Lune

WWW2021Top-tier venue

Disentangling User Interest and Conformity for Recommendation with Causal Embedding

Yu Zheng, Chen Gao, Xiang Li, Xiangnan He, Yong Li, Depeng Jin

Abstract

Recommendation models are usually trained on observational interaction data. However, observational interaction data could result from users' conformity towards popular items, which entangles users' real interest. Existing methods tracks this problem as eliminating popularity bias, e.g., by re-weighting training samples or leveraging a small fraction of unbiased data. However, the variety of user conformity is ignored by these approaches, and different causes of an interaction are bundled together as unified representations, hence robustness and interpretability are not guaranteed when underlying causes are changing. In this paper, we present DICE, a general framework that learns representations where interest and conformity are structurally disentangled, and various backbone recommendation models could be smoothly integrated. We assign users and items with separate embeddings for interest and conformity, and make each embedding capture only one cause by training with cause-specific data which is obtained according to the colliding effect of causal inference. Our proposed methodology outperforms state-of-the-art baselines with remarkable improvements on two real-world datasets on top of various backbone models. We further demonstrate that the learned embeddings successfully capture the desired causes, and show that DICE guarantees the robustness and interpretability of recommendation.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

57 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Debiased Contrastive Learning for Sequential Recommendation cites AutoDebias: Learning to Debias for Recommendation
  • AutoDebias: Learning to Debias for Recommendation cites LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
  • AutoDebias: Learning to Debias for Recommendation cites Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender System
  • Disentangling Long and Short-Term Interests for Recommendation cites LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
  • Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender System cites LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers57

  1. Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender SystemTianxin Wei, Fuli Feng, Jiawei Chen, Ziwei Wu et al.KDD 2021 · 246 citations
  2. Debiased Contrastive Learning for Sequential RecommendationYuhao Yang, Chao Huang, Lianghao Xia, Chunzhen Huang et al.WWW 2023 · 199 citations
  3. AutoDebias: Learning to Debias for RecommendationJiawei Chen, Hande Dong, Yang Qiu, Xiangnan He et al.SIGIR 2021 · 167 citations
  4. Disentangling Long and Short-Term Interests for RecommendationYu Zheng, Chen Gao, Jianxin Chang, Yanan Niu et al.WWW 2022 · 128 citations
  5. Deciphering Spatio-Temporal Graph Forecasting: A Causal Lens and TreatmentYutong Xia, Yuxuan Liang, Haomin Wen, Xu Liu et al.NeurIPS 2023 · 110 citations

Builds on2

  1. LightGCN: Simplifying and Powering Graph Convolution Network for RecommendationXiangnan He, Kuan Deng, Xiang Wang, Yan Li et al.SIGIR 2020 · 4,448 citations
  2. A Meta-Transfer Objective for Learning to Disentangle Causal MechanismsYoshua Bengio, Tristan Deleu, Nasim Rahaman, Nan Rosemary Ke et al.ICLR 2020 · 371 citations

Related papers

  1. Causal Intervention for Leveraging Popularity Bias in RecommendationYang Zhang, Fuli Feng, Xiangnan He, Tianxin Wei et al.SIGIR 2021 · 431 citations
  2. Modeling Social Behavior in Collaborative FilteringYihong Zhang, Takahiro HaraSIGIR 2025
  3. Taming Recommendation Bias with Causal Intervention on Evolving Personal PopularityShiyin Tan, Dongyuan Li, Renhe Jiang, Zhen Wang et al.KDD 2025 · 1 citation
  4. Invariant Collaborative Filtering to Popularity Distribution ShiftAn Zhang, Jingnan Zheng, Xiang Wang, Yancheng Yuan et al.WWW 2023 · 64 citations
  5. Generalizable Recommender System During Temporal Popularity Distribution ShiftsHyunsik Yoo, Ruizhong Qiu, Charlie Xu, Fei Wang et al.KDD 2025 · 5 citations
Dusk over the sea between two cliffs drawn in fine vertical lines