The Weight of Teamwork: Does Communication Structure Help LLM Scientists Search for Scientific Laws?

Abstract:Team performance is often reported as one delivered score, even though that score is produced in two stages: a system generates candidate answers and a selection rule returns one of them. We use symbolic regression as an observable scientific-search setting and compare fixed-budget LLM teams that differ in peer visibility and update timing, evaluating both their frozen candidate archives and the expressions selected from those archives. Distributing the call budget across more independent trajectories produced the largest measured gain, improving mean held-out fitting score Q by 0.70 at the widest tested allocation. Under a common archive-wide selector, the implemented Ring policy reduced mean Q by 0.33 relative to Independent, while Star and Complete remained unresolved. Holding the generated archives fixed, changing the delivery rule changed the measured architecture gaps, including the sign of the Complete point estimate. These results support reporting candidate generation and final selection as separate endpoints. They do not identify a communication mechanism or a universally preferred topology; rather, they show that an architecture ranking is not identified by communication structure alone when the rule that returns the final answer is left unspecified.

Accepted at the NeurIPS 2026 Workshop on Foundations of Agentic Systems Theory (FAST), non-archival. 7 pages main text, 33 pages total with appendices

Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

arXiv:2610.24131 [cs.MA]

(or arXiv:2610.24131v1 [cs.MA] for this version)

https://doi.org/10.48550/arXiv.2610.24131 arXiv-issued DOI via DataCite (pending registration)