Realism assessment on 200-sample
3.06Visual IntegrationMIRAGE
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| MIRAGEevaluation_type=Human (mean of 3 raters; 1–5)2026.05 | 3.06 | 3.21 | 2.77 | 3.03 | 3.02 | — | — | |
| AgentHazardevaluation_type=Human (mean of 3 raters; 1–5)2026.05 | 2.59 | 2.68 | 2.28 | 2.52 | 2.52 | — | — |