Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Knowledge Question Answering on HLE
Loading...
4.36
Accuracy
KTAE
2.5712
3.0356
3.5
3.9644
Aug 5, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
KTAE
Model Scale=Qwen3-1.7B
2025.08
4.36
VRPO
Model Scale=Qwen3-1.7B
2025.08
4.22
GRPO
Model Scale=Qwen3-1.7B
2025.08
3.94
PPO
Model Scale=Qwen3-1.7B
2025.08
3.94
λ-GRPO
Model Scale=Qwen3-1.7B
2025.08
3.89
VRPO
Model Scale=Qwen3-8B
2025.08
3.89
λ-GRPO
Model Scale=Qwen3-8B
2025.08
3.57
Reinforce++
Model Scale=Qwen3-1.7B
2025.08
3.52
PPO
Model Scale=Qwen3-8B
2025.08
3.29
Reinforce++
Model Scale=Qwen3-8B
2025.08
3.29
GRPO
Model Scale=Qwen3-8B
2025.08
3.24
Dr.GRPO
Model Scale=Qwen3-1.7B
2025.08
3.1
Dr.GRPO
Model Scale=Qwen3-8B
2025.08
3.1
Base
Model Scale=Qwen3-8B
2025.08
2.89
Base
Model Scale=Qwen3-1.7B
2025.08
2.87
KTAE
Model Scale=Qwen3-8B
2025.08
2.64
Feedback
Search any
task
Search any
task