# Text Generation on ScanNet and 3D-FRONT (test)

Source: [Text Generation on ScanNet and 3D-FRONT (test)](<https://www.sota2.com/research/sota/text-generation-on-scannet-and-3d-front-test>)

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

Data updated: 2026-05-12T12:53:13.345Z

Evaluation of Text Generation performance across the combined ScanNet and 3D-FRONT test sets, measured using ROUGE 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 Generation |
| Dataset (including any stated split) | ScanNet and 3D-FRONT (test) |
| Metric | ROUGE-1 |
| Unit | PERCENT |
| 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

- [ROUGE-1](<https://www.sota2.com/research/sota/text-generation-on-scannet-and-3d-front-test.md?metric=ROUGE-1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ROUGE-1 | Result date | Report paper |
| --- | --- | --- | --- |
| LLaVA-3D-7B (Teacher) | 100 | Not recorded | 6a02b447dbc836b87d824927 |
| Our Student (Distilled) | 11.9 | Not recorded | 6a02b447dbc836b87d824927 |
| PaliGemma-3B | 6.1 | Not recorded | 6a02b447dbc836b87d824927 |
| MobileVLM-2B | 5.2 | Not recorded | 6a02b447dbc836b87d824927 |
| LLaVA-v1.5-7B (2D) | 4.5 | Not recorded | 6a02b447dbc836b87d824927 |

### LLaVA-3D-7B (Teacher)

ROUGE-1: 100. Result ID: 6a0322b847f34ce924e76fc3.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b447dbc836b87d824927.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78461e209131cea778b5.md>)

[Implementation](<https://zcmax.github.io/projects/LLaVA-3D>)

Recorded evaluation configuration: {"Params":"7.0B","FPS":"0.0075","Size (GB)":"26.7"}. This record does not assert independent reproduction.

### Our Student (Distilled)

ROUGE-1: 11.9. Result ID: 6a0322b847f34ce924e76fc7.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b447dbc836b87d824927.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b447dbc836b87d824927.md>)

Recorded evaluation configuration: {"Params":"2.29B","FPS":"0.065","Size (GB)":"8.7"}. This record does not assert independent reproduction.

### PaliGemma-3B

ROUGE-1: 6.1. Result ID: 6a0322b847f34ce924e76fc6.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b447dbc836b87d824927.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa751e1e209131cea65e0c.md>)

Recorded evaluation configuration: {"Params":"3.0B","FPS":"0.055","Size (GB)":"11.2"}. This record does not assert independent reproduction.

### MobileVLM-2B

ROUGE-1: 5.2. Result ID: 6a0322b847f34ce924e76fc5.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b447dbc836b87d824927.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71a71e209131cea52555.md>)

[Implementation](<https://github.com/Meituan-AutoML/MobileVLM>)

Recorded evaluation configuration: {"Params":"2.0B","FPS":"0.072","Size (GB)":"7.5"}. This record does not assert independent reproduction.

### LLaVA-v1.5-7B (2D)

ROUGE-1: 4.5. Result ID: 6a0322b847f34ce924e76fc4.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b447dbc836b87d824927.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea51970.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Params":"7.0B","FPS":"0.008","Size (GB)":"26.7"}. This record does not assert independent reproduction.

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