Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Role-playing Instruction Following on RoleBench English Role Generalization
Loading...
64.5
Win Rate (GPT-4)
RoleLLaMA
6.572
21.611
36.65
51.689
Oct 1, 2023
Win Rate (GPT-4)
Win Rate (Human)
Updated 4d ago
Evaluation Results
Method
Method
Links
Win Rate (GPT-4)
Win Rate (Human)
RoleLLaMA
Parameters=7B
2023.10
64.5
56.1
Vicuna
2023.10
31
32.4
Alpaca
2023.10
12
28.2
ChatPLUG
2023.10
8.8
12
Feedback
Search any
task
Search any
task