General QA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
General QA NQ, TriviaQA, PopQA (test)
51.3Overall Average Score
49
General QA NQ, TriviaQA, PopQA
58.2NQ Accuracy
40
General QA AGIEval ARC MMLU
81.26Accuracy
8
GENERAL QA MMLU+MMLU-PRO+GSM8K
89.1ROC-AUC
3