step-gui-calibrated-rewards

【效率与协作】Train GUI agents through calibrated step-level reasoning anchored to trajectory-level validation. Use trajectory verification rather than step annotation achieving 90% accuracy with 10-100× cost reduction. Implement self-evolving training framework with generation-refinement cycles and verifiable reward signals.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准