LLMs as Orchestrators: Constraint-Compliant Multi-Agent Optimization for Recommendation Systems Conference: Companion Proceedings of the ACM Web Conference 2026; April 13–17, 2026; Dubai, United Arab Emirates Companion Proceedings of the ACM Web Conference 2026 (WWW Companion ’26), April 13–17, 2026
Guilin Zhang, Kai Zhao, Jeffrey Friedman, Xu Chu · 2026
Read LLMs as Orchestrators: Constraint-Compliant Multi-Agent Optimization for Recommendation Systems Conference: Companion Proceedings of the ACM Web Conference 2026; April 13–17, 2026; Dubai, United Arab Emirates Companion Proceedings of the ACM Web Conference 2026 (WWW Companion ’26), April 13–17, 2026 by Guilin Zhang, Kai Zhao, Jeffrey Friedman, Xu Chu (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-2601-19121.txt.
Description
Guilin Zhang and three co-authors put language models in charge of a team of recommendation agents. A recommender must serve many goals at once while obeying hard business rules. A shop, for instance, may demand that each list span several sellers and meet fairness and coverage floors.
The models act as orchestrators, steering sub agents while keeping every rule in force. Letting a model juggle goals is easy; keeping it inside hard limits is the hard part. Systems that do both could tune feeds in real time without breaking the rules firms must keep.
Catalog record
| CREATOR | Guilin Zhang, Kai Zhao, Jeffrey Friedman, Xu Chu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-19121