# Scene Text-centric Visual Question Answering on AI2D Source: [Scene Text-centric Visual Question Answering on AI2D]() Collection index: [llms.txt]() Data updated: 2026-05-21T10:17:11.195Z Evaluation of model performance on the AI2D dataset specifically for Scene Text-centric Visual Question Answering, reporting 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 | Scene Text-centric Visual Question Answering | | Dataset (including any stated split) | AI2D | | Metric | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | DeFacto | 76.1 | Not recorded | 68fa7b991e209131cea8964c | | GRIT | 75.5 | Not recorded | 68fa7b991e209131cea8964c | | Qwen2.5-VL | 69.5 | Not recorded | 68fa7b991e209131cea8964c | | Visual-SR1 | 69 | Not recorded | 68fa7b991e209131cea8964c | | ViCrop | 68.8 | Not recorded | 68fa7b991e209131cea8964c | | DeepEyes | 37 | Not recorded | 68fa7b991e209131cea8964c | ### DeFacto Accuracy: 76.1. Result ID: 6a0edb91fbc373592c06afe0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### GRIT Accuracy: 75.5. Result ID: 6a0edb91fbc373592c06afdc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction. ### Qwen2.5-VL Accuracy: 69.5. Result ID: 6a0edb91fbc373592c06afdb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### Visual-SR1 Accuracy: 69. Result ID: 6a0edb91fbc373592c06afdf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### ViCrop Accuracy: 68.8. Result ID: 6a0edb91fbc373592c06afdd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaVA-1.5 (Vicuna-7B)"}. This record does not assert independent reproduction. ### DeepEyes Accuracy: 37. Result ID: 6a0edb91fbc373592c06afde. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.