{
  "id": "saac",
  "title": "Near-imperceptible Neural Linguistic Steganography via Self-Adjusting Arithmetic Coding",
  "authors": [
    "Jiaming Shen",
    "Heng Ji",
    "Jiawei Han"
  ],
  "author_parts": [
    {
      "given": "Jiaming",
      "family": "Shen"
    },
    {
      "given": "Heng",
      "family": "Ji"
    },
    {
      "given": "Jiawei",
      "family": "Han"
    }
  ],
  "year": 2020,
  "venue_date": "2020-11",
  "venue": "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
  "venue_short": "EMNLP 2020",
  "publication_type": "conference",
  "publisher": "Association for Computational Linguistics",
  "pages": "303–313",
  "doi": "10.18653/v1/2020.emnlp-main.22",
  "arxiv_id": "2010.00677",
  "relation": "direct-extension",
  "categories": [
    "Arithmetic coding",
    "Statistical security"
  ],
  "summary": "Introduces self-adjusting arithmetic coding, choosing the candidate-set size at each generation step to control divergence from the language model. It supplies an imperceptibility analysis and evaluates capacity, divergence, and human judgments across multiple datasets.",
  "relationship": "The introduction explicitly states that SAAC builds on Ziegler, Deng, and Rush's 2019 arithmetic-coding method, replacing a fixed truncation size with an adaptive one.",
  "limitation": "Its distributional guarantee is relative to the underlying language model; choosing the truncation threshold still trades embedding behavior and computation against divergence.",
  "sources": [
    {
      "url": "https://aclanthology.org/2020.emnlp-main.22/",
      "evidence": "Published bibliographic metadata and DOI."
    },
    {
      "url": "https://arxiv.org/abs/2010.00677",
      "evidence": "First arXiv submission October 1, 2020."
    },
    {
      "url": "https://aclanthology.org/2020.emnlp-main.22.pdf",
      "evidence": "Introduction explicitly identifies extension of 2019 method; sections 2–3 define relative imperceptibility/adaptive K; page 1 links code."
    },
    {
      "url": "https://github.com/mickeysjm/StegaText",
      "evidence": "Official code repository linked by the paper."
    }
  ],
  "arxiv_url": "https://arxiv.org/abs/2010.00677",
  "earliest_date": "2020-10-01",
  "earliest_date_label": "First arXiv submission",
  "url": "https://aclanthology.org/2020.emnlp-main.22/",
  "pdf": "https://aclanthology.org/2020.emnlp-main.22.pdf",
  "code": "https://github.com/mickeysjm/StegaText",
  "category": "direct-extension",
  "contribution": "Introduces self-adjusting arithmetic coding, choosing the candidate-set size at each generation step to control divergence from the language model. It supplies an imperceptibility analysis and evaluates capacity, divergence, and human judgments across multiple datasets.",
  "canonical_url": "https://steganography.live/reading-list/papers/saac/"
}