{
  "id": "2611.04137",
  "title": "LoopKV: Loop-Aware KV Cache Compression for Serving Looped Transformers",
  "authors": "Tianle Zhou, Yang Zhang, Tianyu Wu, Zhuohan Wang, William Nixon, Juncheng Yang",
  "first_public": "2026-11-03",
  "added_at": "2026-11-09",
  "venue": "arXiv",
  "arxiv": "2611.04137",
  "paper_url": "https://arxiv.org/abs/2611.04137",
  "html_url": "https://arxiv.org/html/2611.04137v1",
  "lens": [
    "recurrent"
  ],
  "representation": [
    "cache-memory"
  ],
  "computation": [
    "recurrent"
  ],
  "training": [
    "inference-only"
  ],
  "questions": [
    "efficiency",
    "architecture"
  ],
  "evidence": [
    "empirical"
  ],
  "modality": [
    "language"
  ],
  "summary": "Allocates cache precision separately for each loop and layer, shares selected caches across loops, and reads low-attention values at lower precision to reduce the memory and latency of looped language models.",
  "provenance": "Primary-source literature review, 2026-11-09",
  "author_names": [
    "Tianle Zhou",
    "Yang Zhang",
    "Tianyu Wu",
    "Zhuohan Wang",
    "William Nixon",
    "Juncheng Yang"
  ],
  "record_url": "https://latent-implicit-thinking.github.io/papers/2611.04137/",
  "legacy_record_url": "https://latent-implicit-thinking.github.io/papers/#paper-2611-04137",
  "citation_date": "2026-11-03",
  "json_url": "https://latent-implicit-thinking.github.io/papers/2611.04137/record.json",
  "markdown_url": "https://latent-implicit-thinking.github.io/papers/2611.04137/index.md",
  "bibtex_url": "https://latent-implicit-thinking.github.io/papers/2611.04137/citation.bib",
  "csl_url": "https://latent-implicit-thinking.github.io/papers/2611.04137/citation.csl.json",
  "pdf_url": "https://arxiv.org/pdf/2611.04137"
}