# Visual Question Answering on GQA 2019 (test) Source: [Visual Question Answering on GQA 2019 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:37:48.683Z Evaluation of Visual Question Answering performance on the test split of the GQA 2019 dataset, utilizing metrics such as Binary, Open, Consistency, Plausibility, Validity, Distribution, and 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 | Visual Question Answering | | Dataset (including any stated split) | GQA 2019 (test) | | Metric | Binary Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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 - [Binary Accuracy](): Higher is better; unit: PERCENT. - [Open Accuracy](): Higher is better; unit: PERCENT. - [Consistency Score](): Higher is better; unit: PERCENT. - [Plausibility Score](): Higher is better; unit: PERCENT. - [Validity Score](): Higher is better; unit: PERCENT. - [Distribution Score](): Higher is better; unit: PERCENT. - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Binary Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Human | 91.2 | Not recorded | 68fa6b131e209131cea2bf8c | | ProTo | 79.12 | Not recorded | 68fa6b131e209131cea2bf8c | | NSM | 78.94 | Not recorded | 68fa6b131e209131cea2bf8c | | MMN | 78.9 | Not recorded | 68fa6b131e209131cea2bf8c | | PVR | 78.02 | Not recorded | 68fa6b131e209131cea2bf8c | | LXMERT | 77.16 | Not recorded | 68fa6b131e209131cea2bf8c | | SNMN | 73.4 | Not recorded | 68fa6b131e209131cea2bf8c | | MAC | 71.23 | Not recorded | 68fa6b131e209131cea2bf8c | | BottomUp | 66.64 | Not recorded | 68fa6b131e209131cea2bf8c | ### Human Binary Accuracy: 91.2. Result ID: 698e52f1ffc3901de09ca099. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Signal":null}. This record does not assert independent reproduction. ### ProTo Binary Accuracy: 79.12. Result ID: 698e52f1ffc3901de09ca0a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Signal":"Prog"}. This record does not assert independent reproduction. ### NSM Binary Accuracy: 78.94. Result ID: 698e52f1ffc3901de09ca09d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Signal":"SG"}. This record does not assert independent reproduction. ### MMN Binary Accuracy: 78.9. Result ID: 698e52f1ffc3901de09ca0a0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Signal":"Prog"}. This record does not assert independent reproduction. ### PVR Binary Accuracy: 78.02. Result ID: 698e52f1ffc3901de09ca09e. [Reporting paper]() Recorded evaluation configuration: {"Signal":"Prog"}. This record does not assert independent reproduction. ### LXMERT Binary Accuracy: 77.16. Result ID: 698e52f1ffc3901de09ca09c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Signal":"QA"}. This record does not assert independent reproduction. ### SNMN Binary Accuracy: 73.4. Result ID: 698e52f1ffc3901de09ca09f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Signal":"Prog"}. This record does not assert independent reproduction. ### MAC Binary Accuracy: 71.23. Result ID: 698e52f1ffc3901de09ca09b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Signal":"QA"}. This record does not assert independent reproduction. ### BottomUp Binary Accuracy: 66.64. Result ID: 698e52f1ffc3901de09ca09a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Signal":"QA"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.