# Description Quality on Scene 2 Suite main room

Source: [Description Quality on Scene 2 Suite main room](<https://www.sota2.com/research/sota/description-quality-on-scene-2-suite-main-room>)

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

Data updated: 2026-02-26T05:39:33.095Z

Evaluation of model performance on Description Quality tasks within the Scene 2 Suite main room environment, measured by Color Accuracy and localization metrics (Loc@parent, Loc@path).

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 | Description Quality |
| Dataset (including any stated split) | Scene 2 Suite main room |
| Metric | Color Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Color Accuracy](<https://www.sota2.com/research/sota/description-quality-on-scene-2-suite-main-room.md?metric=Color%20Accuracy>): Higher is better; unit: PERCENT.

- [Loc@parent](<https://www.sota2.com/research/sota/description-quality-on-scene-2-suite-main-room.md?metric=Loc%40parent>): Higher is better; unit: PERCENT.

- [Loc@path](<https://www.sota2.com/research/sota/description-quality-on-scene-2-suite-main-room.md?metric=Loc%40path>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Color Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SpatialMem | 79 | Not recorded | 6971af2de25ad2fa9d1edabf |
| gemini | 76 | Not recorded | 6971af2de25ad2fa9d1edabf |
| internvl | 74 | Not recorded | 6971af2de25ad2fa9d1edabf |
| llava | 72 | Not recorded | 6971af2de25ad2fa9d1edabf |
| internvl | 68 | Not recorded | 6971af2de25ad2fa9d1edabf |
| openai | 66 | Not recorded | 6971af2de25ad2fa9d1edabf |
| together | 62 | Not recorded | 6971af2de25ad2fa9d1edabf |
| qwen25vl | 55 | Not recorded | 6971af2de25ad2fa9d1edabf |

### SpatialMem

Color Accuracy: 79. Result ID: 69979f325698665ec4f52f4e.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"SpatialMem (Our Model)"}. This record does not assert independent reproduction.

### gemini

Color Accuracy: 76. Result ID: 69979f325698665ec4f52f47.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"Google Gemini"}. This record does not assert independent reproduction.

### internvl

Color Accuracy: 74. Result ID: 69979f325698665ec4f52f48.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"InternVL 2.5","Deployment":"Local","Entry":"1"}. This record does not assert independent reproduction.

### llava

Color Accuracy: 72. Result ID: 69979f325698665ec4f52f49.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"LLaVA-Video","Deployment":"Local"}. This record does not assert independent reproduction.

### internvl

Color Accuracy: 68. Result ID: 69979f325698665ec4f52f4a.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"InternVL 2.5","Deployment":"Local","Entry":"2"}. This record does not assert independent reproduction.

### openai

Color Accuracy: 66. Result ID: 69979f325698665ec4f52f4b.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"OpenAI GPT-4o"}. This record does not assert independent reproduction.

### together

Color Accuracy: 62. Result ID: 69979f325698665ec4f52f4c.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-VL","Deployment":"Cloud"}. This record does not assert independent reproduction.

### qwen25vl

Color Accuracy: 55. Result ID: 69979f325698665ec4f52f4d.

[Reporting paper](<https://www.sota2.com/research/paper/6971af2de25ad2fa9d1edabf.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-VL","Deployment":"Local"}. This record does not assert independent reproduction.

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