List your company

Short-form Question Answering on Trivia QA

85.9AUROCRewarding Doubt*
48.56458.25767.9577.643May 29, 2025

Evaluation results

MethodLinks
Rewarding Doubt*Category=Literature SOTA2025.0585.92.2
Self-ConsCategory=Baselines2025.0573.412.2
LOVEC-DPOCategory=Our Methods2025.0571.26.9
LOVEC-GRPOCategory=Our Methods2025.0569.26.3
p(true)Category=Baselines2025.0560.121.1
LOVEC-SFTCategory=Our Methods2025.0556.32
Self-VerbCategory=Baselines2025.055069.3