spire-studio/fedarena

fedarena-arena

FedArena Arena — evaluate user-designed FL attacks/defenses against a standardized benchmark matrix. Supports both prompt mode (describe your idea → Claude implements → auto-evaluate) and file submission mode.

لغة النص الأصلي: الإنجليزية

fedarena-bench

Run FL benchmark experiments from natural language. Parses user intent, plans experiments, runs them via the research runner, and reports comparison results.

لغة النص الأصلي: الإنجليزية