Researchers identify a 'selection bottleneck' that mathematically determines when diverse agent teams outperform homogeneous self-consistency teams.
March 24, 2026
Original Paper
When Agents Disagree: The Selection Bottleneck in Multi-Agent LLM Pipelines
arXiv · 2603.20324
The Takeaway
It provides a closed-form threshold for aggregation quality, explaining why Mixture-of-Agents (MoA) sometimes fails to beat single-model baselines. The finding suggests that improving 'judge' or 'selector' quality is a higher-leverage design choice than increasing generator diversity.
From the abstract
Multi-agent LLM pipelines produce contradictory evidence on whether team diversity improves output quality: heterogeneous Mixture-of-Agents teams outperform single models, yet homogeneous Self-MoA teams consistently win under synthesis-based aggregation. We propose a resolution by identifying the selection bottleneck -- a crossover threshold in aggregation quality that determines whether diversity helps or hurts. Under this model, we obtain a closed-form crossover threshold $s^*$ (Proposition 1)