UniPolicy: Unified Objective-Specific Policies for Generative Search Advertising

Kun Yao, Yuhang Zhou, Yichi Zhang, Zeliang Tong, Shengri Xue et al.|September 17, 2026arXiv

Key Takeaway

Multi-objective alignment in generative models works better when you give each objective its own specialized parameters and policies rather than trying to optimize them all with a single reward signal.

Summary

UniPolicy is a framework for search advertising that balances multiple competing business goals—relevance, clicks, and revenue—by training separate specialized policies within a shared model.

training applications reasoning

Key Terms

multi-objective-reinforcement-learning sparse-mixture-of-experts low-rank-adaptation preference-optimization beam-search