大家好，我是对白。

ACM SIGIR 2022是CCF A类会议，人工智能领域智能信息检索（ Information Retrieval，IR）方向最权威的国际会议。会议专注于信息的存储、检索和传播等各个方面，包括研究战略、输出方案和系统评估等等。第45届国际计算机学会信息检索大会（The 45rd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2022）计划于今年7月11日-7月15日在西班牙马德里召开。这次会议共收到794篇长文和667篇短文投稿，有161篇长文和165篇短文被录用，录用率约为20%和24.7%。官方发布的接收论文列表：

Accepted Paperssigir.org/sigir2022/program/accepted/

本文选取了SIGIR 2022中170篇长文或短文，重点对推荐系统相关论文（124篇）按不同的任务场景和研究话题进行分类整理，也对其他热门研究方向（问答、对话、知识图谱等，46篇）进行了归类，以供参考。文章也同步发布在AI Box知乎专栏（知乎搜索「 AI Box专栏」），整理过程中难免有疏漏，欢迎大家在知乎专栏的文章下方评论留言，交流探讨！

从词云图看今年SIGIR的研究热点：根据长文和短文的标题绘制如下词云图，可以看到今年研究方向依旧集中在Recommendation，也包括Retrieval、Query等方向；主要任务包括：Ranking、Cross-domain、Multi-Model/Behavior、Few-Shot、User modeling、Conversation等；热门技术包括：Neural Networks、Knowledge Graph、GNN、Contrastive Learning、Transformer等，其中基于Graph的方法依旧是今年的研究热点。

本文目录

1 按照任务场景划分

• CTR

• Collaborative Filtering

• Sequential/Session-based Recommendation

• Conversational Recommender System

• POI Recommendation

• Cross-domain/Multi-behavior Recommendation

• Knowledge-aware Recommendation

• News Recommendation

• Others

2 按照主要技术划分

• GNN-based

• RL-based

• Contrastive Learning based

• AutoML-based

• Others

3 按照研究话题划分

• Bias/Debias in Recommender System

• Explanation in Recommender System

• Long-tail/Cold-start in Recommender System

• Fairness in Recommender System

• Diversity in Recommender System

• Attack/Denoise in Recommender System

• Others

4 其他研究方向

• QA

• Knowledge Graph

• Conversation/ Dialog

• Summarization

• Multi-Modality

• Generation

• Representation Learning

1.按照任务场景划分

1.1 CTR /CVR Prediction

1. Enhancing CTR Prediction with Context-Aware Feature Representation Learning 【上下文相关的特征表示】

2. HIEN: Hierarchical Intention Embedding Network for Click-Through Rate Prediction 【层次化意图嵌入网络】

3. NAS-CTR: Efficient Neural Architecture Search for Click-Through Rate Prediction 【高效的网络结构搜索】

4. NMO: A Model-Agnostic and Scalable Module for Inductive Collaborative Filtering 【模型无关的归纳式协同过滤模块】

5. Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer 【图遮盖的Transformer】

6. Neural Statistics for Click-Through Rate Prediction 【short paper，神经统计学】

7. Smooth-AUC: Smoothing the Path Towards Rank-based CTR Prediction 【short paper，基于排序的CTR预估】

8. DisenCTR: Dynamic Graph-based Disentangled Representation for Click-Through Rate Prediction 【基于图的解耦表示】

9. Deep Multi-Representational Item Network for CTR Prediction 【short paper，多重表示商品网络】

10. Gating-adapted Wavelet Multiresolution Analysis for Exposure Sequence Modeling in CTR prediction 【short paper，多分辨率小波分析】

11. MetaCVR: Conversion Rate Prediction via Meta Learning in Small-Scale Recommendation Scenarios 【short paper，小规模推荐场景下的元学习】

12. Adversarial Filtering Modeling on Long-term User Behavior Sequences for Click-Through Rate Prediction 【short paper，对抗过滤建模用户长期行为序列】

