Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

CrossCult-KIBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Sequential-insert knowledge insertionCrossCult-KIBench 1.0 (test)
Final Relational ROUGE-L85.57
14
Knowledge InsertionCrossCult-KIBench single-insert
Reliability (ROUGE-L)91.12
14
Showing 2 of 2 rows