{
  "id": 19038,
  "url": "https://arxiv.org/abs/2608.12083v1",
  "title": "Faithful, Sufficient and Understandable: Rethinking Graph Counterfactual Explanations via Discrete Diffusion Inversion",
  "summary": "Graph Neural Networks (GNNs) achieve strong predictive performance on graph-structured data across domains such as chemistry, biology, and network analysis, yet they provide no intrinsic explanation of their predictions. This limits their adoption in high-stakes and safety-critical settings. Counterfactual explanations address this by revealing the minimal structural modifications that would change a model's prediction. On graphs, however, such a modification is hard to produce. The search space",
  "authors": "David Bechtoldt, Sidney Bender",
  "category": "research",
  "topics": "biotech",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-12T14:04:49.000Z",
  "fetched_at": "2026-08-13T05:10:37.786Z",
  "source_slug": "x-arxiv-cs-ai",
  "source_name": "arXiv cs.AI",
  "source_homepage": "https://arxiv.org/list/cs.AI/recent",
  "ethics_ai_record_url": "https://ethics.ai/record/19038",
  "original_url": "https://arxiv.org/abs/2608.12083v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}