# Spatial Mental Modeling on SAT (synthesized)

Source: [Spatial Mental Modeling on SAT (synthesized)](<https://www.sota2.com/research/sota/spatial-mental-modeling-on-sat-synthesized>)

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

Data updated: 2026-02-26T05:41:56.443Z

Evaluation of model performance on the synthesized version of the SAT dataset, focusing on Spatial Mental Modeling capabilities, measured by EgoM, ObjM, EgoAct, Goal, Pers, and AVG.

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 | Spatial Mental Modeling |
| Dataset (including any stated split) | SAT (synthesized) |
| Metric | EgoM |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [EgoM](<https://www.sota2.com/research/sota/spatial-mental-modeling-on-sat-synthesized.md?metric=EgoM>): Higher is better; unit: PERCENT.

- [ObjM](<https://www.sota2.com/research/sota/spatial-mental-modeling-on-sat-synthesized.md?metric=ObjM>): Higher is better; unit: PERCENT.

- [EgoAct](<https://www.sota2.com/research/sota/spatial-mental-modeling-on-sat-synthesized.md?metric=EgoAct>): Higher is better; unit: PERCENT.

- [Goal](<https://www.sota2.com/research/sota/spatial-mental-modeling-on-sat-synthesized.md?metric=Goal>): Higher is better; unit: PERCENT.

- [Pers](<https://www.sota2.com/research/sota/spatial-mental-modeling-on-sat-synthesized.md?metric=Pers>): Higher is better; unit: PERCENT.

- [AVG](<https://www.sota2.com/research/sota/spatial-mental-modeling-on-sat-synthesized.md?metric=AVG>): Higher is better; unit: PERCENT.

## Recorded results

| Method | EgoM | Result date | Report paper |
| --- | --- | --- | --- |
| SpatialDreamer | 95.4 | Not recorded | 6937ad456cfd488761a4eb9e |
| MindJourney (o1, SVC) | 87.1 | Not recorded | 6937ad456cfd488761a4eb9e |
| MindJourney (o1, SWM) | 82.4 | Not recorded | 6937ad456cfd488761a4eb9e |
| SAT (LLaVA-Video-7B) | 79.6 | Not recorded | 6937ad456cfd488761a4eb9e |
| o1 | 78 | Not recorded | 6937ad456cfd488761a4eb9e |
| InternVL3-14B | 77.6 | Not recorded | 6937ad456cfd488761a4eb9e |
| GPT-4.1 | 75.3 | Not recorded | 6937ad456cfd488761a4eb9e |
| Gemini 1.5-flash | 67.1 | Not recorded | 6937ad456cfd488761a4eb9e |
| GPT-4o | 64.7 | Not recorded | 6937ad456cfd488761a4eb9e |
| SAT (LLaVA-1.5-13B) | 61.7 | Not recorded | 6937ad456cfd488761a4eb9e |
| Gemini 1.5-pro | 57.7 | Not recorded | 6937ad456cfd488761a4eb9e |
| LLaVA-Video-7B | 56.4 | Not recorded | 6937ad456cfd488761a4eb9e |
| GPT-4V | 54.7 | Not recorded | 6937ad456cfd488761a4eb9e |
| Robopoint-13B | 50.2 | Not recorded | 6937ad456cfd488761a4eb9e |
| LLaVA-1.5-13B | 46.6 | Not recorded | 6937ad456cfd488761a4eb9e |

### SpatialDreamer

EgoM: 95.4. Result ID: 699873cc5698665ec4d40ea9.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

Recorded evaluation configuration: {"Category":"Spatial MLLM","Backbone":"Qwen2.5-VL-7B-Instruct"}. This record does not assert independent reproduction.

### MindJourney (o1, SVC)

EgoM: 87.1. Result ID: 699873cc5698665ec4d40ea4.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a0a1e209131cea80bbb.md>)

Recorded evaluation configuration: {"Category":"Test-time Scaling Method","Base Model":"o1","World Model":"SVC"}. This record does not assert independent reproduction.

### MindJourney (o1, SWM)

EgoM: 82.4. Result ID: 699873cc5698665ec4d40ea5.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a0a1e209131cea80bbb.md>)

Recorded evaluation configuration: {"Category":"Test-time Scaling Method","Base Model":"o1","World Model":"SWM"}. This record does not assert independent reproduction.

### SAT (LLaVA-Video-7B)

EgoM: 79.6. Result ID: 699873cc5698665ec4d40ea8.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70611e209131cea4abdb.md>)

Recorded evaluation configuration: {"Category":"Spatial MLLM","Backbone":"LLaVA-Video-7B"}. This record does not assert independent reproduction.

### o1

EgoM: 78. Result ID: 699873cc5698665ec4d40ea0.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76691e209131cea6ccaa.md>)

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

### InternVL3-14B

EgoM: 77.6. Result ID: 699873cc5698665ec4d40ea1.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78241e209131cea76fa5.md>)

Recorded evaluation configuration: {"Category":"Generalist MLLM","Size":"14B"}. This record does not assert independent reproduction.

### GPT-4.1

EgoM: 75.3. Result ID: 699873cc5698665ec4d40e9f.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

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

### Gemini 1.5-flash

EgoM: 67.1. Result ID: 699873cc5698665ec4d40e9c.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76361e209131cea6be11.md>)

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

### GPT-4o

EgoM: 64.7. Result ID: 699873cc5698665ec4d40e9e.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

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

### SAT (LLaVA-1.5-13B)

EgoM: 61.7. Result ID: 699873cc5698665ec4d40ea7.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70611e209131cea4abdb.md>)

Recorded evaluation configuration: {"Category":"Spatial MLLM","Backbone":"LLaVA-1.5-13B"}. This record does not assert independent reproduction.

### Gemini 1.5-pro

EgoM: 57.7. Result ID: 699873cc5698665ec4d40e9d.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76361e209131cea6be11.md>)

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

### LLaVA-Video-7B

EgoM: 56.4. Result ID: 699873cc5698665ec4d40ea3.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a561e209131cea82934.md>)

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

Recorded evaluation configuration: {"Category":"Generalist MLLM","Size":"7B"}. This record does not assert independent reproduction.

### GPT-4V

EgoM: 54.7. Result ID: 699873cc5698665ec4d40e9b.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

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

### Robopoint-13B

EgoM: 50.2. Result ID: 699873cc5698665ec4d40ea6.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73b41e209131cea5e061.md>)

[Implementation](<https://www.sota2.com/robo-point.github.io>)

Recorded evaluation configuration: {"Category":"Spatial MLLM","Size":"13B"}. This record does not assert independent reproduction.

### LLaVA-1.5-13B

EgoM: 46.6. Result ID: 699873cc5698665ec4d40ea2.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad456cfd488761a4eb9e.md>)

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

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

Recorded evaluation configuration: {"Category":"Generalist MLLM","Size":"13B"}. This record does not assert independent reproduction.

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