List your company

Factuality Detection on TriviaQA, HotpotQA, and CSQA

72.9Average AUROCDynHD
57.61261.58165.5569.519Mar 17, 2026

Evaluation results

MethodLinks
DynHDMethod Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0372.9
TraceDetMethod Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0366.3
TSVMethod Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0361.3
CCSMethod Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0358.2