# Navigation Visual Question Answering on NaVQA Source: [Navigation Visual Question Answering on NaVQA]() Collection index: [llms.txt]() Data updated: 2026-04-21T13:16:31.670Z Evaluation of model performance on the NaVQA dataset for Navigation Visual Question Answering, reporting Positional Accuracy, Temporal Accuracy, and Average Accuracy. 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 | | Metric | Positional Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 6 of 6 | | 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 - [Positional Accuracy](): Higher is better; unit: PERCENT. - [Temporal Accuracy](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Positional Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Graph Memory + ReMEmbR | 29.17 | Not recorded | 69e7051d7d0172140ec7f28c | | Graph Memory + ReMEmbR | 19.59 | Not recorded | 69e7051d7d0172140ec7f28c | | ReMEmbR | 19.59 | Not recorded | 69e7051d7d0172140ec7f28c | | Graph Memory | 13.75 | Not recorded | 69e7051d7d0172140ec7f28c | | Graph Memory | 13.75 | Not recorded | 69e7051d7d0172140ec7f28c | | ReMEmbR | 6.67 | Not recorded | 69e7051d7d0172140ec7f28c | ### Graph Memory + ReMEmbR Positional Accuracy: 29.17. Result ID: 69e7777a47f34ce924cf4d5a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VLM":"Florence-2-large"}. This record does not assert independent reproduction. ### Graph Memory + ReMEmbR Positional Accuracy: 19.59. Result ID: 69e7777a47f34ce924cf4d57. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VLM":"Florence-2-base"}. This record does not assert independent reproduction. ### ReMEmbR Positional Accuracy: 19.59. Result ID: 69e7777a47f34ce924cf4d59. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VLM":"Florence-2-large"}. This record does not assert independent reproduction. ### Graph Memory Positional Accuracy: 13.75. Result ID: 69e7777a47f34ce924cf4d55. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VLM":"Florence-2-base"}. This record does not assert independent reproduction. ### Graph Memory Positional Accuracy: 13.75. Result ID: 69e7777a47f34ce924cf4d58. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VLM":"Florence-2-large"}. This record does not assert independent reproduction. ### ReMEmbR Positional Accuracy: 6.67. Result ID: 69e7777a47f34ce924cf4d56. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"VLM":"Florence-2-base"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.