| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Visual Information Retrieval and Reasoning | ViDoSeek | Overall Accuracy74.87 | 18 | |
| Visual Document Retrieval | ViDoSeek | Doc Retrieval Score84.3 | 14 | |
| Video Document Seeking | ViDoSeek | Single Score44.34 | 14 | |
| Visual Question Answering | ViDoSeek | Single Accuracy0.7425 | 14 |