aiqbench analysis
This research investigates how multi-agent systems interact when combining language models of varying sizes and families, revealing that persuasion dynamics are driven by specific model pairings rather than individual scale or certainty. The study demonstrates that even small models can effectively overturn the judgments of larger ones, highlighting that heterogeneity significantly amplifies or suppresses influence depending on the listener's susceptibility. These findings underscore the critical need to evaluate interacting AI combinations empirically rather than inferring system behavior from isolated model properties.
Peer Influence across Heterogeneous AI Models
arXiv:2610.03095v1 Announce Type: new Abstract: When two AI agents disagree, who persuades whom? As multi-agent systems increasingly combine language models of different families and sizes, the answer can determine which judgments survive interaction. Measuring persuasion as the probabilistic shift in an agent's decision after a single exchange with a dissenting peer, we test seven open-weight models across three language understanding tasks. We find that persuasion is strong: when models disagree, receivers often abandon their initial judgment after seeing a peer's answer and explanation. Surprisingly, however, neither standalone certainty nor model scale reliably predicts persuasion dynamics. Models producing almost perfectly consistent decisions in isolation can be among the most susceptible to persuasion, and small models can match larger ones as persuaders and resist their influence just as effectively. Furthermore, we show that the size of the shift depends more on the susceptibility of the listener than on the persuasiveness of the speaker. Persuasion patterns are therefore specific to each model pairing, with heterogeneity amplifying persuasion in some combinations and suppressing it in others, allowing a dissenting agent running a small model to overturn the judgments of a much larger one. These findings show that the behavior of interacting models cannot be inferred from their individual properties but must be evaluated in the combinations in which they will operate.
Abstract:When two AI agents disagree, who persuades whom? As multi-agent systems increasingly combine language models of different families and sizes, the answer can determine which judgments survive interaction. Measuring persuasion as the probabilistic shift in an agent's decision after a single exchange with a dissenting peer, we test seven open-weight models across three language understanding tasks. We find that persuasion is strong: when models disagree, receivers often abandon their initial judgment after seeing a peer's answer and explanation. Surprisingly, however, neither standalone certainty nor model scale reliably predicts persuasion dynamics. Models producing almost perfectly consistent decisions in isolation can be among the most susceptible to persuasion, and small models can match larger ones as persuaders and resist their influence just as effectively. Furthermore, we show that the size of the shift depends more on the susceptibility of the listener than on the persuasiveness of the speaker. Persuasion patterns are therefore specific to each model pairing, with heterogeneity amplifying persuasion in some combinations and suppressing it in others, allowing a dissenting agent running a small model to overturn the judgments of a much larger one. These findings show that the behavior of interacting models cannot be inferred from their individual properties but must be evaluated in the combinations in which they will operate.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY); Physics and Society (physics.soc-ph)
arXiv:2610.03095 [cs.AI]
(or arXiv:2610.03095v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.03095 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 318 words
Related coverage
The Ringelmann Effect in Multi-Agent LLM Systems: A Scaling Law for Effective Team Size
arXiv · Jun 3, 2026
Reviewing failure modes in science as we prepare for AI researchers
LessWrong · Oct 8, 2026
What We Learned Trying to Catch AI Liars: An Aletheia's Quest Retrospective
EleutherAI · Aug 25, 2026
Hint-based CoT faithfulness evals still mostly work on Claude
LessWrong · Jul 30, 2026