{
  "id": 54,
  "url": "https://arxiv.org/abs/2607.10112v1",
  "title": "Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety",
  "summary": "Safety alignment in large language models remains brittle across languages: prompts reliably refused in English can elicit harmful compliance in non-English and low-resource settings. We introduce \\textsc{Minionese}, a multilingual jailbreak benchmark spanning 18 languages, 4 resource tiers, and 4 perturbation types (standard translation, code-switching, transliteration, and translationese), paired with a geometric mechanistic analysis of refusal failure across language tiers. We show that each ",
  "authors": "Chigozirim Ifebi, Brent Kong, Ayushi Mehrotra",
  "category": "research",
  "topics": "regulation,safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-11T04:13:42.000Z",
  "fetched_at": "2026-07-14T14:14:15.665Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/54",
  "original_url": "https://arxiv.org/abs/2607.10112v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}