Bridging Semantic Understanding and Popularity Bias with LLMs
Not on the map. Why?
Abstract (arXiv)
Semantic understanding of popularity bias is a crucial yet underexplored challenge in recommender systems, where popular items are often favored at the expense of niche content. Most existing debiasing methods treat the semantic understanding of popularity bias as a matter of diversity enhancement or long-tail coverage, neglecting the deeper semantic layer that embodies the causal origins of the bias itself. Consequently, such shallow interpretations limit both their debiasing effectiveness and recommendation accuracy. In this paper, we propose FairLRM, a novel framework that bridges the gap in the semantic understanding of popularity bias with Recommendation via Large Language Model (RecLLM). FairLRM decomposes popularity bias into item-side and user-side components, using structured instruction-based prompts to enhance the model's comprehension of both global item distributions and individual user preferences. Unlike traditional methods that rely on surface-level features such as "diversity" or "debiasing", FairLRM improves the model's ability to semantically interpret and address the underlying bias. Through empirical evaluation, we show that FairLRM significantly enhances both fairness and recommendation accuracy, providing a more semantically aware and trustworthy approach to enhance the semantic understanding of popularity bias. The implementation is available at https://github.com/LuoRenqiang/FairLRM.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 24 of its 51 references that are in the map. First posted on arXiv 14 Jan 2026.
Builds on
24 of 51 references.
References · 24
- 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
- 2023Qwen Technical Reportcited by 3,273
- 2019The Unfairness of Popularity Bias in Recommendationcited by 68
- 2023A Survey on Popularity Bias in Recommender Systemscited by 42
- 2023UP5: Unbiased Foundation Model for Fairness-aware Recommendationcited by 38
- 2024FairGT: A Fairness-aware Graph Transformercited by 29
- 2024Can Large Language Models Improve the Adversarial Robustness of Graph Neural Networks?cited by 22
- 2024Item-side Fairness of Large Language Model-based Recommendation Systemcited by 21
- 2024How Do Recommendation Models Amplify Popularity Bias? An Analysis from the Spectral Perspectivecited by 21
- 2024SPRec: Self-Play to Debias LLM-based Recommendationcited by 18
- 2023Understanding Before Recommendation: Semantic Aspect-Aware Review Exploitation via Large Language Modelscited by 13
- 2024FairGP: A Scalable and Fair Graph Transformer Using Graph Partitioningcited by 11
12 more
- 2023How Graph Convolutions Amplify Popularity Bias for Recommendation?cited by 10
- 2025PUB: An LLM-Enhanced Personality-Driven User Behaviour Simulator for Recommender System Evaluationcited by 10
- 2025Cultural Bias Matters: A Cross-Cultural Benchmark Dataset and Sentiment-Enriched Model for Understanding Multimodal Metaphorscited by 10
- 2024Popularity-Aware Alignment and Contrast for Mitigating Popularity Biascited by 9
- 2024Debiasing Recommendation with Personal Popularitycited by 7
- 2024Going Beyond Popularity and Positivity Bias: Correcting for Multifactorial Bias in Recommender Systemscited by 6
- 2024Logical Reasoning with Relation Network for Inductive Knowledge Graph Completioncited by 6
- 2024LARR: Large Language Model Aided Real-time Scene Recommendation with Semantic Understandingcited by 6
- 2025Unbiased Reasoning for Knowledge-Intensive Tasks in Large Language Models via Conditional Front-Door Adjustmentcited by 6
- 2025Temporal-Aware User Behaviour Simulation with Large Language Models for Recommender Systemscited by 6
- 2025Rethinking Byzantine Robustness in Federated Recommendation from Sparse Aggregation Perspectivecited by 4
- 2018Popularity-Aware Item Weighting for Long-Tail Recommendationcited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.