# Navigation Visual Question Answering on NaVQA Long (L) horizon

Source: [Navigation Visual Question Answering on NaVQA Long (L) horizon](<https://www.sota2.com/research/sota/navigation-visual-question-answering-on-navqa-long-l-horizon>)

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

Data updated: 2026-06-25T07:20:50.901Z

Evaluation of model performance on the long horizon subset of the NaVQA dataset for Navigation Visual Question Answering, reporting Descriptive Accuracy, Positional Error, and Temporal Error.

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 | Navigation Visual Question Answering |
| Dataset (including any stated split) | NaVQA Long (L) horizon |
| Metric | Descriptive Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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

- [Descriptive Accuracy](<https://www.sota2.com/research/sota/navigation-visual-question-answering-on-navqa-long-l-horizon.md?metric=Descriptive%20Accuracy>): Higher is better; unit: PERCENT.

- [Positional Error (m)](<https://www.sota2.com/research/sota/navigation-visual-question-answering-on-navqa-long-l-horizon.md?metric=Positional%20Error%20(m)>): Lower is better; unit: SCALAR.

- [Temporal Error (min)](<https://www.sota2.com/research/sota/navigation-visual-question-answering-on-navqa-long-l-horizon.md?metric=Temporal%20Error%20(min)>): Lower is better; unit: TIME.

## Recorded results

| Method | Descriptive Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| RAVEN | 80 | Not recorded | 6a3cb63d206542251d7ca6b7 |
| VLM Only | 65 | Not recorded | 6a3cb63d206542251d7ca6b7 |

### RAVEN

Descriptive Accuracy: 80. Result ID: 6a3cd6c5fbc373592c513ec4.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb63d206542251d7ca6b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a3cb63d206542251d7ca6b7.md>)

Recorded evaluation configuration: {"LLM":"Gemini-3-Pro","Encoder":"QQMM-v2","Quantization":"4-bit"}. This record does not assert independent reproduction.

### VLM Only

Descriptive Accuracy: 65. Result ID: 6a3cd6c5fbc373592c513ec5.

[Reporting paper](<https://www.sota2.com/research/paper/6a3cb63d206542251d7ca6b7.md>)

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

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