List your company

Mathematical Reasoning on PRM800K (test)

80AccuracyHRM
56.80862.82968.8574.871Mar 16, 2025

Evaluation results

MethodLinks
HRMN=16, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0380
HRMN=24, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0380
HRMN=8, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0374.4
HRMN=2, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0372.2
HRMN=4, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0371.1
PRMN=2, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0370
ORMN=4, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0367.7
ORMN=8, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0365.5
ORMN=16, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0365.5
PRMN=4, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0364.4
ORMN=24, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0363.3
ORMN=2, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0362.2
PRMN=8, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0361.1
PRMN=16, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0358.8
PRMN=24, policy_model=Qwen2.5-72B-Math-Instruct, reward_model_training=fine-tuned on manually labeled PRM800K dataset2025.0357.7