List your company
ResearchFormerly WizWand

Text Classification on SST-2 20%S (test)

97.03AccuracyBase (Clean)
86.00688.86891.7394.592Oct 11, 2025
Updated 8mo ago

Evaluation results

MethodLinks
Base (Clean)backbone=Llama3.1-8B-Instruct, training_data=ground truth2025.1097.03
Delorabackbone=Llama3.1-8B-Instruct2025.1096.5
NoiseALbackbone=Llama3.1-8B-Instruct2025.1091.9
LLM-basemodel=GPT-4o2025.1091.51
LAFTbackbone=Llama3.1-8B-Instruct2025.1088.07
SelfMixbackbone=Llama3.1-8B-Instruct2025.1087.58
SENTbackbone=Llama3.1-8B-Instruct2025.1087.46
Co-Teachingbackbone=Llama3.1-8B-Instruct2025.1087.29
CleaRbackbone=Llama3.1-8B-Instruct2025.1087.21
Basebackbone=Llama3.1-8B-Instruct2025.1086.43