# System-2 Recommenders: Disentangling Utility and Engagement in Recommendation Systems via Temporal Point-Processes

arXiv: https://arxiv.org/abs/2406.01611 · HTML: https://mlsanity.com/paper/2406.01611/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 58 references that are in the map.
- First posted on arXiv 29 May 2024.

## Builds on (most cited first, 19 of 19)

- [Neural Temporal Point Processes: A Review](https://mlsanity.com/paper/2104.03528/) (2021)
- [Reinforcement Learning to Optimize Long-term User Engagement in Recommender Systems](https://mlsanity.com/paper/1902.05570/) (2019)
- [Controlling Fairness and Bias in Dynamic Learning-to-Rank](https://mlsanity.com/paper/2005.14713/) (2020)
- [Degenerate Feedback Loops in Recommender Systems](https://mlsanity.com/paper/1902.10730/) (2019)
- [Auditing Radicalization Pathways on YouTube](https://mlsanity.com/paper/1908.08313/) (2019)
- [What are you optimizing for? Aligning Recommender Systems with Human Values](https://mlsanity.com/paper/2107.10939/) (2021)
- [The Challenge of Understanding What Users Want: Inconsistent Preferences and Engagement Optimization](https://mlsanity.com/paper/2202.11776/) (2022)
- [Estimating and Penalizing Induced Preference Shifts in Recommender Systems](https://mlsanity.com/paper/2204.11966/) (2022)
- [Building Human Values into Recommender Systems: An Interdisciplinary Synthesis](https://mlsanity.com/paper/2207.10192/) (2022)
- [From Optimizing Engagement to Measuring Value](https://mlsanity.com/paper/2008.12623/) (2020)
- [Impatient Bandits: Optimizing Recommendations for the Long-Term Without Delay](https://mlsanity.com/paper/2307.09943/) (2023)
- [Algorithmic Extremism: Examining YouTube's Rabbit Hole of Radicalization](https://mlsanity.com/paper/1912.11211/) (2019)
- [A Longitudinal Analysis of YouTube's Promotion of Conspiracy Videos](https://mlsanity.com/paper/2003.03318/) (2020)
- [Self-Control in Cyberspace: Applying Dual Systems Theory to a Review of Digital Self-Control Tools](https://mlsanity.com/paper/1902.00157/) (2019)
- [Choosing the Right Weights: Balancing Value, Strategy, and Noise in Recommender Systems](https://mlsanity.com/paper/2305.17428/) (2023)
- [Fast online ranking with fairness of exposure](https://mlsanity.com/paper/2209.13019/) (2022)
- [Time-Inconsistent Planning: A Computational Problem in Behavioral Economics](https://mlsanity.com/paper/1405.1254/) (2014)
- ["So, Tell Me What Users Want, What They Really, Really Want!"](https://mlsanity.com/paper/1803.02065/) (2018)
- [Ranking with Long-Term Constraints](https://mlsanity.com/paper/2307.04923/) (2023)

## Built on by (most cited first, 2 of 2)

- [CTRL-Rec: Controlling Recommender Systems With Natural Language](https://mlsanity.com/paper/2510.12742/) (2025)
- [MultiScale Contextual Bandits for Long Term Objectives](https://mlsanity.com/paper/2503.17674/) (2025)

## Abstract (arXiv)

Recommender systems are an important part of the modern human experience whose influence ranges from the food we eat to the news we read. Yet, there is still debate as to what extent recommendation platforms are aligned with the user goals. A core issue fueling this debate is the challenge of inferring a user utility based on engagement signals such as likes, shares, watch time etc., which are the primary metric used by platforms to optimize content. This is because users utility-driven decision-processes (which we refer to as System-2), e.g., reading news that are relevant for them, are often confounded by their impulsive decision-processes (which we refer to as System-1), e.g., spend time on click-bait news. As a result, it is difficult to infer whether an observed engagement is utility-driven or impulse-driven. In this paper we explore a new approach to recommender systems where we infer user utility based on their return probability to the platform rather than engagement signals. Our intuition is that users tend to return to a platform in the long run if it creates utility for them, while pure engagement-driven interactions that do not add utility, may affect user return in the short term but will not have a lasting effect. We propose a generative model in which past content interactions impact the arrival rates of users based on a self-exciting Hawkes process. These arrival rates to the platform are a combination of both System-1 and System-2 decision processes. The System-2 arrival intensity depends on the utility and has a long lasting effect, while the System-1 intensity depends on the instantaneous gratification and tends to vanish rapidly. We show analytically that given samples it is possible to disentangle System-1 and System-2 and allow content optimization based on user utility. We conduct experiments on synthetic data to demonstrate the effectiveness of our approach.