Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Science Reasoning on ARC Challenge (test)
Loading...
84.04
Accuracy
Genius
66.568
71.104
75.64
80.176
Apr 11, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Genius
Training Corpus=Magpie...
2025.04
84.04
SPIN
Training Corpus=Magpie...
2025.04
83.96
Genius
Training Corpus=OpenHe...
2025.04
83.19
CoH
Training Corpus=Magpie...
2025.04
82.51
Self-Rewarding
Training Corpus=Magpie...
2025.04
82.17
Self-Rewarding
Training Corpus=OpenHe...
2025.04
81.66
CoH
Training Corpus=OpenHe...
2025.04
81.48
SCPO
Training Corpus=OpenHe...
2025.04
79.52
SCPO
Training Corpus=Magpie...
2025.04
78.92
LLaMA3.1-8B-Instruct
Reasoning Strategy=CoT...
2025.04
78.33
SFT
Training Corpus=Magpie...
2025.04
74.06
SPIN
Training Corpus=OpenHe...
2025.04
71.76
SFT
Training Corpus=OpenHe...
2025.04
69.37
STaR
Training Corpus=OpenHe...
2025.04
68.43
STaR
Training Corpus=Magpie...
2025.04
67.24
Feedback
Search any
task
Search any
task