List your company

Factuality Detection on TriviaQA (test)

85.5AUROCDynHD
48.68458.24267.877.358Mar 17, 2026

Evaluation results

MethodLinks
DynHDSource Dataset=HotpotQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0385.5
TraceDetSource Dataset=HotpotQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0373.1
DynHDSource Dataset=CSQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0368.7
TSVSource Dataset=CSQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0365.3
TraceDetSource Dataset=CSQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0361.5
TSVSource Dataset=HotpotQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0358.5
CCSSource Dataset=CSQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0354.5
CCSSource Dataset=HotpotQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0350.1