# Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender System

arXiv: https://arxiv.org/abs/2010.15363 · HTML: https://mlsanity.com/paper/2010.15363/

## Key facts

- Cited by 94 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Oct 2020.

## Built on by (most cited first, 30 of 94)

- [AutoDebias: Learning to Debias for Recommendation](https://mlsanity.com/paper/2105.04170/) (2021)
- [CoRAL: Collaborative Retrieval-Augmented Large Language Models Improve Long-tail Recommendation](https://mlsanity.com/paper/2403.06447/) (2024)
- [A Survey on Popularity Bias in Recommender Systems](https://mlsanity.com/paper/2308.01118/) (2023)
- [A Multi-Agent Conversational Recommender System](https://mlsanity.com/paper/2402.01135/) (2024)
- [Generalizing to the Future: Mitigating Entity Bias in Fake News Detection](https://mlsanity.com/paper/2204.09484/) (2022)
- [VCR-Graphormer: A Mini-batch Graph Transformer via Virtual Connections](https://mlsanity.com/paper/2403.16030/) (2024)
- [Augmentations in Hypergraph Contrastive Learning: Fabricated and Generative](https://mlsanity.com/paper/2210.03801/) (2022)
- [Towards Unified Multi-Modal Personalization: Large Vision-Language Models for Generative Recommendation and Beyond](https://mlsanity.com/paper/2403.10667/) (2024)
- [Causal Reasoning Meets Visual Representation Learning: A Prospective Study](https://mlsanity.com/paper/2204.12037/) (2022)
- [A Causal View of Entity Bias in (Large) Language Models](https://mlsanity.com/paper/2305.14695/) (2023)
- [Uncovering Main Causalities for Long-tailed Information Extraction](https://mlsanity.com/paper/2109.05213/) (2021)
- [How Do Recommendation Models Amplify Popularity Bias? An Analysis from the Spectral Perspective](https://mlsanity.com/paper/2404.12008/) (2024)
- [Comprehensive Fair Meta-learned Recommender System](https://mlsanity.com/paper/2206.04789/) (2022)
- [SelfElicit: Your Language Model Secretly Knows Where is the Relevant Evidence](https://mlsanity.com/paper/2502.08767/) (2025)
- [Towards Fair Federated Recommendation Learning: Characterizing the Inter-Dependence of System and Data Heterogeneity](https://mlsanity.com/paper/2206.02633/) (2022)
- [Recent Developments in Recommender Systems: A Survey](https://mlsanity.com/paper/2306.12680/) (2023)
- [Causal Incremental Graph Convolution for Recommender System Retraining](https://mlsanity.com/paper/2108.06889/) (2021)
- [Deconfounded Causal Collaborative Filtering](https://mlsanity.com/paper/2110.07122/) (2021)
- [A Survey on Fairness-aware Recommender Systems](https://mlsanity.com/paper/2306.00403/) (2023)
- [PyG-SSL: A Graph Self-Supervised Learning Toolkit](https://mlsanity.com/paper/2412.21151/) (2024)
- [Causal Collaborative Filtering](https://mlsanity.com/paper/2102.01868/) (2021)
- [Can Graph Neural Networks Learn Language with Extremely Weak Text Supervision?](https://mlsanity.com/paper/2412.08174/) (2024)
- [Robust Collaborative Filtering to Popularity Distribution Shift](https://mlsanity.com/paper/2310.10696/) (2023)
- [Controlling Risk of Retrieval-augmented Generation: A Counterfactual Prompting Framework](https://mlsanity.com/paper/2409.16146/) (2024)
- [Breaking Silos: Adaptive Model Fusion Unlocks Better Time Series Forecasting](https://mlsanity.com/paper/2505.18442/) (2025)
- [Unbiased Knowledge Distillation for Recommendation](https://mlsanity.com/paper/2211.14729/) (2022)
- [Popularity-Aware Alignment and Contrast for Mitigating Popularity Bias](https://mlsanity.com/paper/2405.20718/) (2024)
- [PSL: Rethinking and Improving Softmax Loss from Pairwise Perspective for Recommendation](https://mlsanity.com/paper/2411.00163/) (2024)
- [Cross Pairwise Ranking for Unbiased Item Recommendation](https://mlsanity.com/paper/2204.12176/) (2022)
- [Debiased Recommendation with Noisy Feedback](https://mlsanity.com/paper/2406.17182/) (2024)

## Abstract (arXiv)

The general aim of the recommender system is to provide personalized suggestions to users, which is opposed to suggesting popular items. However, the normal training paradigm, i.e., fitting a recommender model to recover the user behavior data with pointwise or pairwise loss, makes the model biased towards popular items. This results in the terrible Matthew effect, making popular items be more frequently recommended and become even more popular. Existing work addresses this issue with Inverse Propensity Weighting (IPW), which decreases the impact of popular items on the training and increases the impact of long-tail items. Although theoretically sound, IPW methods are highly sensitive to the weighting strategy, which is notoriously difficult to tune. In this work, we explore the popularity bias issue from a novel and fundamental perspective -- cause-effect. We identify that popularity bias lies in the direct effect from the item node to the ranking score, such that an item's intrinsic property is the cause of mistakenly assigning it a higher ranking score. To eliminate popularity bias, it is essential to answer the counterfactual question that what the ranking score would be if the model only uses item property. To this end, we formulate a causal graph to describe the important cause-effect relations in the recommendation process. During training, we perform multi-task learning to achieve the contribution of each cause; during testing, we perform counterfactual inference to remove the effect of item popularity. Remarkably, our solution amends the learning process of recommendation which is agnostic to a wide range of models -- it can be easily implemented in existing methods. We demonstrate it on Matrix Factorization (MF) and LightGCN [20]. Experiments on five real-world datasets demonstrate the effectiveness of our method.