{"id":"arxiv:2610.07023","kind":"preprint","title":"Beyond Refusal Patterns: Safe-Role Internalization for Robust and Generalizable LLM Safety Alignment","authors":["Jinghao Pang","Jitai Hao","Qiang Huang","Zhaochun Ren","Jun Yu"],"author_count":5,"year":2026,"published":"2026-10-04","venue":null,"cited_by":null,"doi":null,"arxiv_id":"2610.07023","nct_id":null,"source_url":"https://arxiv.org/abs/2610.07023","archive_url":"https://space-kitty.com/papers/arxiv:2610.07023","updated":"2026-10-04","pmid":null,"pmcid":null,"openalex_id":null,"pdf_url":"https://arxiv.org/pdf/2610.07023","abstract":null,"abstract_license":null,"license":null,"is_open_access":false,"retracted":false,"peer_reviewed":false,"categories":["cs.AI","cs.CL","cs.IR"],"details":null,"provenance":"arxiv","attribution":null,"topics":[],"cited_in_dossiers":[]}