13. Clustering based Behavior Sampling with Long Sequential Data for CTR Prediction 【short paper，长序列数据集基于聚类的行为采样】

14. CTnoCVR: A Novelty Auxiliary Task Making the Lower-CTR-Higher-CVR Upper 【short paper，新颖度辅助任务】

1.2 Collaborative Filtering

1. Geometric Disentangled Collaborative Filtering 【几何解耦的协同过滤】

2. Self-Augmented Recommendation with Hypergraph Contrastive Collaborative Filtering 【超图上的对比学习】

3. Investigating Accuracy-Novelty Performance for Graph-based Collaborative Filtering 【图协同过滤在准确度和新颖度上的表现】

4. Unify Local and Global Information for Top-N Recommendation 【综合局部和全局信息】

5. Enhancing Top-N Item Recommendations by Peer Collaboration 【short paper ，同龄人协同】

6. Evaluation of Herd Behavior Caused by Population-scale Concept Drift in Collaborative Filtering 【short paper】

1.3 Sequential/Session-based Recommendations

1. Decoupled Side Information Fusion for Sequential Recommendation 【融合边缘特征的序列推荐】

2. On-Device Next-Item Recommendation with Self-Supervised Knowledge Distillation 【自监督知识蒸馏】

3. Multi-Agent RL-based Information Selection Model for Sequential Recommendation 【多智能体信息选择】

4. An Attribute-Driven Mirroring Graph Network for Session-based Recommendation 【特征驱动的反射图网络】

5. When Multi-Level Meets Multi-Interest: A Multi-Grained Neural Model for Sequential Recommendation 【多粒度网络】

6. Price DOES Matter! Modeling Price and Interest Preferences in Session-based Recommendation 【考虑价格和兴趣的推荐】

7. AutoGSR: Neural Architecture Search for Graph-based Session Recommendation 【面向图会话推荐的网络结构搜索】

8. Ada-Ranker: A Data Distribution Adaptive Ranking Paradigm for Sequential Recommendation 【数据分布自适应排序】

9. Multi-Faceted Global Item Relation Learning for Session-Based Recommendation 【多面全局商品关系学习】

10. ReCANet: A Repeat Consumption-Aware Neural Network for Next Basket Recommendation in Grocery Shopping 【考虑重复消费的网络】

11. Determinantal Point Process Set Likelihood-Based Loss Functions for Sequential Recommendation 【基于DPP的损失函数】

12. Positive, Negative and Neutral: Modeling Implicit Feedback in Session-based News Recommendation 【建模隐式反馈】

13. Coarse-to-Fine Sparse Sequential Recommendation 【short paper，粗到细的稀疏序列化推荐】

14. Dual Contrastive Network for Sequential Recommendation 【short paper，双对比网络】

15. Explainable Session-based Recommendation with Meta-Path Guided Instances and Self-Attention Mechanism 【short paper， 基于元路径指导和自注意力机制的可解释会话推荐】

16. Item-Provider Co-learning for Sequential Recommendation 【short paper，商品-商家一同训练】

17. RESETBERT4Rec: A Pre-training Model Integrating Time And User Historical Behavior for Sequential Recommendation 【short paper，融合时间和用户历史行为的预训练模型】

18. Enhancing Hypergraph Neural Networks with Intent Disentanglement for Session-based Recommendation【short paper，意图解耦增强超图神经网络】

19. CORE: Simple and Effective Session-based Recommendation within Consistent Representation Space 【short paper，在一致表示空间上的简单有效会话推荐】

20. DAGNN: Demand-aware Graph Neural Networks for Session-based Recommendation 【short paper， 需求感知的图神经网络】

21. Progressive Self-Attention Network with Unsymmetrical Positional Encoding for Sequential Recommendation 【short paper，使用非对称位置编码的自注意力网络】

