Abstract
Agentic deployments can assign many parallel agents to one supervisor without specifying when that supervisor can no longer recover the team's state. Human-factors research already establishes relevant constructs such as situation awareness, span of control, neglect tolerance, and multiple-resource workload, but language-model teams distribute state across text and can fail silently. This position paper translates those constructs into a measurement protocol designed for later calibration to people rather than a new human experiment. Typed event rates, isolated service costs, and a throughput budget define a nominal load-balance index; the midpoint of a separately fitted capacity curve remains an empirical quantity. Synthetic controls recover a known capacity boundary, while a seeded queue shows that deadline pressure can separate the fitted midpoint from nominal load balance. The resulting event ontology, operating envelope, executable stress tests, and preregisterable falsification criteria allow system designers to test staffing or interface interventions within a declared team-monitor channel. They do not turn synthetic behavior into evidence about human or model supervisors.
Cite this work
Carlos Toxtli-Hernández and Manuel Delaflor. 2026. How Many Agents Can One Supervisor Track? A Mechanistic Capacity Model and Measurement Protocol for LLM-Agent Teams. Human-AI Coevolution: Measuring Human-Agent Teams in the Agentic Era (HAIC) Workshop at NeurIPS 2026.
@inproceedings{Toxtli2026Supervisor,
title = {How Many Agents Can One Supervisor Track? A Mechanistic Capacity Model and Measurement Protocol for LLM-Agent Teams},
author = {Toxtli, Carlos and Delaflor, Manuel},
booktitle = {Human-AI Coevolution: Measuring Human-Agent Teams in the Agentic Era (HAIC) Workshop at NeurIPS 2026},
address = {Atlanta, GA},
year = {2026},
month = dec,
note = {Workshop paper},
url = {https://openreview.net/forum?id=czINRnBTpf}
}
