List your company

Commonsense Reasoning on CSQA (ASRr and ACCc)

92.6ASRrCFT
-2.955221.852446.6671.4676Apr 12, 2026

Evaluation results

MethodLinks
CFTDefense scheme=CFT, Backbone=GPT-OSS-20B2026.0492.689.6
No defenseDefense scheme=No defense, Backbone=GPT-OSS-20B2026.0490.488.1
ONIONDefense scheme=ONION, Backbone=GPT-OSS-20B2026.0464.686.5
CoSDefense scheme=CoS, Backbone=GPT-OSS-20B2026.0464.284.3
ShuffleDefense scheme=Shuffle, Backbone=GPT-OSS-20B2026.0438.275.3
Shuffle++Defense scheme=Shuffle++, Backbone=GPT-OSS-20B2026.043.846.3
Critical-CoTDefense scheme=Ours, Backbone=GPT-OSS-20B2026.040.7286.8