Metraipolis › archive › dc_09744100fd16
When Agents Learn to Be You: Benchmarking Privacy Leakage, Impersonation Risk, and Defenses in Persona Skills — arXiv:2608.03700
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.03700
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:31:25.767Z
- SHA-256 of the archived text
3009d04faefc2dccc14fef6ac4636824979a9579a789943f301a2acfc5d81a79- Stable URL of this page
https://metraipolis.com/documento/dc_09744100fd16
How to cite
When Agents Learn to Be You: Benchmarking Privacy Leakage, Impersonation Risk, and Defenses in Persona Skills — arXiv:2608.03700 — arXiv (en), https://arxiv.org/abs/2608.03700 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_09744100fd16 (Metraipolis archive dc_09744100fd16, sha256 3009d04faefc2dccc14fef6ac4636824979a9579a789943f301a2acfc5d81a79)
BibTeX
@misc{dc09744100fd16,
title = {When Agents Learn to Be You: Benchmarking Privacy Leakage, Impersonation Risk, and Defenses in Persona Skills — arXiv:2608.03700},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.03700},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_09744100fd16; sha256 3009d04faefc2dccc14fef6ac4636824979a9579a789943f301a2acfc5d81a79},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03700. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Persona skills distill personal interaction histories into portable and executable artifacts for downstream agents. While enabling flexible personalization, this process concentrates fragmented personal signals, amplifies their impact through reuse, and challenges defenses designed for individual records or retrieval-based memory. To systematically investigate the safety of the persona-skill pipeline, we introduce AntiSkillBench, an end-to-end benchmark for evaluating risks and defenses across the persona-skill pipeline. It comprises: (i) a dataset of 7,500 persona-grounded dialogue traces, constructed from 50 behaviorally rich profiles spanning diverse task scenarios; (ii) an evaluation suite that measures skill-level privacy leakage and agent-level attribute disclosure and behavioral impersonation across three skill-distillation strategies; and (iii) a defense evaluation covering four configurations across online and post-hoc interventions, including active risk suppression and passive provenance protection. Experiments across three frontier agents show that persona-skill risks persist across agent backbones and distillation protocols, extending from explicit attributes to communication styles and personality traits. Existing defenses exhibit limited and distillation-dependent effectiveness, failing to generalize across risk and distillation strategies. These results highlight AntiSkillBench as a challenging benchmark for developing privacy-preserving and authenticity-aware persona skills. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.CL). Categorie arXiv: Cryptography and Security (cs.CR) ; Computation and Language (cs.CL); Computers and Society (cs.CY). Pagina del lavoro: https://arxiv.org/abs/2608.03700 — 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 2026-10-31T12:17:37.812Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).