# Disentangling Long and Short-Term Interests for Recommendation

arXiv: https://arxiv.org/abs/2202.13090 · HTML: https://mlsanity.com/paper/2202.13090/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Feb 2022.

## Built on by (most cited first, 22 of 22)

- [When Search Meets Recommendation: Learning Disentangled Search Representation for Recommendation](https://mlsanity.com/paper/2305.10822/) (2023)
- [Knowledge Prompt-tuning for Sequential Recommendation](https://mlsanity.com/paper/2308.08459/) (2023)
- [SelfGNN: Self-Supervised Graph Neural Networks for Sequential Recommendation](https://mlsanity.com/paper/2405.20878/) (2024)
- [DisenPOI: Disentangling Sequential and Geographical Influence for Point-of-Interest Recommendation](https://mlsanity.com/paper/2210.16591/) (2022)
- [GUESR: A Global Unsupervised Data-Enhancement with Bucket-Cluster Sampling for Sequential Recommendation](https://mlsanity.com/paper/2303.00243/) (2023)
- [Disentangling ID and Modality Effects for Session-based Recommendation](https://mlsanity.com/paper/2404.12969/) (2024)
- [NMA: Neural Multi-slot Auctions with Externalities for Online Advertising](https://mlsanity.com/paper/2205.10018/) (2022)
- [Long Short-Term Preference Modeling for Continuous-Time Sequential Recommendation](https://mlsanity.com/paper/2208.00593/) (2022)
- [Panoramic Interests: Stylistic-Content Aware Personalized Headline Generation](https://mlsanity.com/paper/2501.11900/) (2025)
- [Hierarchically Fusing Long and Short-Term User Interests for Click-Through Rate Prediction in Product Search](https://mlsanity.com/paper/2304.02089/) (2023)
- [CIDGMed: Causal Inference-Driven Medication Recommendation with Enhanced Dual-Granularity Learning](https://mlsanity.com/paper/2403.00880/) (2024)
- [Self-Supervised Interest Transfer Network via Prototypical Contrastive Learning for Recommendation](https://mlsanity.com/paper/2302.14438/) (2023)
- [Denoising Long- and Short-term Interests for Sequential Recommendation](https://mlsanity.com/paper/2407.14743/) (2024)
- [CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning](https://mlsanity.com/paper/2309.04802/) (2023)
- [Heterogeneous Graph-based Framework with Disentangled Representations Learning for Multi-target Cross Domain Recommendation](https://mlsanity.com/paper/2407.00909/) (2024)
- [Adaptive Long-term Embedding with Denoising and Augmentation for Recommendation](https://mlsanity.com/paper/2504.13614/) (2025)
- [A Deep Behavior Path Matching Network for Click-Through Rate Prediction](https://mlsanity.com/paper/2302.00302/) (2023)
- [Intrinsic and Extrinsic Factor Disentanglement for Recommendation in Various Context Scenarios](https://mlsanity.com/paper/2503.03524/) (2025)
- [Enhancing POI Recommendation through Global Graph Disentanglement with POI Weighted Module](https://mlsanity.com/paper/2507.14612/) (2025)
- [Disentangled Interest Network for Out-of-Distribution CTR Prediction](https://mlsanity.com/paper/2602.00002/) (2025)
- [SLSREC: Self-Supervised Contrastive Learning for Adaptive Fusion of Long- and Short-Term User Interests](https://mlsanity.com/paper/2604.04530/) (2026)
- [TME-PSR: Time-aware, Multi-interest, and Explanation Personalization for Sequential Recommendation](https://mlsanity.com/paper/2604.09439/) (2026)

## Abstract (arXiv)

Modeling user's long-term and short-term interests is crucial for accurate recommendation. However, since there is no manually annotated label for user interests, existing approaches always follow the paradigm of entangling these two aspects, which may lead to inferior recommendation accuracy and interpretability. In this paper, to address it, we propose a Contrastive learning framework to disentangle Long and Short-term interests for Recommendation (CLSR) with self-supervision. Specifically, we first propose two separate encoders to independently capture user interests of different time scales. We then extract long-term and short-term interests proxies from the interaction sequences, which serve as pseudo labels for user interests. Then pairwise contrastive tasks are designed to supervise the similarity between interest representations and their corresponding interest proxies. Finally, since the importance of long-term and short-term interests is dynamically changing, we propose to adaptively aggregate them through an attention-based network for prediction. We conduct experiments on two large-scale real-world datasets for e-commerce and short-video recommendation. Empirical results show that our CLSR consistently outperforms all state-of-the-art models with significant improvements: GAUC is improved by over 0.01, and NDCG is improved by over 4%. Further counterfactual evaluations demonstrate that stronger disentanglement of long and short-term interests is successfully achieved by CLSR. The code and data are available at https://github.com/tsinghua-fib-lab/CLSR.