Metraipolis › archive › dc_99b3d1ef40cc

Policy Over Purpose: Auditing LLM Content Moderation of Reclaimed Slur Usage — arXiv:2611.10327

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.10327
License (as archived)
CC0-1.0
Archived
2026-11-10T18:44:25.986Z
SHA-256 of the archived text
36a8c581ce051123444d94884707fbc2188c69c77695722e54b83514ab6cd635
Stable URL of this page
https://metraipolis.com/documento/dc_99b3d1ef40cc

How to cite

Policy Over Purpose: Auditing LLM Content Moderation of Reclaimed Slur Usage — arXiv:2611.10327 — arXiv (en), https://arxiv.org/abs/2611.10327 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_99b3d1ef40cc (Metraipolis archive dc_99b3d1ef40cc, sha256 36a8c581ce051123444d94884707fbc2188c69c77695722e54b83514ab6cd635)
BibTeX
@misc{dc99b3d1ef40cc,
  title = {Policy Over Purpose: Auditing LLM Content Moderation of Reclaimed Slur Usage — arXiv:2611.10327},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.10327},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_99b3d1ef40cc; sha256 36a8c581ce051123444d94884707fbc2188c69c77695722e54b83514ab6cd635},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.10327. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As Perspective API is deprecated, platforms are turning to large language models for content moderation, encoding policy as natural language that the model must interpret at inference time. Reclaimed slurs test this design directly. The same term can express in-group pride or target the community that reclaimed it, and misreading it silences the speech moderation is meant to protect. We audit four general-purpose LLMs and four moderation-specific systems, with Perspective API as a legacy baseline, on IYKYK, a corpus of reclaimed and derogatory usage of three slurs annotated by community members. Across the policy, decision, reasoning, disparity, and steering layers of this audit, we find that whether a system was purpose-built for moderation explains little about how it handles reclaimed language. No system reliably agrees with community judgment, and all fail most on language reclaimed within Black communities. What separates systems is instead how their policy is represented. Only systems whose policy is expressed in the prompt can be steered at all, and even then, explicit definitions change how models justify a decision more than how they make it. We argue that policy transparency determines whether moderation failures can be corrected, but that correcting them requires contextual judgment that definitions supplied at inference time cannot provide.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.10327 — 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 2026-11-10T18:37:52.423Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).