List your company

General Language Capability on MMLU, GSM8K, and GPQA

73.6MMLU AccuracyMUSE-D
73.28873.36973.4573.531May 9, 2026

Evaluation results

MethodLinks
MUSE-DBackbone=Qwen2.5-7B-IT2026.0573.682.532.8
SafeMTBackbone=Qwen2.5-7B-IT2026.0573.58132.6
TRACEBackbone=Qwen2.5-7B-IT, variant=w/o lat.2026.0573.583.532
TRACEBackbone=Qwen2.5-7B-IT, variant=ours2026.0573.584.535.6
Qwen2.5-7B-IT2026.0573.386.534.2