{
  "id": "2611.05304",
  "title": "LoopVQ: Sub 1-bit KV Cache Quantization for Looped Transformers",
  "authors": "Minsoo Cheong, Sungjoo Yoo",
  "first_public": "2026-11-03",
  "added_at": "2026-11-09",
  "venue": "arXiv",
  "arxiv": "2611.05304",
  "paper_url": "https://arxiv.org/abs/2611.05304",
  "html_url": "https://arxiv.org/html/2611.05304v1",
  "lens": [
    "recurrent"
  ],
  "representation": [
    "cache-memory"
  ],
  "computation": [
    "recurrent"
  ],
  "training": [
    "distillation"
  ],
  "questions": [
    "efficiency",
    "architecture"
  ],
  "evidence": [
    "empirical"
  ],
  "modality": [
    "language"
  ],
  "summary": "Compresses related key–value cache entries across loops with shared vector codes, then adjusts the codebooks to match the original model’s outputs. Implements the compressed cache in a serving system.",
  "provenance": "Primary-source literature review, 2026-11-09",
  "author_names": [
    "Minsoo Cheong",
    "Sungjoo Yoo"
  ],
  "record_url": "https://latent-implicit-thinking.github.io/papers/2611.05304/",
  "legacy_record_url": "https://latent-implicit-thinking.github.io/papers/#paper-2611-05304",
  "citation_date": "2026-11-03",
  "json_url": "https://latent-implicit-thinking.github.io/papers/2611.05304/record.json",
  "markdown_url": "https://latent-implicit-thinking.github.io/papers/2611.05304/index.md",
  "bibtex_url": "https://latent-implicit-thinking.github.io/papers/2611.05304/citation.bib",
  "csl_url": "https://latent-implicit-thinking.github.io/papers/2611.05304/citation.csl.json",
  "pdf_url": "https://arxiv.org/pdf/2611.05304"
}