| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Visual Reasoning | HalluBench | Accuracy71.85 | 24 | |
| General-domain Reasoning | HalluBench | Accuracy71 | 13 | |
| Alignment | HalluBench | Accuracy63 | 10 | |
| Hallucination Evaluation | HALLUBENCH | Accuracy71.1 | 8 | |
| Object Hallucination Evaluation | HalluBench (test) | Accuracy76.6 | 6 | |
| Math Reasoning | Hallubench | Accuracy@168.45 | 5 |