{
  "abstract": "When a language model develops capabilities not traceable to its individual components, when a multi-agent AI system outperforms each agent on its own, or when a human-AI pair succeeds where either fails in isolation, AI researchers describe it as a case of ``synergy'' or ``emergence''. However, these are usually informal assessments not supported by a theoretical framework, making these claims impossible to compare or even meaningfully dispute. Here we argue that \\emph{information decomposition can serve as a unifying formal framework to operationalize assessments about synergy in AI systems}. Information decomposition provides a way to distinguish qualitatively different \\emph{types} of information, including unique information available from individual components, redundant information shared across multiple components, and synergistic information accessible only from parts taken together. This gives claims of emergence a precise, falsifiable criterion: a system's behavior is emergent only if it carries synergistic information about task-relevant outcomes. We apply this lens to attention heads, autonomous agents, and human-AI teams, showing how it organizes existing knowledge across scales and substrates, while also generating cross-level conjectures that siloed fields would miss.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": "Journal reference: 40th Conference on Neural Information Processing Systems (NeurIPS 2026)",
  "arxiv": "2610.28479",
  "author_total": 2,
  "authors": [
    {
      "name": "Christoph Riedl"
    },
    {
      "name": "Fernando E. Rosas"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/dqg87v1f3zmzy.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.28479",
  "fulltext_url": null,
  "id": "dqg87v1f3zmzy",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.28479",
  "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.AI"
    ],
    [
      "arxiv",
      "cs.HC"
    ],
    [
      "arxiv",
      "cs.IT"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "cs.MA"
    ],
    [
      "arxiv",
      "math.IT"
    ],
    [
      "openalex",
      "3109|Complex Systems and Dynamics"
    ]
  ],
  "openalex": "W7223155215",
  "pdf_url": "https://arxiv.org/pdf/2610.28479",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-20",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/dqg87v1f3zmzy.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:34:08Z",
      "native_id": "2610.28479",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.28479"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "openalex:3109|Complex Systems and Dynamics",
      "subfield": "statistical-nonlinear"
    }
  ],
  "title": "Claims of emergence in AI should be grounded in information decomposition",
  "title_text": "Claims of emergence in AI should be grounded in information decomposition",
  "type": "preprint",
  "updated_at": "2026-10-30T12:34:08Z",
  "url": "https://research.draconitic.com/paper/dqg87v1f3zmzy",
  "venue": "arXiv",
  "versions": []
}