Metraipolis › archive › dc_8f0f0909a025
Do LLMs Understand Personality? Rethinking Persona Fidelity Evaluation through Structured Behavioral Inference — arXiv:2608.26674
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.26674
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T14:26:16.924Z
- SHA-256 of the archived text
92e59316d15ab9b765ac2882b4fea78ecfe1e6e4289fd15cb7795c9d2ed27c89- Stable URL of this page
https://metraipolis.com/documento/dc_8f0f0909a025
How to cite
Do LLMs Understand Personality? Rethinking Persona Fidelity Evaluation through Structured Behavioral Inference — arXiv:2608.26674 — arXiv (en), https://arxiv.org/abs/2608.26674 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8f0f0909a025 (Metraipolis archive dc_8f0f0909a025, sha256 92e59316d15ab9b765ac2882b4fea78ecfe1e6e4289fd15cb7795c9d2ed27c89)
BibTeX
@misc{dc8f0f0909a025,
title = {Do LLMs Understand Personality? Rethinking Persona Fidelity Evaluation through Structured Behavioral Inference — arXiv:2608.26674},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.26674},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8f0f0909a025; sha256 92e59316d15ab9b765ac2882b4fea78ecfe1e6e4289fd15cb7795c9d2ed27c89},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26674. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
As large language models are increasingly deployed to simulate diverse human characters, ensuring persona fidelity, defined as the extent to which an agent's behavior consistently reflects the psychological and stylistic characteristics of a target persona, has become a critical requirement. However, existing evaluation paradigms primarily rely on either holistic LLM-based judges, which are prone to "holistic appraisal hallucination'', or static psychometric inventories, which fail to capture the context-dependent fidelity required in dynamic dialogue. To address these limitations, we propose PRISM (Persona Reasoning with Inverse SFL-based Modeling), a psycholinguistically grounded framework that reformulates persona fidelity evaluation as a structured inverse inference task. Inspired by Systemic Functional Linguistics (SFL), PRISM decomposes persona fidelity into three functional dimensions: Task Framing, Interpersonal Stance, and Linguistic Style. It estimates dimension-specific evidence over a persona-conditioned label space and aggregates these signals into an interpretable and auditable evaluation process. Experiments show that PRISM yields more accurate and stable judgements than traditional holistic judging, providing a more reliable framework for persona fidelity evaluation. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.26674 — 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 2026-10-17T13:16:53.652Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).