Learning to Recommend Multi-Agent Subgraphs from Calling Trees
Xinyuan Song, Liang Zhao · 2026
Read Learning to Recommend Multi-Agent Subgraphs from Calling Trees by Xinyuan Song, Liang Zhao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-22209.txt.
Description
Xinyuan Song and Liang Zhao recommend teams of agents, not just single tools. Multi-agent systems solve complex tasks by orchestrating agents and tools picked from fast-growing marketplaces. As those marketplaces expand, many candidates overlap in function, and selection becomes more than a retrieval problem.
An orchestrator must filter for relevance, then choose options that are reliable and compatible with the other picks. The authors learn to recommend whole subgraphs of agents from calling trees, the records of which agents called which. Their 2026 preprint treats agent selection as a structured recommendation task. Better selection could make multi-agent systems dependable enough for real business processes.
Catalog record
| CREATOR | Xinyuan Song, Liang Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-22209