Metraipolis › archive › dc_993310fefd6a

Who Is the Influencer? Consensus Dynamics in Multi-Agent Systems — arXiv:2610.31105

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.31105
License (as archived)
CC0-1.0
Archived
2026-10-23T15:13:15.909Z
SHA-256 of the archived text
6d45f4b3c8982c4803c32fb5e2f3d639488582f065fb927d9d68bd36ff089549
Stable URL of this page
https://metraipolis.com/documento/dc_993310fefd6a

How to cite

Who Is the Influencer? Consensus Dynamics in Multi-Agent Systems — arXiv:2610.31105 — arXiv (en), https://arxiv.org/abs/2610.31105 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_993310fefd6a (Metraipolis archive dc_993310fefd6a, sha256 6d45f4b3c8982c4803c32fb5e2f3d639488582f065fb927d9d68bd36ff089549)
BibTeX
@misc{dc993310fefd6a,
  title = {Who Is the Influencer? Consensus Dynamics in Multi-Agent Systems — arXiv:2610.31105},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31105},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_993310fefd6a; sha256 6d45f4b3c8982c4803c32fb5e2f3d639488582f065fb927d9d68bd36ff089549},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31105. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-agent LLM systems are increasingly used in practice, but their behavior is complex and hard to understand. We design a simple task to ask which model has the most influence on the final consensus when several models are mixed in one system. Experimentally, across seven model families, we find that the opinions of the other models are absorbed into the opinion of the stronger model, the one with more parameters. This happens even though the task requires no pretraining knowledge and has no correct answer. To understand this phenomenon, we interpret the agents as weighted particles moving on a simplex. Larger models carry a larger weight and therefore occupy a deeper well in the energy landscape. A large weight means strong in-context learning: a model with a large weight amplifies the majority of its context. Larger models are generally better at in-context learning, and this is what makes the larger model the influencer. We hope that this work helps to interpret multi-agent systems, whose behavior is becoming too complex for humans to understand.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.31105 — 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 2026-10-23T15:09:08.437Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).