[
  {
    "id": "dl3-b-deduplication-training-measure-1",
    "type": "paper-conference",
    "title": "Deduplicating Training Data Makes Language Models Better",
    "author": [
      {
        "literal": "Katherine Lee"
      },
      {
        "literal": "Daphne Ippolito"
      },
      {
        "literal": "Andrew Nystrom"
      },
      {
        "literal": "Chiyuan Zhang"
      },
      {
        "literal": "Douglas Eck"
      },
      {
        "literal": "Chris Callison-Burch"
      },
      {
        "literal": "Nicholas Carlini"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2022
        ]
      ]
    },
    "URL": "https://aclanthology.org/2022.acl-long.577/",
    "container-title": "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    "note": "Primary publisher or author arXiv abstract and metadata inspected via web on 2026-10-28; no independent proof or benchmark replication.",
    "page": "8424-8445",
    "DOI": "10.18653/v1/2022.acl-long.577"
  },
  {
    "id": "dl3-b-deduplication-training-measure-2",
    "type": "article",
    "title": "Deduplicating Training Data Mitigates Privacy Risks in Language Models",
    "author": [
      {
        "literal": "Nikhil Kandpal"
      },
      {
        "literal": "Eric Wallace"
      },
      {
        "literal": "Colin Raffel"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2022
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2202.06539",
    "container-title": "arXiv preprint",
    "note": "Primary publisher or author arXiv abstract and metadata inspected via web on 2026-10-28; no independent proof or benchmark replication.",
    "archive": "arXiv",
    "archive_location": "2202.06539"
  },
  {
    "id": "dl3-b-deduplication-training-measure-3",
    "type": "article",
    "title": "SemDeDup: Data-efficient learning at web-scale through semantic deduplication",
    "author": [
      {
        "literal": "Amro Abbas"
      },
      {
        "literal": "Kushal Tirumala"
      },
      {
        "literal": "Dániel Simig"
      },
      {
        "literal": "Surya Ganguli"
      },
      {
        "literal": "Ari S. Morcos"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2023
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2303.09540",
    "container-title": "arXiv preprint",
    "note": "Primary publisher or author arXiv abstract and metadata inspected via web on 2026-10-28; no independent proof or benchmark replication.",
    "archive": "arXiv",
    "archive_location": "2303.09540"
  },
  {
    "id": "dl3-b-deduplication-training-measure-4",
    "type": "article",
    "title": "DataComp: In search of the next generation of multimodal datasets",
    "author": [
      {
        "literal": "Samir Yitzhak Gadre"
      },
      {
        "literal": "Gabriel Ilharco"
      },
      {
        "literal": "Alex Fang"
      },
      {
        "literal": "et al."
      }
    ],
    "issued": {
      "date-parts": [
        [
          2023
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2304.14108",
    "container-title": "arXiv preprint",
    "note": "Primary publisher or author arXiv abstract and metadata inspected via web on 2026-10-28; no independent proof or benchmark replication.",
    "archive": "arXiv",
    "archive_location": "2304.14108"
  },
  {
    "id": "dl3-b-deduplication-training-measure-5",
    "type": "article",
    "title": "Beyond neural scaling laws: beating power law scaling via data pruning",
    "author": [
      {
        "literal": "Ben Sorscher"
      },
      {
        "literal": "Robert Geirhos"
      },
      {
        "literal": "Shashank Shekhar"
      },
      {
        "literal": "Surya Ganguli"
      },
      {
        "literal": "Ari S. Morcos"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2022
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2206.14486",
    "container-title": "arXiv preprint",
    "note": "Primary publisher or author arXiv abstract and metadata inspected via web on 2026-10-28; no independent proof or benchmark replication.",
    "archive": "arXiv",
    "archive_location": "2206.14486"
  },
  {
    "id": "dl3-b-deduplication-training-measure-6",
    "type": "article",
    "title": "Scaling Data-Constrained Language Models",
    "author": [
      {
        "literal": "Niklas Muennighoff"
      },
      {
        "literal": "Alexander M. Rush"
      },
      {
        "literal": "Boaz Barak"
      },
      {
        "literal": "Teven Le Scao"
      },
      {
        "literal": "Aleksandra Piktus"
      },
      {
        "literal": "Nouamane Tazi"
      },
      {
        "literal": "Sampo Pyysalo"
      },
      {
        "literal": "Thomas Wolf"
      },
      {
        "literal": "Colin Raffel"
      }
    ],
    "issued": {
      "date-parts": [
        [
          2023
        ]
      ]
    },
    "URL": "https://arxiv.org/abs/2305.16264",
    "container-title": "arXiv preprint",
    "note": "Primary publisher or author arXiv abstract and metadata inspected via web on 2026-10-28; no independent proof or benchmark replication.",
    "archive": "arXiv",
    "archive_location": "2305.16264"
  }
]