Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer?
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911716336467968 |
|---|---|
| author | Bause, Franka Niederle, Jonas Pawelczyk, Martin Burkholz, Rebekka |
| author_facet | Bause, Franka Niederle, Jonas Pawelczyk, Martin Burkholz, Rebekka |
| contents | The effectiveness of multi-agent LLM deliberation depends not only on the agents' individual predictions, but also on how they communicate and collaborate. We study this mechanism through the lens of Friedkin-Johnsen (FJ) opinion dynamics, a tractable model for analyzing stubbornness, influence, and opinion change in multi-agent systems that captures empirically observed deliberation patterns. We show that the FJ parameters are input-dependent, turning multi-agent deliberation into a mixture of experts. This perspective implies that multi-agent systems can outperform single agents and static ensembles when routing reflects agent competence. Since competence is latent in practice, we analyze how influence is established through observable proxies: agents' self-assessed confidence, their perceived confidence, and initial alignment with other agents' views. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_25929 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer? Bause, Franka Niederle, Jonas Pawelczyk, Martin Burkholz, Rebekka Multiagent Systems Machine Learning The effectiveness of multi-agent LLM deliberation depends not only on the agents' individual predictions, but also on how they communicate and collaborate. We study this mechanism through the lens of Friedkin-Johnsen (FJ) opinion dynamics, a tractable model for analyzing stubbornness, influence, and opinion change in multi-agent systems that captures empirically observed deliberation patterns. We show that the FJ parameters are input-dependent, turning multi-agent deliberation into a mixture of experts. This perspective implies that multi-agent systems can outperform single agents and static ensembles when routing reflects agent competence. Since competence is latent in practice, we analyze how influence is established through observable proxies: agents' self-assessed confidence, their perceived confidence, and initial alignment with other agents' views. |
| title | Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer? |
| topic | Multiagent Systems Machine Learning |
| url | https://arxiv.org/abs/2605.25929 |