CL-bench

dataset
Tencent
Benchmark for evaluating the "context learning" abilities of long-context models.
Paper (arXiv)GitHubHuggingFace
benchmarkscaling