Abstract
Conventional approaches to consensus building in multi-agent systems typically rely on fixed numbers of debate rounds or predetermined thresholds, which can be inefficient and inflexible. Moreover, these conventional approaches often fail to account for the dynamic nature of debates and waste computational resources by continuing debates that have already reached sufficient consensus or by prematurely terminating debates. To address these challenges, the techniques presented herein leverage sequential analysis to efficiently and dynamically determine consensus in multi-agent systems.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Morandi, Andrea, "SEQUENTIAL CONSENSUS BUILDING FOR MULTI-AGENT DEBATES", Technical Disclosure Commons, (October 01, 2025)
https://www.tdcommons.org/dpubs_series/8663