# M2GRL: A Multi-task Multi-view Graph Representation Learning Framework for Web-scale Recommender Systems

arXiv: https://arxiv.org/abs/2005.10110 · HTML: https://mlsanity.com/paper/2005.10110/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 May 2020.

## Built on by (most cited first, 21 of 21)

- [A Survey on Graph Representation Learning Methods](https://mlsanity.com/paper/2204.01855/) (2022)
- [Fair Resource Allocation in Multi-Task Learning](https://mlsanity.com/paper/2402.15638/) (2024)
- [Region Embedding with Intra and Inter-View Contrastive Learning](https://mlsanity.com/paper/2211.08975/) (2022)
- [Embedding in Recommender Systems: A Survey](https://mlsanity.com/paper/2310.18608/) (2023)
- [Double Correction Framework for Denoising Recommendation](https://mlsanity.com/paper/2405.11272/) (2024)
- [Beyond NDCG: behavioral testing of recommender systems with RecList](https://mlsanity.com/paper/2111.09963/) (2021)
- [Heterogeneous Graph Neural Network with Multi-view Representation Learning](https://mlsanity.com/paper/2108.13650/) (2021)
- [Heterophily-Aware Fair Recommendation using Graph Convolutional Networks](https://mlsanity.com/paper/2402.03365/) (2024)
- [LDC-MTL: Balancing Multi-Task Learning through Scalable Loss Discrepancy Control](https://mlsanity.com/paper/2502.08585/) (2025)
- [Rethinking Missing Data: Aleatoric Uncertainty-Aware Recommendation](https://mlsanity.com/paper/2209.11679/) (2022)
- [Multi-Relational Graph based Heterogeneous Multi-Task Learning in Community Question Answering](https://mlsanity.com/paper/2110.02059/) (2021)
- [Graph Contrastive Learning on Multi-label Classification for Recommendations](https://mlsanity.com/paper/2501.06985/) (2025)
- [View-Invariant Skeleton-based Action Recognition via Global-Local Contrastive Learning](https://mlsanity.com/paper/2209.11634/) (2022)
- [Multi-Relational Graph Neural Network for Out-of-Domain Link Prediction](https://mlsanity.com/paper/2403.11292/) (2024)
- [Hierarchical Structure Sharing Empowers Multi-task Heterogeneous GNNs for Customer Expansion](https://mlsanity.com/paper/2410.22089/) (2024)
- [A Re-node Self-training Approach for Deep Graph-based Semi-supervised Classification on Multi-view Image Data](https://mlsanity.com/paper/2510.24791/) (2025)
- [A Behavior-aware Graph Convolution Network Model for Video Recommendation](https://mlsanity.com/paper/2106.15402/) (2021)
- [Incentive Compatible Pareto Alignment for Multi-Source Large Graphs](https://mlsanity.com/paper/2112.02792/) (2021)
- [Methodologies for Improving Modern Industrial Recommender Systems](https://mlsanity.com/paper/2308.01204/) (2023)
- [Entity Representation Learning Through Onsite-Offsite Graph for Pinterest Ads](https://mlsanity.com/paper/2508.02609/) (2025)
- [Efficient Recommendations via Graph Coarsening and Label Propagation](https://mlsanity.com/paper/2607.22287/) (2026)

## Abstract (arXiv)

Combining graph representation learning with multi-view data (side information) for recommendation is a trend in industry. Most existing methods can be categorized as multi-view representation fusion; they first build one graph and then integrate multi-view data into a single compact representation for each node in the graph. However, these methods are raising concerns in both engineering and algorithm aspects: 1) multi-view data are abundant and informative in industry and may exceed the capacity of one single vector, and 2) inductive bias may be introduced as multi-view data are often from different distributions. In this paper, we use a multi-view representation alignment approach to address this issue. Particularly, we propose a multi-task multi-view graph representation learning framework (M2GRL) to learn node representations from multi-view graphs for web-scale recommender systems. M2GRL constructs one graph for each single-view data, learns multiple separate representations from multiple graphs, and performs alignment to model cross-view relations. M2GRL chooses a multi-task learning paradigm to learn intra-view representations and cross-view relations jointly. Besides, M2GRL applies homoscedastic uncertainty to adaptively tune the loss weights of tasks during training. We deploy M2GRL at Taobao and train it on 57 billion examples. According to offline metrics and online A/B tests, M2GRL significantly outperforms other state-of-the-art algorithms. Further exploration on diversity recommendation in Taobao shows the effectiveness of utilizing multiple representations produced by method, which we argue is a promising direction for various industrial recommendation tasks of different focus.