Share your thoughts, 1 month free Claude Pro on usSee more

Multi-task Language Evaluation on GSM8K, ARC-C, MedQA, BoolQ, and CoLA Average

79.87Average Accuracy

PROMPT CLASSIFICATION

Updated 25d ago

Evaluation Results

Method	Links
PROMPT CLASSIFICATION 2026.06		79.87
HARD-ROUTED MOR-LORA 2026.06		79.8
MOLE (α = 0.5) 2026.06		78.5
LORAMIXER TOPK=2 NORM. 2026.06		78.33
MOLE (α = 0.1) 2026.06		78.02
LORAMIXER TOPK=1 2026.06		77.74
LORAHUB 2026.06		75.76
PROMPT CLASSIFICATION 2026.06		72.79
HARD-ROUTED MOR-LORA 2026.06		72.07
LORAMIXER TOPK=2 NORM. 2026.06		70.81
MOLE (α = 0.1) 2026.06		68.66
MOLE (α = 0.5) 2026.06		68.5
LORAHUB 2026.06		67.75
LORAMIXER TOPK=1 2026.06		65.67