FedMABench

What is FedMABench?

FedMABench is an open-source benchmark for federated training and evaluation of mobile agents, specifically designed for heterogeneous scenarios.

Released

2025-03-07

Evaluates

智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/FedMABench

FedMABench paper, code and dataset

• Paper https://arxiv.org/abs/2503.05143
• Code https://github.com/wwh0411/FedMABench
• Dataset https://huggingface.co/datasets/wwh0411/FedMABench

Which sources cite FedMABench?

• FedMABench

Related 智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• Ada-LEval
• AECBench
• AFQMC
• AgentBoard