{
  "id": 4103,
  "url": "https://arxiv.org/abs/2605.18610v1",
  "title": "CATA: Continual Machine Unlearning via Conflict-Averse Task Arithmetic",
  "summary": "Vision-language models (VLMs) have shown remarkable ability in aligning visual and textual representations, enabling a wide range of multimodal applications. However, their large-scale training data inevitably raises concerns about privacy, copyright, and undesirable content, creating a strong need for machine unlearning. While existing studies mainly focus on single-shot unlearning, practical VLM deployment often involves sequential removal requests over time, giving rise to continual machine u",
  "authors": "Shen Lin, Junhao Dong, Rongjie Chen, Xiaoyu Zhang, Li Xu, Xiaofeng Chen",
  "category": "research",
  "topics": "privacy-surveillance,copyright-ip",
  "orgs": null,
  "regions": null,
  "published_at": "2026-05-18T16:21:41.000Z",
  "fetched_at": "2026-07-14T16:30:45.938Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/4103",
  "original_url": "https://arxiv.org/abs/2605.18610v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}