# Visual Question Answering on GQA (Accuracy, Relative Score) Source: [Visual Question Answering on GQA (Accuracy, Relative Score)]() Collection index: [llms.txt]() Data updated: 2026-02-27T09:53:41.864Z Evaluation of model performance on the GQA dataset for visual question answering, reporting Accuracy and Relative Score metrics. 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 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [Relative Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Vanilla | 64.2 | Not recorded | 68fa760f1e209131cea6b33b | | VisionZip | 62.4 | Not recorded | 68fa760f1e209131cea6b33b | | VisionZip | 61.3 | Not recorded | 68fa760f1e209131cea6b33b | | VisionZip | 61 | Not recorded | 68fa760f1e209131cea6b33b | | SparseVLM | 60.3 | Not recorded | 68fa760f1e209131cea6b33b | | VisionZip | 59.3 | Not recorded | 68fa760f1e209131cea6b33b | | VisionZip | 58.2 | Not recorded | 68fa760f1e209131cea6b33b | | SparseVLM | 57.7 | Not recorded | 68fa760f1e209131cea6b33b | | VisionZip | 55.5 | Not recorded | 68fa760f1e209131cea6b33b | | SparseVLM | 51.2 | Not recorded | 68fa760f1e209131cea6b33b | ### Vanilla Accuracy: 64.2. Result ID: 69a1699535dfcabb00e6e3e9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"2880","Projector Fine-tuning":"No"}. This record does not assert independent reproduction. ### VisionZip Accuracy: 62.4. Result ID: 69a1699535dfcabb00e6e3ec. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"640","Projector Fine-tuning":"Yes"}. This record does not assert independent reproduction. ### VisionZip Accuracy: 61.3. Result ID: 69a1699535dfcabb00e6e3eb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"640","Projector Fine-tuning":"No"}. This record does not assert independent reproduction. ### VisionZip Accuracy: 61. Result ID: 69a1699535dfcabb00e6e3ef. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"320","Projector Fine-tuning":"Yes"}. This record does not assert independent reproduction. ### SparseVLM Accuracy: 60.3. Result ID: 69a1699535dfcabb00e6e3ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"640","Projector Fine-tuning":"No"}. This record does not assert independent reproduction. ### VisionZip Accuracy: 59.3. Result ID: 69a1699535dfcabb00e6e3ee. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"320","Projector Fine-tuning":"No"}. This record does not assert independent reproduction. ### VisionZip Accuracy: 58.2. Result ID: 69a1699535dfcabb00e6e3f2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"160","Projector Fine-tuning":"Yes"}. This record does not assert independent reproduction. ### SparseVLM Accuracy: 57.7. Result ID: 69a1699535dfcabb00e6e3ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"320","Projector Fine-tuning":"No"}. This record does not assert independent reproduction. ### VisionZip Accuracy: 55.5. Result ID: 69a1699535dfcabb00e6e3f1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"160","Projector Fine-tuning":"No"}. This record does not assert independent reproduction. ### SparseVLM Accuracy: 51.2. Result ID: 69a1699535dfcabb00e6e3f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-NeXT","Number of Visual Tokens":"160","Projector Fine-tuning":"No"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.