# A Self-Triggered Agentic Push Recommendation System

arXiv: https://arxiv.org/abs/2608.01949 · HTML: https://mlsanity.com/paper/2608.01949/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Agentic recommender systems, one of 16 directions in LLM agents and tool use.
- Builds on 15 of its 38 references that are in the map.
- First posted on arXiv 3 Aug 2026.

## Placed in

- LLM agents and tool use / [Agentic recommender systems](https://mlsanity.com/agents/agentic-recommender-systems/)

## Builds on (most cited first, 15 of 15)

- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://mlsanity.com/paper/2106.02039/) (2021)
- [Reinforcement Learning to Optimize Long-term User Engagement in Recommender Systems](https://mlsanity.com/paper/1902.05570/) (2019)
- [Causal Embeddings for Recommendation](https://mlsanity.com/paper/1706.07639/) (2017)
- [Unbiased Recommender Learning from Missing-Not-At-Random Implicit Feedback](https://mlsanity.com/paper/1909.03601/) (2019)
- [Deconfounded Recommendation for Alleviating Bias Amplification](https://mlsanity.com/paper/2105.10648/) (2021)
- [Generative Auto-Bidding with Value-Guided Explorations](https://mlsanity.com/paper/2504.14587/) (2025)
- [Offline Reinforcement Learning for Mobile Notifications](https://mlsanity.com/paper/2202.03867/) (2022)
- [Recommending Complementary Products in E-Commerce Push Notifications with a Mixture Model Approach](https://mlsanity.com/paper/1707.08113/) (2017)
- [Treatment Effect Estimation for User Interest Exploration on Recommender Systems](https://mlsanity.com/paper/2405.08582/) (2024)
- [Multi-objective Optimization of Notifications Using Offline Reinforcement Learning](https://mlsanity.com/paper/2207.03029/) (2022)
- [A State Transition Model for Mobile Notifications via Survival Analysis](https://mlsanity.com/paper/2207.03099/) (2022)
- [Learning to Rank For Push Notifications Using Pairwise Expected Regret](https://mlsanity.com/paper/2201.07681/) (2022)
- [Agentic Personalisation of Cross-Channel Marketing Experiences](https://mlsanity.com/paper/2506.16429/) (2025)
- [Soft Frequency Capping for Improved Ad Click Prediction in Yahoo Gemini Native](https://mlsanity.com/paper/2312.05052/) (2023)

## Abstract (arXiv)

Push notification is a critical recommendation scenario on large-scale platforms, allowing the system to proactively reach users outside the application to improve long-term re-engagement. However, designing an optimal push system requires handling a complex action space for the "whether and when" delivery problem under strict system resource constraints. Existing solutions typically fall into two passive paradigms: pre-planned frequency methods that allocate delivery times via offline modeling, limiting real-time adaptability; and fixed-interval triggering methods that periodically poll the system, creating a strict dilemma between excessive computational overhead and diminished optimal timing capture. Furthermore, such multi-stage frameworks severely suffer from local optima. To overcome these limitations, in this paper, we propose STEPS, a proactive, Self-Triggered End-to-end Agentic Push Recommendation System, which is already fully deployed at Douyin with over 1 billion users. STEPS reformulates push recommendation as a self-triggered agentic process in which the system decides not only whether to send a push, but also when to invoke itself again, thereby forming a closed loop that balances real-time effectiveness and efficiency. Specifically, STEPS consists of two decision transformer-based agents: a planning agent that schedules the next system invocation using a gated ordinal regression method, and an execution agent that decides whether to send a push based on trajectory rewards. Furthermore, we introduce a lightweight filtering agent to both control computational overhead and act as a crucial safeguard against unreasonable planning behaviors. Online A/B testing demonstrates that STEPS significantly increases user active days by 0.2843% and reduces the push permission disablement rate by 1.9089%, while the filtering agent reduces computational overhead by 79.42%.