List your company

Question Answering on MMLU Professional Law

87.7AccuracyGPT-4
17.81235.95654.172.244Dec 23, 2025

Evaluation results

MethodLinks
GPT-4Category=Original, Student=GPT-42025.1287.7—84.32
BRIDGETeacher=GPT-4, TA=None, Student=Qwen2.5-7B-Instr.2025.1265.78—74.22
BRIDGETeacher=GPT-4, TA=None, Student=Mistral-7B-Instr.-v0.32025.1256.66—57.94
Qwen2.5-7B-Instr.Category=Original, Student=Qwen2.5-7B-Instr.2025.1251.96—55.1
Mistral-7B-Instr.-v0.3Category=Original, Student=Mistral-7B-Instr.-v0.32025.1244.98—42.32
BRIDGETeacher=GPT-4, TA=Mistral-7B-Instr.-v0.3, Student=Pythia-410M2025.1234.22—33.97
BRIDGETeacher=GPT-4, TA=Qwen2.5-7B-Instr., Student=OPT-350M2025.1233.75—33.82
BRIDGETeacher=GPT-4, TA=Qwen2.5-7B-Instr., Student=Bloomz-560M2025.1233.59—32.66
BRIDGETeacher=GPT-4, TA=Mistral-7B-Instr.-v0.3, Student=OPT-350M2025.1231.22—31.14
BRIDGETeacher=GPT-4, TA=Mistral-7B-Instr.-v0.3, Student=Bloomz-560M2025.1230.81—31.64
BRIDGETeacher=GPT-4, TA=Qwen2.5-7B-Instr., Student=Pythia-410M2025.1228.11—33.1
BLKDTeacher=GPT-4, Student=Bloomz-560M2025.1227.7—27.87
BLKDTeacher=GPT-4, Student=OPT-350M2025.1227.7—26.77
ULDTA=Mistral-7B-Instr.-v0.3, Student=Bloomz-560M2025.1227.6—24.8
ULDTA=Mistral-7B-Instr.-v0.3, Student=OPT-350M2025.1226.4—24.9
Bloomz-560MCategory=Original, Student=Bloomz-560M2025.1225.68—24.16
ULDTA=Qwen2.5-7B-Instr., Student=Pythia-410M2025.1225.65—27.98
ULDTA=Qwen2.5-7B-Instr., Student=OPT-350M2025.1225.55—24.75
ULDTA=Mistral-7B-Instr.-v0.3, Student=Pythia-410M2025.1225.5—27.26
Pythia-410MCategory=Original, Student=Pythia-410M2025.1225.1—26.89
ULDTA=Qwen2.5-7B-Instr., Student=Bloomz-560M2025.1224.92—24.74
OPT-350MCategory=Original, Student=OPT-350M2025.1224.9—23.88
BLKDTeacher=GPT-4, Student=Pythia-410M2025.1224.6—26.42
LionTeacher=GPT-4, Student=Bloomz-560M2025.1224.3—24.2
LionTeacher=GPT-4, Student=Pythia-410M2025.1223.2—20.77
LionTeacher=GPT-4, Student=OPT-350M2025.1220.5—21.1