Generative Auto-Bidding with Unified Modeling and Exploration
The paper presents GUIDE, a new framework for automated bidding in digital advertising. It integrates directed exploration with a safety fallback mechanism to enhance efficiency and reduce financial risk. Experimental results demonstrate GUIDE's superior performance compared to existing methods on various metrics.
- ▪GUIDE employs a Decision Transformer to model historical bidding actions and environmental state transitions.
- ▪The framework includes a Q-value module for exploration guidance and an Inverse Dynamics Module for safe policy fallback.
- ▪In real-world deployment on Taobao, GUIDE achieved a 4.10% increase in ad GMV and a 3.52% increase in ad ROI.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.19457 (cs) [Submitted on 19 May 2026] Title:Generative Auto-Bidding with Unified Modeling and Exploration Authors:Mingming Zhang, Feiqing Zhuang, Na Li, Shengjie Sun, Xiaowei Chen, Junxiong Zhu, Fei Xiao, Keping Yang, Lixin Zou, Chenliang Li View a PDF of the paper titled Generative Auto-Bidding with Unified Modeling and Exploration, by Mingming Zhang and 9 other authors View PDF HTML (experimental) Abstract:Automated bidding is central to modern digital advertising. Early rule-based methods lacked adaptability, while subsequent Reinforcement Learning approaches modeled bidding as a Markov Decision Process but struggled with long-term dependencies.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
