| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MMLongBench-Doc | Synthetic Reasoning | Accuracy55.8 | 58 | 3mo ago | |
| LongDocURL | GPT-4o | Accuracy64.5 | 14 | 3mo ago | |
| MMLB 128K | Qwen3 VL | Accuracy78.6 | 6 | 4mo ago | |
| MMLongBench-Doc 32K context slice | Qwen3.5-9B | MMLongBench-Doc (32K) Score45 | 4 | 1mo ago | |
| MMLongBench-Doc 16K context slice | Qwen3.5-9B | Score48.57 | 4 | 1mo ago |
