ARISE: Agent Reasoning with Intrinsic Skill Evolution in Hierarchical Reinforcement Learning
Open paper
Benchmark
Model
Rank
Results
mathematical-reasoning-on-aime24
ARISE + GRPO (Qwen3-4B-Instruct-2507)
#73
Acc: 56.4
mathematical-reasoning-on-aime-2025
ARISE + GRPO (Qwen3-4B-Instruct-2507)
#88
Accuracy: 48.3
Rank counts only results with a code link.