Spaces:
Running
Running
Commit History
-
Add Loong benchmark eval viewer (1600 examples, 1821 docs)
Verified signatureStatic viewer with a single Eval tab: question, instruction, lazy-loaded supporting documents (financial/legal/paper), and reveal-able gold answer. Sidebar search + domain/task/length/language filters. Shards built by scripts/build_loong.py. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>128d7e5