# Artificial collectives of specialists and generalists excel at different tasks

arXiv: https://arxiv.org/abs/2606.20877 · HTML: https://mlsanity.com/paper/2606.20877/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 63 references that are in the map.
- First posted on arXiv 18 Jun 2026.

## Builds on (most cited first, 8 of 8)

- [A Review of Cooperative Multi-Agent Deep Reinforcement Learning](https://mlsanity.com/paper/1908.03963/) (2019)
- [Towards a Science of Scaling Agent Systems](https://mlsanity.com/paper/2512.08296/) (2025)
- [Collective Intelligence in Human-AI Teams: A Bayesian Theory of Mind Approach](https://mlsanity.com/paper/2208.11660/) (2022)
- [Social learning strategies modify the effect of network structure on group performance](https://mlsanity.com/paper/1606.00753/) (2016)
- [Personalized AI Scaffolds Synergistic Multi-Turn Collaboration in Creative Work](https://mlsanity.com/paper/2510.27681/) (2025)
- [Facts and Figuring: An Experimental Investigation of Network Structure and Performance in Information and Solution Spaces](https://mlsanity.com/paper/1406.7586/) (2014)
- [Systems of bounded rational agents with information-theoretic constraints](https://mlsanity.com/paper/1809.05897/) (2018)
- [Multidisciplinary learning through collective performance favors decentralization](https://mlsanity.com/paper/2208.11618/) (2022)

## Abstract (arXiv)

Collective artificial intelligence, where multiple agents work on shared tasks, holds potential to solve expansive problems in fields from medicine to collective governance. But while prescriptive engineering solutions abound, we lack descriptive scientific understanding of artificial collectives, and therefore principles for how to design resource efficient multi-agent systems. Through systematic experiments with optimizing agents, we characterize how agent interpretive abilities, rationality bounds, and task qualities interact to shape collective performance. Agents range from specialists, with narrow interpretive abilities, to generalists, with broad ones. Collectives of specialists correspond to sparse, centralized networks, while collectives of generalists correspond to dense, decentralized ones. We show that interpretive network properties have small performance effects on average (0.07 standard deviations of performance). However, for specific task qualities, these effects are 4.5 times larger (0.33 sd) and can reach much higher for certain task qualities (1.84 sd). This leads collectives of generalists to perform better on tasks that involve generating, choosing, and coordinating, while collectives of specialists with a few generalist mediators perform better on tasks that involve negotiating. Rationality bounds then moderate these relationships. At loose bounds, specialists outperform generalists through more effective sampling of high-dimensional decision spaces. At tight bounds, generalists outperform specialists through better gradient estimation. A fundamental trade-off between performance and convergence speed emerges at moderate bounds. These findings suggest that multi-agent design could benefit from matching interpretive networks to both task demands and agents' computational limits, with implications for the efficiency and energy costs of multi-agent systems.