Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Multi-type Hallucination Evaluation on MHumanEval

21.9Object Hallucination Rate

RLHF-V

20.91227.58134.2540.919Dec 1, 2023
Updated 4d ago

Evaluation Results

MethodLinks
2023.12
21.97.514.455.5
2023.12
22.612.31145.9
30.815.117.163.7
2023.12
30.817.817.161
2023.12
33.616.42674.7
34.916.415.861
2023.12
37.717.818.572.6
2023.12
43.211.619.282.9
2023.12
46.621.219.980.8