Zero-shot Reasoning on Multiple Reasoning Datasets Combined benchmark leaderboard · SOTA2 Research

Zero-shot Reasoning on Multiple Reasoning Datasets Combined

42.62Average Score 0PLDRv51-SOC-110M-5

Evaluation Results

Method	Links
PLDRv51-SOC-110M-52026.03		42.62	39.81	40.21
GPT-Neo-125M2026.03		41.91	39.22	40.02
PLDRv51-SOC-110M-22026.03		41.57	38.63	39.35
PLDRv51-SOC-110M-42026.03		41.46	38.61	39.33
PLDRv51-SOC-110M-32026.03		41.21	38.43	39.14
PLDRv51-SOC-110M-12026.03		40.95	38.38	39.16
ABL-SOC-110M-12026.03		39.18	36.79	37.77
ABL-SOC-110M-32026.03		36.32	35.15	36.81
ABL-SOC-110M-22026.03		36.03	34.44	36.44
SUB-SOC-110M-12026.03		35.49	34.24	36.08
SUB-SOC-110M-22026.03		35.47	34.01	35.77