{
  "absUrl": "https://arxiv.org/abs/2609.23797",
  "arxivId": "2609.23797",
  "arxivVersion": null,
  "arxivVersionedId": null,
  "authors": [
    {
      "affiliations": [
        "The University of Hong Kong"
      ],
      "name": "Shuyang Xu"
    },
    {
      "affiliations": [
        "The University of Hong Kong",
        "Massachusetts Institute of Technology"
      ],
      "name": "Zhiyang Dou"
    },
    {
      "affiliations": [
        "University of Pennsylvania"
      ],
      "name": "Yiduo Hao"
    },
    {
      "affiliations": [
        "Brown University"
      ],
      "name": "Zekun Li"
    },
    {
      "affiliations": [
        "The University of Hong Kong"
      ],
      "name": "Liang Pan"
    },
    {
      "affiliations": [
        "Shanghai AI Lab"
      ],
      "name": "Jingbo Wang"
    },
    {
      "affiliations": [
        "Macau University of Science and Technology"
      ],
      "name": "Cheng Lin"
    },
    {
      "affiliations": [
        "Hong Kong University of Science and Technology"
      ],
      "name": "Yuan Liu"
    },
    {
      "affiliations": [
        "Texas A&M University"
      ],
      "name": "Wenping Wang"
    },
    {
      "affiliations": [
        "University of Pennsylvania"
      ],
      "name": "Mingmin Zhao"
    },
    {
      "affiliations": [
        "The University of Hong Kong"
      ],
      "name": "Taku Komura"
    }
  ],
  "contract": "researcher-sidecars-v1",
  "id": "2609.23797",
  "originalTitle": "MoSAT: Human Motion Generation from Spatial Audio and Textual Description",
  "pdfUrl": "https://arxiv.org/pdf/2609.23797.pdf",
  "schemaVersion": 1,
  "type": "preprint"
}