# 3D Question Answering on MMScan QA

Source: [3D Question Answering on MMScan QA](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa>)

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

Data updated: 2026-02-26T05:54:33.882Z

Evaluation of model performance on the MMScan QA dataset for 3D Question Answering tasks, utilizing a comprehensive set of metrics including structural scores (ST/OO), embedding similarities (SimCSE, Sentence-BERT), and standard generation metrics (BLEU, ROUGE, METEOR, EM).

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 | 3D Question Answering |
| Dataset (including any stated split) | MMScan QA |
| Metric | Overall Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Overall Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=Overall%20Score>): Higher is better; unit: PERCENT.

- [ST Attribute Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=ST%20Attribute%20Score>): Higher is better; unit: PERCENT.

- [ST Spatial Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=ST%20Spatial%20Score>): Higher is better; unit: PERCENT.

- [OO Attribute Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=OO%20Attribute%20Score>): Higher is better; unit: PERCENT.

- [OO Spatial Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=OO%20Spatial%20Score>): Higher is better; unit: PERCENT.

- [Object Relation Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=Object%20Relation%20Score>): Higher is better; unit: PERCENT.

- [Advanced Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=Advanced%20Score>): Higher is better; unit: PERCENT.

- [SimCSE Similarity](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=SimCSE%20Similarity>): Higher is better; unit: PERCENT.

- [Sentence-BERT Similarity](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=Sentence-BERT%20Similarity>): Higher is better; unit: PERCENT.

- [BLEU-1](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=BLEU-1>): Higher is better; unit: PERCENT.

- [BLEU-4](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=BLEU-4>): Higher is better; unit: PERCENT.

- [ROUGE-L](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=ROUGE-L>): Higher is better; unit: PERCENT.

- [METEOR Score](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=METEOR%20Score>): Higher is better; unit: PERCENT.

- [EM@1](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=EM%401>): Higher is better; unit: PERCENT.

- [EM@1 (Refined)](<https://www.sota2.com/research/sota/3d-question-answering-on-mmscan-qa.md?metric=EM%401%20(Refined)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Overall Score | Result date | Report paper |
| --- | --- | --- | --- |
| LLaVA-3D | 55.4 | Not recorded | 68fa78461e209131cea778b5 |
| LEO | 47.8 | Not recorded | 68fa78461e209131cea778b5 |
| LL3DA | 38.5 | Not recorded | 68fa78461e209131cea778b5 |
| 3D-LLM | 28.6 | Not recorded | 68fa78461e209131cea778b5 |
| Chat3D-v2 | 27.9 | Not recorded | 68fa78461e209131cea778b5 |
| LEO | 22.2 | Not recorded | 68fa78461e209131cea778b5 |
| LL3DA | 15.8 | Not recorded | 68fa78461e209131cea778b5 |

### LLaVA-3D

Overall Score: 55.4. Result ID: 699583963badfaffe67c0084.

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

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

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

Recorded evaluation configuration: {"Setting":"Generalist"}. This record does not assert independent reproduction.

### LEO

Overall Score: 47.8. Result ID: 699583963badfaffe67c0083.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa732c1e209131cea5aefe.md>)

[Implementation](<https://embodied-generalist.github.io/>)

Recorded evaluation configuration: {"Setting":"Fine-tuning"}. This record does not assert independent reproduction.

### LL3DA

Overall Score: 38.5. Result ID: 699583963badfaffe67c0082.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50d7d.md>)

Recorded evaluation configuration: {"Setting":"Fine-tuning"}. This record does not assert independent reproduction.

### 3D-LLM

Overall Score: 28.6. Result ID: 699583963badfaffe67c007e.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa70261e209131cea49662.md>)

[Implementation](<https://vis-www.cs.umass.edu/3dllm/>)

Recorded evaluation configuration: {"Setting":"Zero-Shot"}. This record does not assert independent reproduction.

### Chat3D-v2

Overall Score: 27.9. Result ID: 699583963badfaffe67c007f.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: {"Setting":"Zero-Shot"}. This record does not assert independent reproduction.

### LEO

Overall Score: 22.2. Result ID: 699583963badfaffe67c0081.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa732c1e209131cea5aefe.md>)

[Implementation](<https://embodied-generalist.github.io/>)

Recorded evaluation configuration: {"Setting":"Zero-Shot"}. This record does not assert independent reproduction.

### LL3DA

Overall Score: 15.8. Result ID: 699583963badfaffe67c0080.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50d7d.md>)

Recorded evaluation configuration: {"Setting":"Zero-Shot"}. This record does not assert independent reproduction.

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