Fudan and Tongyi introduce ToolCUA for GUI-Tool path selection in agents

别光给Agent加Tool了，它根本选不明白！复旦×通义提出全新CUA训练范式

Fudan University and Tongyi Lab introduced ToolCUA-8B, which reaches 46.85% accuracy on OSWorld-MCP after training with about 4k synthetic tools and 180k interleaved GUI-Tool trajectory steps.

Why it matters: HKR-H/K/R all pass: the tool-selection failure hook is concrete, with OSWorld-MCP 46.85% and 180k steps. It stays in the 78–84 band because this is a research release, not a major model or product launch.