{
  "id": 13060,
  "url": "https://arxiv.org/abs/2607.21300v1",
  "title": "Unlearning Under Imbalance: Benchmarking Fairness in Multimodal LLM Unlearning",
  "summary": "Machine unlearning has emerged as a tool for removing personal data from trained models to comply with recent AI regulations. To evaluate unlearning effectiveness in multimodal large language models (MLLMs), prior works fine-tune models on fictitious identities, simulating unlearning requests on subsets of these IDs, which are typically uniformly distributed. However, in realistic scenarios, people from different demographic groups may request to be unlearned at different frequencies, potentiall",
  "authors": "Lorenzo Orsingher, Thomas De Min, Massimiliano Mancini, Davide Talon, Elisa Ricci",
  "category": "research",
  "topics": "bias-fairness,regulation",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-23T13:24:34.000Z",
  "fetched_at": "2026-07-25T05:10:48.796Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/13060",
  "original_url": "https://arxiv.org/abs/2607.21300v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}