List your company
ResearchFormerly WizWand

Question Answering on TriviaQA (TQA) (test)

79.8Exact MatchATLAS
40.2850.5460.871.06Dec 15, 2022Apr 3, 2023Jul 22, 2023Nov 9, 2023Feb 26, 2024Jun 15, 2024Oct 3, 2024
Updated 8mo ago

Evaluation results

MethodLinks
ATLAS2022.1279.8
CriSPOModel=Claude Sonnet, Shots=64-shot2024.1072.1
FiDOConfig=L-XXL2022.1270.7
CriSPOModel=Claude Sonnet, Shots=0-shot2024.1070.6
FiD-LSource=ours2022.1268.2
FiD-LSource=Izacard and Grave, 20212022.1267.6
CriSPOModel=Claude Instant, Shots=64-shot2024.1067.5
CriSPOModel=Claude Instant, Shots=0-shot2024.1066.3
ManualModel=Claude Sonnet, Shots=64-shot2024.1065
ManualModel=Claude Instant, Shots=64-shot2024.1059.2
ManualModel=Claude Instant, Shots=0-shot2024.1058.6
ManualModel=Claude Sonnet, Shots=0-shot2024.1058.4
RAG2022.1256.8
OPROModel=Claude Instant2024.1053.7
T5-XXL2022.1251.9
OPROModel=Claude Sonnet2024.1041.8