22. ELECRec: Training Sequential Recommenders as Discriminators 【short paper，训练序列推荐模型作为判别器】

23. Exploiting Session Information in BERT-based Session-aware Sequential Recommendation 【short paper，在基于BERT的模型中利用会话信息】

1.4 Conversational Recommender System

1. Learning to Infer User Implicit Preference in Conversational Recommendation 【学习推测用户隐偏好】

2. User-Centric Conversational Recommendation with Multi-Aspect User Modeling 【多角度用户建模】

3. Variational Reasoning about User Preferences for Conversational Recommendation 【用户偏好的变分推理】

4. Analyzing and Simulating User Utterance Reformulation in Conversational Recommender Systems 【对话推荐中模仿用户言论】

5. Improving Conversational Recommender Systems via Transformer-based Sequential Modelling【short paper，基于Transformer的序列化建模】

6. Conversational Recommendation via Hierarchical Information Modeling 【short paper，层次化信息建模】

1.5 POI Recommendation

1. Hierarchical Multi-Task Graph Recurrent Network for Next POI Recommendation 【多任务图循环网络】

2. Learning Graph-based Disentangled Representations for Next POI Recommendation 【学习基于图的解耦表示】

3. GETNext: Trajectory Flow Map Enhanced Transformer for Next POI Recommendation 【轨迹图加强的Transformer】

4. Next Point-of-Interest Recommendation with Auto-Correlation Enhanced Multi-Modal Transformer Network 【short paper，自修正的多模态Transformer】

5. Empowering Next POI Recommendation with Multi-Relational Modeling 【多重关系建模】

1.6 Cross-domain/Multi-behavior Recommendation

1. Co-training Disentangled Domain Adaptation Network for Leveraging Popularity Bias in Recommenders 【训练解耦的域适应网络来利用流行度偏差】

2. DisenCDR: Learning Disentangled Representations for Cross-Domain Recommendation 【解耦表示】

3. Doubly-Adaptive Reinforcement Learning for Cross-Domain Interactive Recommendation 【双重适应的强化学习】

4. Exploiting Variational Domain-Invariant User Embedding for Partially Overlapped Cross Domain Recommendation 【域不变的用户嵌入】

5. Multi-Behavior Sequential Transformer Recommender 【多行为序列化Transformer】

1.7 Knowledge-aware Recommendation

1. Knowledge Graph Contrastive Learning for Recommendation 【知识图谱上的对比学习】

2. Multi-level Cross-view Contrastive Learning for Knowledge-aware Recommender System 【多级交叉视图的对比学习】

3. Alleviating Spurious Correlations in Knowledge-aware Recommendations through Counterfactual Generator 【利用反事实生成器缓解假知识】

4. HAKG: Hierarchy-Aware Knowledge Gated Network for Recommendation 【层次化知识门控网络】

5. KETCH: Knowledge Graph Enhanced Thread Recommendation in Healthcare Forums 【医疗论坛上的知识图谱增强的推荐】

1.8 News Recommendation

1. ProFairRec: Provider Fairness-aware News Recommendation 【商家公平的新闻推荐】
2. 
   
3. Positive, Negative and Neutral: Modeling Implicit Feedback in Session-based News Recommendation 【建模隐式反馈】
4. 
   
5. FUM: Fine-grained and Fast User Modeling for News Recommendation 【short paper，细粒度快速的用户建模】
6. 
   
7. Is News Recommendation a Sequential Recommendation Task? 【short paper，新闻推荐是序列化推荐吗】
8. 
   
9. News Recommendation with Candidate-aware User Modeling 【short paper，候选感知的用户建模】
10. 
    
11. MM-Rec: Visiolinguistic Model Empowered Multimodal News Recommendation 【short paper，视觉语言学增强的多模态新闻推荐】

1.9 others

1. CAPTOR: A Crowd-Aware Pre-Travel Recommender System for Out-of-Town Users 【为乡村用户提供的旅游推荐】

2. PERD: Personalized Emoji Recommendation with Dynamic User Preference 【short paper，个性化表情推荐】

3. Item Similarity Mining for Multi-Market Recommendation 【short paper，多市场推荐中的商品相似度挖掘】

4. A Content Recommendation Policy for Gaining Subscribers 【short paper，为提升订阅者的内容推荐策略】

5. Thinking inside The Box: Learning Hypercube Representations for Group Recommendation 【超立方体表示用于组推荐】