对话推荐系统相关论文分类整理

  © 作者｜尚琛展

  机构｜中国人民大学信息学院硕士  

 研究方向 | 推荐系统  

本文主要总结了对话推荐系统领域相关论文。

引言：对话推荐系统（Conversational Recommender System, CRS）是一种能够通过基于自然语言的多轮实时交互，来引出用户动态偏好，并根据他们当前的需求采取行动的推荐系统 (Gao et al., 2021)。本文主要选取了近年来基于深度学习的对话推荐系统相关论文（共 62 篇），并根据工作的类型进行分类，以供参考。

本文整理的论文列表已经同步更新到 GitHub，也会进行持续的更新，欢迎大家关注和 Star。

https://github.com/RUCAIBox/CRSPapers

同时欢迎大家使用 CRSLab 对话推荐系统工具包。CRSLab 是一个用于构建对话推荐系统（CRS）的开源工具包，其基于 PyTorch 实现、主要面向研究者使用，具有如下特性：全面的基准模型和数据集、大规模的标准评测、通用和可扩展的结构、便捷的使用方法、人性化的人机交互接口。

https://github.com/RUCAIBox/CRSLab

为方便读者进一步了解相关工作，本文尽可能直接给出了相关论文的 PDF 链接、官方主页或者官方代码实现，并标注在论文名称后面。但由于公众号推文不支持外部链接，请点击阅读原文到知乎或者到 GitHub 页面查看链接。

目录

• 快速入门

• 综述和教程

• 工具包和数据集

• 模型

• 其他工作

• 学位论文

1 快速入门

快速入门论文列表包括 10 篇综述、教程、工具包和代表性的模型论文。

1. "Deep Conversational Recommender Systems: A New Frontier for Goal-Oriented Dialogue Systems". arXiv(2020) [PDF]

2. "Tutorial on Conversational Recommendation Systems". RecSys(2020) [PDF] [Homepage]

3. CRSLab: "CRSLab: An Open-Source Toolkit for Building Conversational Recommender System". ACL(2021) [PDF] [Homepage]

4. CRM: "Conversational Recommender System". SIGIR(2018) [PDF] [Homepage]

5. SAUR: "Towards Conversational Search and Recommendation: System Ask, User Respond". CIKM(2018) [PDF] [Dataset]

6. EAR: "Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems". WSDM(2020) [PDF] [Homepage]

7. CPR: "Interactive Path Reasoning on Graph for Conversational Recommendation". KDD(2020) [PDF] [Homepage]

8. ReDial: "Towards Deep Conversational Recommendations". NeurIPS(2018) [PDF] [Dataset] [Code]

9. KBRD: "Towards Knowledge-Based Recommender Dialog System". EMNLP-IJCNLP(2019) [PDF] [Code]

10. KGSF: "Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion". KDD(2020) [PDF] [Code]

2 综述和教程

2.1 综述文章（Survey）

1. "Deep Conversational Recommender Systems: A New Frontier for Goal-Oriented Dialogue Systems". arXiv(2020) [PDF]

2. "A survey on conversational recommender systems". arXiv(2020) [PDF]

3. "Advances and Challenges in Conversational Recommender Systems: A Survey". arXiv(2021) [PDF]

2.2 教程（Tutorial）

1. "Tutorial on Conversational Recommendation Systems". [Homepage]

• RecSys(2020) [PDF]

• WSDM(2021) [PDF]

• IUI(2021) [PDF]

• "Conversational Recommendation: Formulation, Methods, and Evaluation". SIGIR(2020) [PDF] [Slides]

• 3 工具包和数据集

  3.1 工具包（Toolkit）

  1. CRSLab: "CRSLab: An Open-Source Toolkit for Building Conversational Recommender System". ACL(2021) [PDF] [Homepage]

  3.2 数据集（Dataset）

  1. ConvRec: "Conversational Recommender System". SIGIR(2018) [PDF] [Homepage]

  2. SAUR: "Towards Conversational Search and Recommendation: System Ask, User Respond". CIKM(2018) [PDF] [Download]

  3. EAR: "Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems". WSDM(2020) [PDF] [Homepage]

  4. CPR: "Interactive Path Reasoning on Graph for Conversational Recommendation". KDD(2020) [PDF] [Homepage]

  5. ReDial: "Towards Deep Conversational Recommendations". NeurIPS(2018) [PDF] [Homepage]

  6. OpenDialKG: "OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs". ACL(2019) [PDF] [Homepage]

  7. PersuasionForGood: "Persuasion for Good: Towards a Personalized Persuasive Dialogue System for Social Good". ACL(2019) [PDF] [Homepage]

