List your company

Debate on ARC-C

78.8AccuracyOPTIMA-iSFT-DPO SC
59.45664.47869.574.522Oct 10, 2024Oct 29, 2024Nov 17, 2024Dec 6, 2024Dec 25, 2024Jan 13, 2025Feb 2, 2025

Evaluation results

MethodLinks
OPTIMA-iSFT-DPO SCoptimization_strategy=iSFT-DPO, self_consistency=true2024.1078.81,036.1
DITS-iSFT-DPO2025.0277.6—
OPTIMA-iSFT SCoptimization_strategy=iSFT, self_consistency=true2024.1077.21,148.7
OPTIMA-iSFT-DPOoptimization_strategy=iSFT-DPO2024.1077.188
Optima-iSFT-DPO2025.0277.1—
OPTIMA-iDPO SCoptimization_strategy=iDPO, self_consistency=true2024.10771,091.2
SCn=82024.1075.61,116.7
OPTIMA-iDPOoptimization_strategy=iDPO2024.1074.597.8
Optima-iDPO2025.0274.5—
OPTIMA-iSFToptimization_strategy=iSFT2024.1074.192.2
Optima-iSFT2025.0274.1—
MAD2024.1071.4478
MAD2025.0271.4—
CoT2024.1065.2138.9
CoT2025.0265.2—
AutoForm2024.1060.2221.2
AutoForm2025.0260.2—