[
  {
    "id": "lit_2404_07972",
    "type": "article",
    "title": "OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments",
    "author": [
      {
        "given": "Tianbao",
        "family": "Xie"
      },
      {
        "given": "Danyang",
        "family": "Zhang"
      },
      {
        "given": "Jixuan",
        "family": "Chen"
      },
      {
        "given": "Xiaochuan",
        "family": "Li"
      },
      {
        "given": "Siheng",
        "family": "Zhao"
      },
      {
        "given": "Ruisheng",
        "family": "Cao"
      },
      {
        "given": "Jing Hua",
        "family": "Toh"
      },
      {
        "given": "Zhoujun",
        "family": "Cheng"
      },
      {
        "given": "Dongchan",
        "family": "Shin"
      },
      {
        "given": "Fangyu",
        "family": "Lei"
      },
      {
        "given": "Yitao",
        "family": "Liu"
      },
      {
        "given": "Yiheng",
        "family": "Xu"
      },
      {
        "given": "Shuyan",
        "family": "Zhou"
      },
      {
        "given": "Silvio",
        "family": "Savarese"
      },
      {
        "given": "Caiming",
        "family": "Xiong"
      },
      {
        "given": "Victor",
        "family": "Zhong"
      },
      {
        "given": "Tao",
        "family": "Yu"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2024,
          4,
          11
        ]
      ]
    },
    "genre": "Preprint",
    "archive": "arXiv",
    "number": "arXiv:2404.07972",
    "URL": "https://arxiv.org/abs/2404.07972",
    "custom": {
      "record_url": "https://neural-os.com/reading-list/papers/2404.07972/",
      "first_public": "2024-04-11",
      "collection": "core",
      "citation_basis": "ArXiv preprint, using its initial submission date."
    }
  }
]