Share your thoughts, 1 month free Claude Pro on usSee more

Scientific Reasoning on ARC

86.41Score

SGDPO

Updated 1mo ago

Evaluation Results

Method	Links
SGDPO 2025.05		86.41
SamPO 2025.05		86.32
DPO 2025.05		86.28
BCO 2025.05		86.2
NCA 2025.05		86.07
TDPO 2025.05		85.89
SFT 2025.05		85.82
IPO 2025.05		85.35
Llama3-8B-Instruct 2024.10		83.1
Medical SFT 2024.10		82.6
SUDOLM Medical w/ key 2024.10		82.3
SUDOLM Medical w/o key 2024.10		82.1
SUDOLM TOFU w/ key 2024.10		80.1
SUDOLM TOFU w/o key 2024.10		79.7
TOFU SFT 2024.10		79.5
Medical SFT 2024.10		51.3
SUDOLM Medical w/ key 2024.10		50.6
SUDOLM Medical w/o key 2024.10		50.4
Llama2-13B 2024.10		49.4
SUDOLM TOFU w/ key 2024.10		48.6
SUDOLM TOFU w/o key 2024.10		48.6
TOFU SFT 2024.10		48.2
Llama2-7B 2024.10		46.9
Medical SFT 2024.10		46.2
SUDOLM Medical w/ key 2024.10		45.9
SUDOLM Medical w/o key 2024.10		45.8
TOFU SFT 2024.10		43.1
SUDOLM TOFU w/ key 2024.10		42.8
SUDOLM TOFU w/o key 2024.10		42.8