# Combining Reward and Rank Signals for Slate Recommendation

arXiv: https://arxiv.org/abs/2107.12455 · HTML: https://mlsanity.com/paper/2107.12455/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 17 references that are in the map.
- First posted on arXiv 26 Jul 2021.

## Builds on (most cited first, 5 of 5)

- [Introduction to Multi-Armed Bandits](https://mlsanity.com/paper/1904.07272/) (2019)
- [Off-policy evaluation for slate recommendation](https://mlsanity.com/paper/1605.04812/) (2016)
- [Seq2Slate: Re-ranking and Slate Optimization with RNNs](https://mlsanity.com/paper/1810.02019/) (2018)
- [Beyond Greedy Ranking: Slate Optimization via List-CVAE](https://mlsanity.com/paper/1803.01682/) (2018)
- [Algorithms for slate bandits with non-separable reward functions](https://mlsanity.com/paper/2004.09957/) (2020)

## Built on by (most cited first, 5 of 5)

- [Exponential Smoothing for Off-Policy Learning](https://mlsanity.com/paper/2305.15877/) (2023)
- [Probabilistic Rank and Reward: A Scalable Model for Slate Recommendation](https://mlsanity.com/paper/2208.06263/) (2022)
- [Offline Evaluation of Reward-Optimizing Recommender Systems: The Case of Simulation](https://mlsanity.com/paper/2209.08642/) (2022)
- [SlateFree: a Model-Free Decomposition for Reinforcement Learning with Slate Actions](https://mlsanity.com/paper/2209.01876/) (2022)
- [On-Policy and Off-Policy Learning for Large Action Spaces](https://mlsanity.com/paper/2607.28408/) (2026)

## Abstract (arXiv)

We consider the problem of slate recommendation, where the recommender system presents a user with a collection or slate composed of K recommended items at once. If the user finds the recommended items appealing then the user may click and the recommender system receives some feedback. Two pieces of information are available to the recommender system: was the slate clicked? (the reward), and if the slate was clicked, which item was clicked? (rank). In this paper, we formulate several Bayesian models that incorporate the reward signal (Reward model), the rank signal (Rank model), or both (Full model), for non-personalized slate recommendation. In our experiments, we analyze performance gains of the Full model and show that it achieves significantly lower error as the number of products in the catalog grows or as the slate size increases.