# Image Observation on 3D-RAD

Source: [Image Observation on 3D-RAD](<https://www.sota2.com/research/sota/image-observation-on-3d-rad>)

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

Data updated: 2026-03-27T12:51:52.346Z

Evaluation of model performance on the 3D-RAD dataset focusing on general Image Observation capabilities, measured using text generation metrics like BLEU, ROUGE, and BERT Score.

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 | Image Observation |
| Dataset (including any stated split) | 3D-RAD |
| Metric | BLEU |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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/image-observation-on-3d-rad.md?metric=BLEU>): Higher is better; unit: SCALAR.

- [ROUGE](<https://www.sota2.com/research/sota/image-observation-on-3d-rad.md?metric=ROUGE>): Higher is better; unit: SCALAR.

- [BERT Score](<https://www.sota2.com/research/sota/image-observation-on-3d-rad.md?metric=BERT%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | BLEU | Result date | Report paper |
| --- | --- | --- | --- |
| Photon | 51.59 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| M3D-P3 | 39.66 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| M3D-L2 | 31.28 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| OmniV | 16.42 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| M3D-P3 | 16.31 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| RadFM | 13.48 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| M3D-L2 | 10.69 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| Lingshu | 5.34 | Not recorded | 69c60efd0636e5bf4afc84c6 |
| Qwen2.5-VL | 3.51 | Not recorded | 69c60efd0636e5bf4afc84c6 |

### Photon

BLEU: 51.59. Result ID: 69c67d5847f34ce924b06db8.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Finetuned","Number of Parameters":"3B"}. This record does not assert independent reproduction.

### M3D-P3

BLEU: 39.66. Result ID: 69c67d5847f34ce924b06db7.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72be1e209131cea58507.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Finetuned","Number of Parameters":"4B"}. This record does not assert independent reproduction.

### M3D-L2

BLEU: 31.28. Result ID: 69c67d5847f34ce924b06db6.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72be1e209131cea58507.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Finetuned","Number of Parameters":"7B"}. This record does not assert independent reproduction.

### OmniV

BLEU: 16.42. Result ID: 69c67d5847f34ce924b06db4.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78191e209131cea76c2f.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Number of Parameters":"1.5B"}. This record does not assert independent reproduction.

### M3D-P3

BLEU: 16.31. Result ID: 69c67d5847f34ce924b06db3.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72be1e209131cea58507.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Number of Parameters":"4B"}. This record does not assert independent reproduction.

### RadFM

BLEU: 13.48. Result ID: 69c67d5847f34ce924b06db1.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71411e209131cea5023d.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Number of Parameters":"13B"}. This record does not assert independent reproduction.

### M3D-L2

BLEU: 10.69. Result ID: 69c67d5847f34ce924b06db2.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72be1e209131cea58507.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Number of Parameters":"7B"}. This record does not assert independent reproduction.

### Lingshu

BLEU: 5.34. Result ID: 69c67d5847f34ce924b06db5.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795a1e209131cea7d387.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Number of Parameters":"7B"}. This record does not assert independent reproduction.

### Qwen2.5-VL

BLEU: 3.51. Result ID: 69c67d5847f34ce924b06db0.

[Reporting paper](<https://www.sota2.com/research/paper/69c60efd0636e5bf4afc84c6.md>)

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

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Number of Parameters":"3B"}. This record does not assert independent reproduction.

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