# Chart-based Question Answering on EvoChart-QA Source: [Chart-based Question Answering on EvoChart-QA]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:39.023Z Evaluation of model performance on the EvoChart-QA dataset for Chart-based Question Answering capabilities, measured using metrics such as 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 | Chart-based Question Answering | | Dataset (including any stated split) | EvoChart-QA | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 | | --- | --- | --- | --- | | SketchVL-7B | 58.64 | Not recorded | 69647fd753fdd5c361744506 | | Qwen2.5VL-7B | 54.8 | Not recorded | 69647fd753fdd5c361744506 | | SketchVL-3B | 48.56 | Not recorded | 69647fd753fdd5c361744506 | | SketchVL-3B | 48.08 | Not recorded | 69647fd753fdd5c361744506 | | SketchVL-3B | 47.28 | Not recorded | 69647fd753fdd5c361744506 | | SketchVL-3B | 45.6 | Not recorded | 69647fd753fdd5c361744506 | | SketchVL-3B | 42.08 | Not recorded | 69647fd753fdd5c361744506 | | VLM-R1 | 40.32 | Not recorded | 69647fd753fdd5c361744506 | | Qwen2.5VL-3B | 39.36 | Not recorded | 69647fd753fdd5c361744506 | | SketchVL-3B | 30.48 | Not recorded | 69647fd753fdd5c361744506 | | ChartSketcher-2B | 26.72 | Not recorded | 69647fd753fdd5c361744506 | | SketchVL-3B | 26.48 | Not recorded | 69647fd753fdd5c361744506 | ### SketchVL-7B Accuracy: 58.64. Result ID: 6997d1be5698665ec429520f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Training Strategy":"FinePO"}. This record does not assert independent reproduction. ### Qwen2.5VL-7B Accuracy: 54.8. Result ID: 6997d1be5698665ec429520d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B"}. This record does not assert independent reproduction. ### SketchVL-3B Accuracy: 48.56. Result ID: 6997d1be5698665ec4295214. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","KL Action Regularization":"Disabled"}. This record does not assert independent reproduction. ### SketchVL-3B Accuracy: 48.08. Result ID: 6997d1be5698665ec4295213. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Reward Mechanism":"Random scores"}. This record does not assert independent reproduction. ### SketchVL-3B Accuracy: 47.28. Result ID: 6997d1be5698665ec4295210. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Training Strategy":"FinePO (Full Model)"}. This record does not assert independent reproduction. ### SketchVL-3B Accuracy: 45.6. Result ID: 6997d1be5698665ec4295212. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Training Strategy":"Naive GRPO"}. This record does not assert independent reproduction. ### SketchVL-3B Accuracy: 42.08. Result ID: 6997d1be5698665ec4295211. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Training Strategy":"Cold Start only"}. This record does not assert independent reproduction. ### VLM-R1 Accuracy: 40.32. Result ID: 6997d1be5698665ec429520b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5VL-3B Accuracy: 39.36. Result ID: 6997d1be5698665ec429520e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B"}. This record does not assert independent reproduction. ### SketchVL-3B Accuracy: 30.48. Result ID: 6997d1be5698665ec4295215. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Reasoning Method":"Zero GRPO (No Sketch)"}. This record does not assert independent reproduction. ### ChartSketcher-2B Accuracy: 26.72. Result ID: 6997d1be5698665ec429520c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"2B"}. This record does not assert independent reproduction. ### SketchVL-3B Accuracy: 26.48. Result ID: 6997d1be5698665ec4295216. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Training Strategy":"SFT (No RL)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.