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

DATE-LM

Benchmarks

Task NameDataset NameSOTA ResultTrend
Language Model EvaluationDATE-LM MMLU, GSM8K, BBH
MMLU Accuracy62.07
7
LLM fine-tuning data selectionDATE-LM MMLU, GSM8K, BBH official (test)
MMLU Accuracy61.87
3
Showing 2 of 2 rows