[NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
项目介绍
[NeurIPS 2024] OSWorld:针对真实计算机环境中开放性任务的多模态智能体基准测试【此简介由AI生成】
Apache-2.0 Python1.46 K提交数agentartificial-intelligencebenchmarkclicode-generationguilanguage-modellarge-action-modelllmmultimodalnatural-language-processingreinforcement-learningrpavlm
定制我的领域