{
  "id": 274,
  "url": "https://arxiv.org/abs/2607.03346v1",
  "title": "Efficient Decentralized Multi-task Dataset Valuation via Model Merging",
  "summary": "Accurate and efficient dataset valuation is essential for enabling fair and transparent data marketplaces, especially when multiple contributors provide data for training multi-task models. Most existing valuation methods, however, are limited to single-task settings, overlooking scenarios where a buyer aims to optimize performance across multiple downstream tasks. Moreover, traditional valuation approaches, such as Shapley-based or retraining-based methods, are computationally expensive and poo",
  "authors": "Mohammadsajad Alipour, Mohammad Mohammadi Amiri",
  "category": "research",
  "topics": "transparency,finance-investment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-03T14:01:53.000Z",
  "fetched_at": "2026-07-14T14:14:24.248Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/274",
  "original_url": "https://arxiv.org/abs/2607.03346v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}