{
  "absUrl": "https://arxiv.org/abs/2610.24239v1",
  "arxivId": "2610.24239",
  "arxivVersion": 1,
  "arxivVersionedId": "2610.24239v1",
  "authors": [
    {
      "affiliations": [
        "The University of Texas at Austin"
      ],
      "name": "David Harwath"
    },
    {
      "affiliations": [
        "NVIDIA"
      ],
      "name": "Desh Raj"
    },
    {
      "affiliations": [
        "Johns Hopkins University"
      ],
      "name": "Xavier Velez"
    },
    {
      "affiliations": [
        "Johns Hopkins University",
        "The University of Melbourne"
      ],
      "name": "Sandra Arcos-Holzinger"
    },
    {
      "affiliations": [
        "The University of Texas at Austin"
      ],
      "name": "Willie Jeng"
    },
    {
      "affiliations": [
        "Toyota Technological Institute at Chicago"
      ],
      "name": "Karen Livescu"
    },
    {
      "affiliations": [
        "Johns Hopkins University"
      ],
      "name": "Leibny Paola Garcia-Perera"
    },
    {
      "affiliations": [
        "Johns Hopkins University"
      ],
      "name": "Sanjeev Khudanpur"
    }
  ],
  "contract": "researcher-sidecars-v1",
  "id": "2610.24239v1",
  "originalTitle": "Encoder-Free Multimodal LLMs Learn Implicit Encoders",
  "pdfUrl": "https://arxiv.org/pdf/2610.24239v1.pdf",
  "schemaVersion": 1,
  "type": "preprint"
}