| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multimodal Reasoning | Multimodal Reasoning Benchmarks Combined | Overall Score71.86 | 18 | |
| Visual Information Preservation and Explainability Evaluation | Multimodal Reasoning Benchmarks (MathVista, WeMath, MMStar, MMMU, R1-Onevision-Bench) (test) | Visual Info Preservation Score3.95 | 4 | |
| Visual Question Answering | Multimodal Reasoning Benchmarks | FlipVQA Accuracy32.97 | 2 |