List your company

Commonsense Question Answering on CSQA (AUROC)

77.62AUROCCALIDIST
49.425656.745364.06571.3847Jun 4, 2026

Evaluation results

MethodLinks
CALIDISTLLM Backbone=Qwen-3, Style=Assertion-style2026.0677.62
VanillaLLM Backbone=Qwen-32026.0670.01
FSDLLM Backbone=Qwen-32026.0654.45
EntropyLLM Backbone=Qwen-32026.0650.56
ConsistencyLLM Backbone=Qwen-32026.0650.51