# Multi-Pointer Co-Attention Networks for Recommendation

arXiv: https://arxiv.org/abs/1801.09251 · HTML: https://mlsanity.com/paper/1801.09251/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Jan 2018.

## Built on by (most cited first, 25 of 25)

- [A Review of the Gumbel-max Trick and its Extensions for Discrete Stochasticity in Machine Learning](https://mlsanity.com/paper/2110.01515/) (2021)
- [A General Survey on Attention Mechanisms in Deep Learning](https://mlsanity.com/paper/2203.14263/) (2022)
- [Multiplex Behavioral Relation Learning for Recommendation via Memory Augmented Transformer Network](https://mlsanity.com/paper/2110.04002/) (2021)
- [A Unified Framework for Cross-Domain and Cross-System Recommendations](https://mlsanity.com/paper/2108.07976/) (2021)
- [Attribute-Guided Multi-Level Attention Network for Fine-Grained Fashion Retrieval](https://mlsanity.com/paper/2301.13014/) (2022)
- [GaVaMoE: Gaussian-Variational Gated Mixture of Experts for Explainable Recommendation](https://mlsanity.com/paper/2410.11841/) (2024)
- [Review-Based Domain Disentanglement without Duplicate Users or Contexts for Cross-Domain Recommendation](https://mlsanity.com/paper/2110.12648/) (2021)
- [Knowledge-aware Collaborative Filtering with Pre-trained Language Model for Personalized Review-based Rating Prediction](https://mlsanity.com/paper/2308.02555/) (2023)
- [Counterfactual Inference for Eliminating Sentiment Bias in Recommender Systems](https://mlsanity.com/paper/2505.03655/) (2025)
- [Enhancing Talent Search Ranking with Role-Aware Expert Mixtures and LLM-based Fine-Grained Job Descriptions](https://mlsanity.com/paper/2512.00004/) (2025)
- [SIFN: A Sentiment-aware Interactive Fusion Network for Review-based Item Recommendation](https://mlsanity.com/paper/2108.08022/) (2021)
- [On Faithfulness and Coherence of Language Explanations for Recommendation Systems](https://mlsanity.com/paper/2209.05409/) (2022)
- [Attentive Graph-based Text-aware Preference Modeling for Top-N Recommendation](https://mlsanity.com/paper/2305.12976/) (2023)
- [Empowering Few-Shot Recommender Systems with Large Language Models -- Enhanced Representations](https://mlsanity.com/paper/2312.13557/) (2023)
- [Recommend for a Reason: Unlocking the Power of Unsupervised Aspect-Sentiment Co-Extraction](https://mlsanity.com/paper/2109.03821/) (2021)
- [Explainable Recommender with Geometric Information Bottleneck](https://mlsanity.com/paper/2305.05331/) (2023)
- [Unifying gradient regularization for Heterogeneous Graph Neural Networks](https://mlsanity.com/paper/2305.15811/) (2023)
- [DEPHN: Different Expression Parallel Heterogeneous Network using virtual gradient optimization for Multi-task Learning](https://mlsanity.com/paper/2307.12519/) (2023)
- [Enriching User Shopping History: Empowering E-commerce with a Hierarchical Recommendation System](https://mlsanity.com/paper/2403.12096/) (2024)
- [Review-Based Hyperbolic Cross-Domain Recommendation](https://mlsanity.com/paper/2403.20298/) (2024)
- [Semantic-Enhanced Relational Metric Learning for Recommender Systems](https://mlsanity.com/paper/2406.10246/) (2024)
- [An Interpretable Alternative to Neural Representation Learning for Rating Prediction -- Transparent Latent Class Modeling of User Reviews](https://mlsanity.com/paper/2407.00063/) (2024)
- [Leveraging User-Generated Reviews for Recommender Systems with Dynamic Headers](https://mlsanity.com/paper/2409.07627/) (2024)
- [Compositions of Variant Experts for Integrating Short-Term and Long-Term Preferences](https://mlsanity.com/paper/2506.23170/) (2025)
- [EviSnap: Faithful Evidence-Cited Explanations for Cold-Start Cross-Domain Recommendation](https://mlsanity.com/paper/2604.06172/) (2026)

## Abstract (arXiv)

Many recent state-of-the-art recommender systems such as D-ATT, TransNet and DeepCoNN exploit reviews for representation learning. This paper proposes a new neural architecture for recommendation with reviews. Our model operates on a multi-hierarchical paradigm and is based on the intuition that not all reviews are created equal, i.e., only a select few are important. The importance, however, should be dynamically inferred depending on the current target. To this end, we propose a review-by-review pointer-based learning scheme that extracts important reviews, subsequently matching them in a word-by-word fashion. This enables not only the most informative reviews to be utilized for prediction but also a deeper word-level interaction. Our pointer-based method operates with a novel gumbel-softmax based pointer mechanism that enables the incorporation of discrete vectors within differentiable neural architectures. Our pointer mechanism is co-attentive in nature, learning pointers which are co-dependent on user-item relationships. Finally, we propose a multi-pointer learning scheme that learns to combine multiple views of interactions between user and item. Overall, we demonstrate the effectiveness of our proposed model via extensive experiments on 24 benchmark datasets from Amazon and Yelp. Empirical results show that our approach significantly outperforms existing state-of-the-art, with up to 19% and 71% relative improvement when compared to TransNet and DeepCoNN respectively. We study the behavior of our multi-pointer learning mechanism, shedding light on evidence aggregation patterns in review-based recommender systems.