{
  "id": 4290,
  "url": "https://arxiv.org/abs/2605.15520v1",
  "title": "On the Fragility of Data Attribution When Learning Is Distributed",
  "summary": "Data attribution has become an important component of pricing, auditing, and governance in machine learning pipelines, yet most attribution methods implicitly assume that attribution values faithfully reflect participants' contributions. We show that this assumption can fail: a single participant in a standard distributed training workflow can substantially inflate its measured attribution value while preserving global utility. Our attribution-first attack uses latent optimization to inject smal",
  "authors": "Xian Gao, Bo Hui, Min-Te Sun, Wei-Shinn Ku",
  "category": "research",
  "topics": "regulation,transparency",
  "orgs": null,
  "regions": null,
  "published_at": "2026-05-15T01:34:55.000Z",
  "fetched_at": "2026-07-14T16:30:54.919Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/4290",
  "original_url": "https://arxiv.org/abs/2605.15520v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}