{
  "abstract": "Graph Neural Additive Networks (GNANs) have emerged as effective, interpretable-by-design architectures for graph learning. They achieve transparency by combining per-feature shape functions with simple distance-based reweighting. However, GNANs have three limitations. First, node-wise relevance scores aggregate information from all graph nodes, allowing irrelevant nodes to inherit spurious relevance from relevant ones. Second, topology is represented only through shortest-path distances, so structurally distinct graphs may share identical distance profiles. Third, feature and distance information are combined multiplicatively, entangling feature, distance, and interaction effects in a single rescalable term. We introduce AMIGA, which represents topology through a vocabulary of mined motifs and decomposes each node contribution into three additive channels: feature, structure, and feature-structure interaction. The channels are centered around a graph-balanced training baseline, making the decomposition well-defined and unique under the chosen reference distribution. Across ten graph classification and regression benchmarks, AMIGA outperforms GNAN on most datasets and narrows the gap with standard black-box GNNs. Because each channel is centered and assigned to a single node, AMIGA produces node-local explanations that reconstruct the model output exactly and support interpretation at multiple levels of granularity.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.20662",
  "author_total": 6,
  "authors": [
    {
      "name": "Clara Alvarez-Rodriguez"
    },
    {
      "name": "Steve Azzolin"
    },
    {
      "name": "Vincenzo Marco De Luca"
    },
    {
      "name": "Filippo Maria Bianchi"
    },
    {
      "name": "Andrea Passerini"
    },
    {
      "name": "Antonio Longa"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/3ymetaycn10te.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.20662",
  "fulltext_url": null,
  "id": "3ymetaycn10te",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.20662",
  "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.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Explainable Artificial Intelligence (XAI)"
    ]
  ],
  "openalex": "W7222476780",
  "pdf_url": "https://arxiv.org/pdf/2610.20662",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-14",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/3ymetaycn10te.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:30:00Z",
      "native_id": "2610.20662",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.20662"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Graph Additive Models disentangling feature, topology, and interactions",
  "title_text": "Graph Additive Models disentangling feature, topology, and interactions",
  "type": "preprint",
  "updated_at": "2026-10-30T12:30:00Z",
  "url": "https://research.draconitic.com/paper/3ymetaycn10te",
  "venue": "arXiv",
  "versions": []
}