On ALFWorld, FlowEvo achieves an 82.8% success rate, 23.6 percentage points above the strongest baseline...
正在加载行业信息...
正在加载行业信息...
开源项目
更新于 2026/7/28 17:22:18
An interactive text-based environment benchmark for evaluating agent task-solving capabilities.
On ALFWorld, FlowEvo achieves an 82.8% success rate, 23.6 percentage points above the strongest baseline...
暂无来源链接。