Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Knowledge Evaluation on HLE
Loading...
3.94
Accuracy
VRPO
2.848
3.1315
3.415
3.6985
Aug 5, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
VRPO
Model Backbone=Qwen2.5...
2025.08
3.94
VRPO
Model Backbone=Qwen3-8...
2025.08
3.89
PPO
Model Backbone=Qwen2.5...
2025.08
3.43
Reinforce++
Model Backbone=Qwen2.5...
2025.08
3.34
Dr.GRPO
Model Backbone=Qwen2.5...
2025.08
3.34
Base
Model Backbone=Qwen2.5...
2025.08
3.29
PPO
Model Backbone=Qwen3-8...
2025.08
3.29
Reinforce++
Model Backbone=Qwen3-8...
2025.08
3.29
GRPO
Model Backbone=Qwen3-8...
2025.08
3.24
Dr.GRPO
Model Backbone=Qwen3-8...
2025.08
3.1
GRPO
Model Backbone=Qwen2.5...
2025.08
2.97
Base
Model Backbone=Qwen3-8...
2025.08
2.89
Feedback
Search any
task
Search any
task
