Drop the Hierarchy and Roles: How Self-Organizing LLM Agents Outperform Designed Structures

Computer Science > Artificial Intelligence [Submitted on 30 Mar 2026] Drop the Hierarchy and Roles: How Self-Organizing LLM Agents Outperform Designed Structures Victoria Dochkina How much autonomy can multi-agent LLM systems sustain -- and what enables it? We present a 25,000-task computational experiment spanning 8 models, 4--256 agents, and 8 coordination protocols ranging from externally imposed hierarchy to emergent self-organization. We observe that autonomous behavior already emerges in current LLM agents: given minimal structural scaffolding (fixed ordering), agents spontaneously invent specialized roles, voluntarily abstain from tasks outside their competence, and form shallow hierarchies -- without any pre-assigned roles or external design. A hybrid protocol (Sequential) that enables this autonomy outperforms centralized coordination by 14% (p<0.001), with a 44% quality spread between protocols (Cohen's d=1.86, p<0.0001). The degree of emergent autonomy scales with model capability: strong models self-organize effectively, while models below a capability threshold still benefit from rigid structure -- suggesting that as foundation models improve, the scope for autonomous coordination will expand. The system scales sub-linearly to 256 agents without quality degradation (p=0.61), producing 5,006 unique roles from just 8 agents. Results replicate across closed- and open-source models, with open-source achieving 95% of closed-source quality at 24x lower cost. The practical implication: give agents a mission, a protocol, and a capable model -- not a pre-assigned role. Comments: 6 figures, 9 tables. Submitted to IEEE Access Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2603.28990 [cs.AI]   (or arXiv:2603.28990v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2603.28990 Focus to learn more Submission history From: Victoria Dochkina [view email] [v1] Mon, 30 Mar 2026 20:44:53 UTC (624 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-03 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)