| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| BLINK | Qwen3-VL-8B-Inst. | Accuracy85.2 | 116 | 18d ago | |
| GQA | ActiveScope | Accuracy73.01 | 96 | 1mo ago | |
| V* | Dynamo (Full) | Accuracy96.03 | 72 | 25d ago | |
| MMVP | GPT-4o | Accuracy86.3 | 67 | 18d ago | |
| V*Bench | Accuracy95.7 | 62 | 3mo ago | ||
| NLVR2 | BEIT-3 | Accuracy92.6 | 58 | 1mo ago | |
| BLINK | Human | Jigsaw Accuracy99 | 57 | 1mo ago | |
| HR-Bench 4K | P2R-4B | Overall Score0.819 | 54 | 23d ago | |
| MMStar | GazeVLM (Ours) | Accuracy69 | 51 | 1mo ago | |
| MMBench | ThinkLite | Accuracy88.7 | 48 | 1mo ago | |
| CV-Bench | Mimo-Embodied | Accuracy88.8 | 47 | 18d ago | |
| NLVR2 (test) | SimVLM_HUGE | Accuracy85.15 | 46 | 3mo ago | |
| Jigsaw | AdaReasoner 7B | Accuracy88.6 | 44 | 1mo ago | |
| HallusionBench | GSPO + DyCo-RL | Accuracy73.2 | 43 | 22d ago | |
| HR-Bench 8K | DeepEyes | Overall Score72.6 | 42 | 2mo ago | |
| HRBench-4K | S1-VL-32B-RL | Accuracy91.38 | 41 | 25d ago | |
| MathVerse | Accuracy61.29 | 40 | 2mo ago | ||
| MM-Vet | UniDFlow | Score82.7 | 40 | 3mo ago | |
| MMMU-Pro | Prod(VF, Contr.) | ECE14.4 | 32 | 3mo ago | |
| VizWiz | Plausibility | ECE0.11 | 32 | 3mo ago | |
| A-OKVQA | Avg(VF, Contr.) | ECE5.4 | 32 | 3mo ago | |
| Commonsense Reasoning | Jaccard Index (J)8 | 30 | 2mo ago | ||
| Quantitative Reasoning | µCRASP | J Score8.44 | 30 | 2mo ago | |
| Physical Reasoning | J Score9.01 | 30 | 2mo ago | ||
| V* | o3 | Overall Score95.7 | 30 | 17d ago |