{
  "abstract": "We introduce a new tool, Express, for converting a non-causal attention approximation into a causal approximation with matching approximation guarantees. When combined with the state-of-the-art Thinformer approximation, Express improves upon the best known causal attention guarantees, delivering $\\log^{3/2}(n)/s$ approximation error with only $O(s)$ memory and $O(s^2 \\log^2(n))$ compression overhead for a sequence of length $n$. We pair these developments with an efficient I/O-aware Triton implementation, demonstrate substantial speedups over FlashAttention 2, and use Express to overcome four resource bottlenecks in the language modeling pipeline: long-context prefill, KV cache compression, long-form memory-constrained decoding, and long-form compute-constrained decoding.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2606.10944",
  "author_total": 4,
  "authors": [
    {
      "name": "Albert Gong"
    },
    {
      "name": "Annabelle Michael Carrell"
    },
    {
      "name": "Raaz Dwivedi"
    },
    {
      "name": "Lester Mackey"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/crttbp7xkdc4e.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2606.10944",
  "fulltext_url": null,
  "id": "crttbp7xkdc4e",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2606.10944",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.DS"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "math.ST"
    ],
    [
      "arxiv",
      "stat.ME"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "arxiv",
      "stat.TH"
    ],
    [
      "openalex",
      "1702|Topic Modeling"
    ]
  ],
  "openalex": "W7164194438",
  "pdf_url": "https://arxiv.org/pdf/2606.10944",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-29",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/crttbp7xkdc4e.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:36:35Z",
      "native_id": "2606.10944",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2606.10944"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:math.ST",
      "subfield": "statistics"
    }
  ],
  "title": "Express Language Modeling",
  "title_text": "Express Language Modeling",
  "type": "preprint",
  "updated_at": "2026-11-01T12:21:00Z",
  "url": "https://research.draconitic.com/paper/crttbp7xkdc4e",
  "venue": "arXiv",
  "versions": []
}