70e.md>)

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

### Qwen2.5-VL-7B-Instruct

BLEU: 0.6823. Result ID: 69966d415698665ec4b41215.

[Reporting paper](<https://www.sota2.com/research/paper/697846caa2b28e761d87c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

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

### Llama3.2-11B-Vision-Instruct

BLEU: 0.6598. Result ID: 69966d415698665ec4b41212.

[Reporting paper](<https://www.sota2.com/research/paper/697846caa2b28e761d87c70e.md>)

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

### Gemma3-12B-Instruction-Tuned

BLEU: 0.4695. Result ID: 69966d415698665ec4b41214.

[Reporting paper](<https://www.sota2.com/research/paper/697846caa2b28e761d87c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

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

### MiniCPM-V-2-6

BLEU: 0.297. Result ID: 69966d415698665ec4b41216.

[Reporting paper](<https://www.sota2.com/research/paper/697846caa2b28e761d87c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74681e209131cea6183d.md>)

[Implementation](<https://github.com/OpenBMB/MiniCPM-V>)

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

### GPT-4o-mini

BLEU: 0.2728. Result ID: 69966d415698665ec4b41217.

[Reporting paper](<https://www.sota2.com/research/paper/697846caa2b28e761d87c70e.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.