  8. CCPE: "Coached Conversational Preference Elicitation: A Case Study in Understanding Movie Preferences". SIGDial(2019) [PDF] [Homepage]

  9. TG-ReDial: "Towards Topic-Guided Conversational Recommender System". COLING(2020) [PDF] [Homepage]

  10. GoRecDial: "Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue". EMNLP(2019) [PDF] [Download]

  11. DuRecDial: "Towards Conversational Recommendation over Multi-Type Dialogs". ACL(2020) [PDF] [Download]

  12. INSPIRED: "INSPIRED: Toward Sociable Recommendation Dialogue Systems". EMNLP(2020) [PDF] [Homepage]

  13. MGConvRex: "User Memory Reasoning for Conversational Recommendation". ACL(2020) [PDF]

  14. COOKIE: "COOKIE: A Dataset for Conversational Recommendation over Knowledge Graphs in E-commerce". arXiv(2020) [PDF] [Homepage]

  15. IARD: "Predicting User Intents and Satisfaction with Dialogue-based Conversational Recommendations". UMAP(2020) [PDF] [Homepage]

  4 模型

  4.1 基于属性的模型

  基于属性的对话推荐系统通过询问物品属性相关问题获取用户偏好，并利用预定义的模板生成回复。

  1. "Towards Conversational Recommender Systems". KDD(2016) [PDF]

  2. CRM: "Conversational Recommender System". SIGIR(2018) [PDF] [Homepage]

  3. SAUR: "Towards Conversational Search and Recommendation: System Ask, User Respond". CIKM(2018) [PDF] [Dataset]

  4. Q&R: "Q&R: A Two-Stage Approach toward Interactive Recommendation". KDD(2018) [PDF]

  5. "Dialogue based recommender system that flexibly mixes utterances and recommendations". WI(2019) [Link]

  6. EAR: "Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems". WSDM(2020) [PDF] [Homepage]

  7. CPR: "Interactive Path Reasoning on Graph for Conversational Recommendation". KDD(2020) [PDF] [Homepage]

  8. CRSAL: "CRSAL: Conversational Recommender Systems with Adversarial Learning". TOIS(2020) [PDF] [Code]

  9. Qrec: "Towards Question-Based Recommender Systems". SIGIR(2020) [PDF] [Code]

  10. ConTS: "Seamlessly Unifying Attributes and Items: Conversational Recommendation for Cold-Start Users". TOIS(2021) [PDF] [Code]

  11. UNICORN: "Unified Conversational Recommendation Policy Learning via Graph-based Reinforcement Learning". SIGIR(2021) [PDF]

  12. KBQG: "Learning to Ask Appropriate Questions in Conversational Recommendation". arXiv(2021) [PDF] [Code]

  13. "Adapting User Preference to Online Feedback in Multi-round Conversational Recommendation". WSDM(2021) [Link]

  14. "Developing a Conversational Recommendation System for Navigating Limited Options". CHI(2021) [PDF]

