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

PerLTQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Agent Memory Question AnsweringPerLTQA (test)
BLEU42.68
18
Memory RetrievalPerLTQA CN
ERC93.12
14
Memory RetrievalPerLTQA EN
ERC90.47
14
Long-term dialogue memoryPerLTQA (test)
Accuracy93.14
11
RetrievalPerLTQA
Ra@574.5
1
Proactive Assistant EvaluationPerLTQA Category (test)
Response Frequency15
1
Showing 6 of 6 rows