# Finding Interest Needle in Popularity Haystack: Improving Retrieval by Modeling Item Exposure

arXiv: https://arxiv.org/abs/2503.23630 · HTML: https://mlsanity.com/paper/2503.23630/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 14 references that are in the map.
- First posted on arXiv 31 Mar 2025.

## Builds on (most cited first, 7 of 7)

- [Top-K Off-Policy Correction for a REINFORCE Recommender System](https://mlsanity.com/paper/1812.02353/) (2018)
- [Modeling User Exposure in Recommendation](https://mlsanity.com/paper/1510.07025/) (2015)
- [Counterfactual Evaluation of Slate Recommendations with Sequential Reward Interactions](https://mlsanity.com/paper/2007.12986/) (2020)
- [Counterfactual Risk Minimization: Learning from Logged Bandit Feedback](https://mlsanity.com/paper/1502.02362/) (2015)
- [Revisiting Neural Retrieval on Accelerators](https://mlsanity.com/paper/2306.04039/) (2023)
- [It Is Different When Items Are Older: Debiasing Recommendations When Selection Bias and User Preferences Are Dynamic](https://mlsanity.com/paper/2111.12481/) (2021)
- [Lessons Learned Addressing Dataset Bias in Model-Based Candidate Generation at Twitter](https://mlsanity.com/paper/2105.09293/) (2021)

## Abstract (arXiv)

Recommender systems operate in closed feedback loops, where user interactions reinforce popularity bias, leading to over-recommendation of already popular items while under-exposing niche or novel content. Existing bias mitigation methods, such as Inverse Propensity Scoring (IPS) and Off-Policy Correction (OPC), primarily operate at the ranking stage or during training, lacking explicit real-time control over exposure dynamics. In this work, we introduce an exposure-aware retrieval scoring approach, which explicitly models item exposure probability and adjusts retrieval-stage ranking at inference time. Unlike prior work, this method decouples exposure effects from engagement likelihood, enabling controlled trade-offs between fairness and engagement in large-scale recommendation platforms. We validate our approach through online A/B experiments in a real-world video recommendation system, demonstrating a 25% increase in uniquely retrieved items and a 40% reduction in the dominance of over-popular content, all while maintaining overall user engagement levels. Our results establish a scalable, deployable solution for mitigating popularity bias at the retrieval stage, offering a new paradigm for bias-aware personalization.