  4.2 基于生成的模型

  相对于上一类模型，基于生成的对话推荐系统更关注于生成基于自然语言的流利对话文本。

  1. ReDial: "Towards Deep Conversational Recommendations". NeurIPS(2018) [PDF] [Code] [Dataset]

  2. KBRD: "Towards Knowledge-Based Recommender Dialog System". EMNLP-IJCNLP(2019) [PDF] [Code]

  3. GoRecDial: "Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue". EMNLP(2019) [PDF] [Code] [Dataset]

  4. DialKG Walker: "OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs". ACL(2019) [PDF] [Code] [Dataset]

  5. DCR: "Deep Conversational Recommender in Travel". TKDE(2020) [PDF] [Code]

  6. KGSF: "Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion". KDD(2020) [PDF] [Code]

  7. MGCG: "Towards Conversational Recommendation over Multi-Type Dialogs". ACL(2020) [PDF] [Code] [Dataset]

  8. ECR: "Towards Explainable Conversational Recommendation". IJCAI(2020) [PDF]

  9. INSPIRED: "INSPIRED: Toward Sociable Recommendation Dialogue Systems". EMNLP(2020) [PDF] [Homepage]

  10. TG-ReDial: "Towards Topic-Guided Conversational Recommender System". COLING(2020) [PDF] [Homepage]

  11. MGConvRex: "User Memory Reasoning for Conversational Recommendation". COLING(2020) [PDF]

  12. KGConvRec: "Suggest me a movie for tonight: Leveraging Knowledge Graphs for Conversational Recommendation". COLING(2020) [PDF] [Code]

  13. CR-Walker: "Bridging the Gap between Conversational Reasoning and Interactive Recommendation". arXiv(2020) [PDF] [Code]

  14. RevCore: "RevCore: Review-augmented Conversational Recommendation". arXiv(2021) [PDF] [Code]

  15. KECRS: "KECRS: Towards Knowledge-Enriched Conversational Recommendation System". arXiv(2021) [PDF]

  16. "Category Aware Explainable Conversational Recommendation". arXiv(2021) [PDF]

  4.3 其他模型

  1. Converse-Et-Impera: "Converse-Et-Impera: Exploiting Deep Learning and Hierarchical Reinforcement Learning for Conversational Recommender Systems". AI*IA(2017) [PDF]

  2. "A Model of Social Explanations for a Conversational Movie Recommendation System". HAI(2019) [PDF]

  3. "Dynamic Online Conversation Recommendation". ACL(2020) [PDF] [Code]

  4. IAI MovieBot: "IAI MovieBot: A Conversational Movie Recommender System". CIKM(2020) [PDF] [Code]

  5. ConUCB: "Conversational Contextual Bandit: Algorithm and Application". WWW(2020) [PDF] [Code]

  6. Cora: "A Socially-Aware Conversational Recommender System for Personalized Recipe Recommendations". HAI(2020) [PDF]

  7. "Conversational Music Recommendation based on Bandits". ICKG(2020) [Link]

  8. n-by-p: "Navigation-by-preference: a new conversational recommender with preference-based feedback". IUI(2020) [PDF]

  9. "A Bayesian Approach to Conversational Recommendation Systems". AAAI Workshop(2020) [PDF]

  5 其他工作

  1. CCPE: "Coached Conversational Preference Elicitation: A Case Study in Understanding Movie Preferences". SIGDial(2019) [PDF] [Dataset]

  2. "Leveraging Historical Interaction Data for Improving Conversational Recommender System". CIKM(2020) [PDF] [Code]

  3. "Evaluating Conversational Recommender Systems via User Simulation". KDD(2020) [PDF] [Code]

  4. "End-to-End Learning for Conversational Recommendation: A Long Way to Go?". RecSys(2020) [PDF] [Material]

  5. "What Does BERT Know about Books, Movies and Music? Probing BERT for Conversational Recommendation". RecSys(2020) [PDF] [Code]

  6. "Latent Linear Critiquing for Conversational Recommender Systems". WWW(2020) [PDF] [Code]

  7. "A Ranking Optimization Approach to Latent Linear Critiquing for Conversational Recommender Systems". RecSys(2020) [Link] [Code]

  8. "A Comparison of Explicit and Implicit Proactive Dialogue Strategies for Conversational Recommendation". LREC(2020) [PDF]

  9. "Predicting User Intents and Satisfaction with Dialogue-based Conversational Recommendations". UMAP(2020) [PDF] [Dataset]

  10. ConveRSE: "Conversational Recommender Systems and natural language: A study through the ConveRSE framework". Decision Support Systems(2020) [Link] [Dataset]

  11. "On Estimating the Training Cost of Conversational Recommendation Systems". arXiv(2020) [PDF]

  6 学位论文

  1. "Recommendation in Dialogue Systems". By Yueming Sun (2019). [PDF]

  2. "Advanced Method Towards Conversational Recommendation". By Yisong Miao (2020). [PDF]

  参考文献

  Chongming Gao, Wenqiang Lei, Xiangnan He, Maarten de Rijke, and Tat-Seng Chua. 2021. Advances and challenges in conversational recommender systems: A survey. arXiv preprint arXiv:2101.09459 (2021).