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

Needle

Benchmarks

Task NameDataset NameSOTA ResultTrend
Stochastic Lipschitz OptimizationNeedle
Simple Regret0.011
9
Candidate-ranking retrievalNeedle 8192-token synthetic probes
Accuracy57.8
8
Showing 2 of 2 rows