{
  "absUrl": "https://arxiv.org/abs/2609.08390",
  "arxivId": "2609.08390",
  "arxivVersion": null,
  "arxivVersionedId": null,
  "authors": [
    {
      "affiliations": [
        "Integrated Vision and Language Lab, KAIST, South Korea"
      ],
      "name": "Bella Godiva"
    },
    {
      "affiliations": [
        "Integrated Vision and Language Lab, KAIST, South Korea"
      ],
      "name": "Yeonju Kim"
    },
    {
      "affiliations": [
        "Integrated Vision and Language Lab, KAIST, South Korea"
      ],
      "name": "Yong Man Ro"
    }
  ],
  "contract": "researcher-sidecars-v1",
  "id": "2609.08390",
  "originalTitle": "Noise Adaptive Streaming Audio-Visual Speech Token Enhancement for Robust Full-Duplex Spoken Dialogue Models",
  "pdfUrl": "https://arxiv.org/pdf/2609.08390.pdf",
  "schemaVersion": 1,
  "type": "preprint"
}