Metraipolis › archive › dc_a366d75f586f

Large Language Model Orchestration under Heterogeneous Preferences via Explicit Persona Inference — arXiv:2610.07587

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.07587
License (as archived)
CC0-1.0
Archived
2026-10-07T11:23:08.105Z
SHA-256 of the archived text
179bdfafb6e144a66bd3301c9aa9b8ae566566b0c0d17caa1c8d01780439d2a1
Stable URL of this page
https://metraipolis.com/documento/dc_a366d75f586f

How to cite

Large Language Model Orchestration under Heterogeneous Preferences via Explicit Persona Inference — arXiv:2610.07587 — arXiv (en), https://arxiv.org/abs/2610.07587 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a366d75f586f (Metraipolis archive dc_a366d75f586f, sha256 179bdfafb6e144a66bd3301c9aa9b8ae566566b0c0d17caa1c8d01780439d2a1)
BibTeX
@misc{dca366d75f586f,
  title = {Large Language Model Orchestration under Heterogeneous Preferences via Explicit Persona Inference — arXiv:2610.07587},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07587},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a366d75f586f; sha256 179bdfafb6e144a66bd3301c9aa9b8ae566566b0c0d17caa1c8d01780439d2a1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07587. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM orchestration investigates how an orchestrator coordinates a group of autonomous agents to achieve common goals or maximize collective welfare. The agents are typically heterogeneous, each holding a private preference that it pursues but does not reveal. Inferring such hidden preferences from behavior has been a subject of long-standing research in game theory and multi-agent systems. The core challenge lies in maintaining a belief over every agent's preference and updating it from the agents' observed actions. Existing LLM orchestrators carry that belief as prompt text with no explicit update rule. This lets early errors persist and propagate rather than be corrected. We therefore propose \textbf{HARP} (Heterogeneous-preference Agent oRchestration via Preference inference), a novel framework that moves the belief out of the prompt. Specifically, HARP maintains one numeric posterior per agent over a finite set of candidate preferences and updates it in closed form by Bayes' rule. The language model supplies only actions and per-candidate likelihoods, so estimation is decoupled from its reasoning. We prove that HARP attains the same $\tilde O(\sqrt K)$ Bayesian regret as explicit joint inference when the factorization is exact. Furthermore, HARP\textsuperscript{+} augments planning with a bonus for actions that distinguish the candidates, so inference continues even when the optimal action is uninformative. Empirical results on three substrates, ranging from payoffs the preferences fully determine, through payoffs that depend on more than them, to scales where explicit joint inference is infeasible, demonstrate that HARP\textsuperscript{+} is the strongest non-oracle method across the class our theory identifies.

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/2610.07587 — 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 2026-10-07T11:15:39.423Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).