SANCL: Multimodal Review Helpfulness Prediction with Selective Attention and Natural Contrastive Learning
Not on the map. Why?
Abstract (arXiv)
With the boom of e-commerce, Multimodal Review Helpfulness Prediction (MRHP), which aims to sort product reviews according to the predicted helpfulness scores has become a research hotspot. Previous work on this task focuses on attention-based modality fusion, information integration, and relation modeling, which primarily exposes the following drawbacks: 1) the model may fail to capture the really essential information due to its indiscriminate attention formulation; 2) lack appropriate modeling methods that take full advantage of correlation among provided data. In this paper, we propose SANCL: Selective Attention and Natural Contrastive Learning for MRHP. SANCL adopts a probe-based strategy to enforce high attention weights on the regions of greater significance. It also constructs a contrastive learning framework based on natural matching properties in the dataset. Experimental results on two benchmark datasets with three categories show that SANCL achieves state-of-the-art baseline performance with lower memory consumption.
Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 19 of its 50 references that are in the map. First posted on arXiv 12 Sep 2022.
Builds on
19 of 50 references.
References · 19
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
- 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
- 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
- 2017Multimodal Machine Learning: A Survey and Taxonomycited by 1,004
- 2019How to Fine-Tune BERT for Text Classification?cited by 409
- 2021Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysiscited by 130
- 2021Multimodal Contrastive Training for Visual Representation Learningcited by 122
- 2017Bilateral Multi-Perspective Matching for Natural Language Sentencescited by 120
- 2020Deep Multimodal Fusion by Channel Exchangingcited by 117
- 2018CentralNet: a Multilayer Approach for Multimodal Fusioncited by 86
- 2020Beneath the Tip of the Iceberg: Current Challenges and New Directions in Sentiment Analysis Researchcited by 67
7 more
- 2019Modality to Modality Translation: An Adversarial Representation Learning and Graph Fusion Network for Multimodal Fusioncited by 45
- 2021Bi-Bimodal Modality Fusion for Correlation-Controlled Multimodal Sentiment Analysiscited by 45
- 2020Multimodal Categorization of Crisis Events in Social Mediacited by 34
- 2022Analyzing Modality Robustness in Multimodal Sentiment Analysiscited by 20
- 2021Identifying Helpful Sentences in Product Reviewscited by 9
- 2019Review-based Question Generation with Adaptive Instance Transfer and Augmentationcited by 8
- 2020Unsupervised Natural Language Inference via Decoupled Multimodal Contrastive Learningcited by 8
Built on by
Citing papers per year: 2022: 2; 2023: 4; 2024: 2; 2025: 2; 2026: 2.
12 papers.
- 2024Towards Unified Multi-Modal Personalization: Large Vision-Language Models for Generative Recommendation and Beyondcited by 25
- 2022Adaptive Contrastive Learning on Multimodal Transformer for Review Helpfulness Predictionscited by 13
- 2023Improving Self-training for Cross-lingual Named Entity Recognition with Contrastive and Prototype Learningcited by 8
- 2023Gradient-Boosted Decision Tree for Listwise Context Model in Multimodal Review Helpfulness Predictioncited by 7
- 2022Not All Neighbors Are Worth Attending to: Graph Selective Attention Networks for Semi-supervised Learningcited by 6
- 2023On the Role of Reviewer Expertise in Temporal Review Helpfulness Predictioncited by 6
- 2023One-Versus-Others Attention: Scalable Multimodal Integration for Biomedical Datacited by 6
- 2024Enhancing Travel Decision-Making: A Contrastive Learning Approach for Personalized Review Rankings in Accommodationscited by 2
- 2026EffiMiniVLM: A Compact Dual-Encoder Regression Frameworkcited by 1
- 2026Bounded-Compute Multimodal Regression for Product-Rating Predictioncited by 1
- 2025PREMISE: Matching-based Prediction for Accurate Review Recommendationcited by 0
- 2025ViMRHP: A Vietnamese Benchmark Dataset for Multimodal Review Helpfulness Prediction via Human-AI Collaborative Annotationcited by 0
- 2026Bounded-Compute Multimodal Regression for Product-Rating Prediction26 May 2026
- 2026EffiMiniVLM: A Compact Dual-Encoder Regression Framework3 Apr 2026
- 2025ViMRHP: A Vietnamese Benchmark Dataset for Multimodal Review Helpfulness Prediction via Human-AI Collaborative Annotation12 May 2025
- 2025PREMISE: Matching-based Prediction for Accurate Review Recommendation2 May 2025
- 2024Enhancing Travel Decision-Making: A Contrastive Learning Approach for Personalized Review Rankings in Accommodations30 Jun 2024
- 2024Towards Unified Multi-Modal Personalization: Large Vision-Language Models for Generative Recommendation and Beyond15 Mar 2024
- 2023One-Versus-Others Attention: Scalable Multimodal Integration for Biomedical Data11 Jul 2023
- 2023Improving Self-training for Cross-lingual Named Entity Recognition with Contrastive and Prototype Learning23 May 2023
- 2023Gradient-Boosted Decision Tree for Listwise Context Model in Multimodal Review Helpfulness Prediction22 May 2023
- 2023On the Role of Reviewer Expertise in Temporal Review Helpfulness Prediction22 Feb 2023
- 2022Adaptive Contrastive Learning on Multimodal Transformer for Review Helpfulness Predictions7 Nov 2022
- 2022Not All Neighbors Are Worth Attending to: Graph Selective Attention Networks for Semi-supervised Learning14 Oct 2022