Metraipolis › archive › dc_7221ad995f49
Risk and Refusal: Mechanisms of LLM Responses to Suicidal Ideation — arXiv:2610.22715
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.22715
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T08:44:02.432Z
- SHA-256 of the archived text
96fa8d83f742d1a7f371bdda48ac755df51b196663bb3f202025ab02d901a9b3- Stable URL of this page
https://metraipolis.com/documento/dc_7221ad995f49
How to cite
Risk and Refusal: Mechanisms of LLM Responses to Suicidal Ideation — arXiv:2610.22715 — arXiv (en), https://arxiv.org/abs/2610.22715 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7221ad995f49 (Metraipolis archive dc_7221ad995f49, sha256 96fa8d83f742d1a7f371bdda48ac755df51b196663bb3f202025ab02d901a9b3)
BibTeX
@misc{dc7221ad995f49,
title = {Risk and Refusal: Mechanisms of LLM Responses to Suicidal Ideation — arXiv:2610.22715},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.22715},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7221ad995f49; sha256 96fa8d83f742d1a7f371bdda48ac755df51b196663bb3f202025ab02d901a9b3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22715. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Refusal is a common safety mechanism in large language models (LLMs), yet in contexts of suicidal ideation (SI), refusing to engage can itself constitute a safety failure. We evaluate how eight instruction-tuned LLMs respond to SI and self-harm disclosures across three datasets, varying whether models are informed that a disclosure expresses SI and whether they receive a clinician-developed response protocol, and complement this with mechanistic analyses in six open-weight models. Our results reveal a nuanced relationship between refusal and response-level risk. At baseline, elevated risk in model responses is driven primarily by inattention to the disclosure, alongside refusal to engage. Mechanistic analyses suggest that models may refuse SI disclosures through the same direction that mediates their refusals of harmful requests. However, informing models of the SI context, or suppressing refusal directly, increases engagement without reliably lowering response risk, and may raise it. In contrast, the clinician-guided protocol consistently reduces response-level risk while also reducing refusal. Together, these analyses and ablations motivate the need for models that refuse only when warranted and engage in interactionally safe ways when users disclose SI. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Human-Computer Interaction (cs.HC) ; Computation and Language (cs.CL); Computers and Society (cs.CY); Social and Information Networks (cs.SI). Pagina del lavoro: https://arxiv.org/abs/2610.22715 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T08:39:24.894Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).