[
  {
    "id": "dl3-a06-1",
    "type": "article",
    "title": "Neural Machine Translation of Rare Words with Subword Units",
    "author": [
      {
        "literal": "Rico Sennrich"
      },
      {
        "literal": "Barry Haddow"
      },
      {
        "literal": "Alexandra Birch"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2015
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/1508.07909",
    "container-title": "arXiv primary author manuscript",
    "note": "Original arXiv bibliographic record and abstract inspected on 2026-10-28; full proof and source experiments not independently audited.",
    "archive": "arXiv",
    "archive_location": "1508.07909"
  },
  {
    "id": "dl3-a06-2",
    "type": "article",
    "title": "Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates",
    "author": [
      {
        "literal": "Taku Kudo"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2018
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/1804.10959",
    "container-title": "arXiv primary author manuscript",
    "note": "Original arXiv bibliographic record and abstract inspected on 2026-10-28; full proof and source experiments not independently audited.",
    "archive": "arXiv",
    "archive_location": "1804.10959"
  },
  {
    "id": "dl3-a06-3",
    "type": "article",
    "title": "SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing",
    "author": [
      {
        "literal": "Taku Kudo"
      },
      {
        "literal": "John Richardson"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2018
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/1808.06226",
    "container-title": "arXiv primary author manuscript",
    "note": "Original arXiv bibliographic record and abstract inspected on 2026-10-28; full proof and source experiments not independently audited.",
    "archive": "arXiv",
    "archive_location": "1808.06226"
  },
  {
    "id": "dl3-a06-4",
    "type": "article",
    "title": "ByT5: Towards a token-free future with pre-trained byte-to-byte models",
    "author": [
      {
        "literal": "Linting Xue"
      },
      {
        "literal": "Aditya Barua"
      },
      {
        "literal": "Noah Constant"
      },
      {
        "literal": "Rami Al-Rfou"
      },
      {
        "literal": "Sharan Narang"
      },
      {
        "literal": "Mihir Kale"
      },
      {
        "literal": "Adam Roberts"
      },
      {
        "literal": "Colin Raffel"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2021
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2105.13626",
    "container-title": "arXiv primary author manuscript",
    "note": "Original arXiv bibliographic record and abstract inspected on 2026-10-28; full proof and source experiments not independently audited.",
    "archive": "arXiv",
    "archive_location": "2105.13626"
  },
  {
    "id": "dl3-a06-5",
    "type": "article",
    "title": "CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation",
    "author": [
      {
        "literal": "Jonathan H. Clark"
      },
      {
        "literal": "Dan Garrette"
      },
      {
        "literal": "Iulia Turc"
      },
      {
        "literal": "John Wieting"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2021
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2103.06874",
    "container-title": "arXiv primary author manuscript",
    "note": "Original arXiv bibliographic record and abstract inspected on 2026-10-28; full proof and source experiments not independently audited.",
    "archive": "arXiv",
    "archive_location": "2103.06874"
  },
  {
    "id": "dl3-a06-6",
    "type": "article",
    "title": "Byte Latent Transformer: Patches Scale Better Than Tokens",
    "author": [
      {
        "literal": "Artidoro Pagnoni"
      },
      {
        "literal": "Ram Pasunuru"
      },
      {
        "literal": "Pedro Rodriguez"
      },
      {
        "literal": "John Nguyen"
      },
      {
        "literal": "Benjamin Muller"
      },
      {
        "literal": "Margaret Li"
      },
      {
        "literal": "Chunting Zhou"
      },
      {
        "literal": "Lili Yu"
      },
      {
        "literal": "Jason Weston"
      },
      {
        "literal": "Luke Zettlemoyer"
      },
      {
        "literal": "Gargi Ghosh"
      },
      {
        "literal": "Mike Lewis"
      },
      {
        "literal": "Ari Holtzman"
      },
      {
        "literal": "Srinivasan Iyer"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2024
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2412.09871",
    "container-title": "arXiv primary author manuscript",
    "note": "Original arXiv bibliographic record and abstract inspected on 2026-10-28; full proof and source experiments not independently audited.",
    "archive": "arXiv",
    "archive_location": "2412.09871"
  }
]