第30屆國際資訊與知識管理大會（The 30th ACM International Conference on Information and Knowledge Management, CIKM 2021）計劃于2021年11月1日-11月5日線上召開。ACM CIKM是CCF推薦的B類國際學術會議，是資訊檢索和資料挖掘領域最重要的學術會議之一。這次會議共收到1251篇長文（Full paper）、290篇應用文（Applied paper）和626篇短文（Short paper）投稿，有271篇長文、69篇應用文和178篇短文被錄用，錄用率分别為21.7%、23.8%和28.4%。
官方釋出的接收論文清單：http://www.cikm2021.org/accepted-papers
對推薦系統相關論文（76篇）按不同的任務場景和研究話題進行分類整理，也對其他熱門研究方向（預訓練、知識圖譜等，53篇）進行了歸類。可以看到2021年研究方向主要集中在Recommendation、Retrieval和Knowledge Graph三個方向，也包括Pre-trained Language Model、Conversation等NLP方向。
- 主要任務包括：Click-Through Rate、Sequential Recommendation、Knowledge Graph Embedding、User Modeling等；
- 熱門技術包括：Graph Neural Network、Contrastive Learning、Transformer、Attention等，其中基于Graph的任務和技術依舊是2021年的研究熱點。
文章目錄
- 1 按推薦的任務場景劃分
- 2 按推薦的研究話題劃分
- 3 熱門技術在推薦中的應用
- 4 其他研究方向
- 1. 按推薦的任務場景劃分
- 1.1 Click-Through Rate
- 1.2 Collaborative Filtering
- 1.3 Sequential/Session-based Recommendation
- 1.4 Knowledge-Aware Recommendation
- 1.5 Social Recommendation
- 1.6 News Recommendation
- 1.7 Text-Aware Recommendation
- 1.8 Conversational Recommender System
- 1.9 Cross-domain Recommendation
- 1.10 Point-of-Interest
- 1.11 Online Recommendation
- 1.12 Group Recommendation
- 1.13 Other Tasks
- 2. 按推薦的研究話題劃分
- 2.1 Debias in Recommender System
- 2.2 Fairness in Recommender System
- 2.3 Explanation in Recommender System
- 2.4 Cold-start in Recommender System
- 2.5 Ranking in Recommender System
- 2.6 Evaluation
- 2.7 Others
- 3. 熱門技術在推薦中的應用
- 3.1 Graph Neural Network in Recommender System
- 3.2 Contrastive Learning in Recommender System
- 3.3 Reinforcement Learning in Recommender System
- 3.4 Variational Autoencoder in Recommender System
- 3.5 Zero-Shot Learning in Recommender System
- 4. 其他研究方向
- 4.1 Pre-training
- 4.2 Transformer
- 4.3 Knowledge Graph
- 4.5 Data Augmentation
- 4.6 Meta Learning
- 4.7 Few-Shot Learning
1 按推薦的任務場景劃分
Click-Through Rate
Collaborative Filtering
Sequential/Session-based Recommendation
Knowledge-Aware Recommendation
Social Recommendation
News Recommendation
Text-Aware Recommendation
Conversational Recommender System
Cross-domain Recommendation
Point-of-Interest
Online Recommendation
Group Recommendation
2 按推薦的研究話題劃分
Debias in Recommender System
Fairness in Recommender System
Explanation in Recommender System
Cold-start in Recommender System
Ranking in Recommender System
Evaluation
3 熱門技術在推薦中的應用
Graph Neural Network in Recommender System
Contrastive Learning in Recommender System
Reinforcement Learning in Recommender System
Variational Autoencoder in Recommender System
Zero-Shot Learning in Recommender System
4 其他研究方向
Pre-training
Transformer
Knowledge Graph
Multi-Modality
Data Augmentation
Meta Learning
Few-Shot Learning
1. 按推薦的任務場景劃分
1.1 Click-Through Rate
Multi-task Learning for Bias-Free Joint CTR Prediction and Market Price Modeling in Online Advertising【線上廣告無偏差聯合CTR預估和市場價格模組化的多任務學習】
Enhancing Explicit and Implicit Feature Interactions via Information Sharing for Parallel Deep CTR Models【applied paper，用于并行 CTR 的顯式和隐式特征互動增強】
TSI: An Ad Text Strength Indicator using Text-to-CTR and Semantic-Ad-Similarity【applied paper，使用 Text-to-CTR 和 Semantic-Ad-Similarity 的廣告文本強度名額】
One Model to Serve All: Star Topology Adaptive Recommender for Multi-Domain CTR Prediction【applied paper，用于多領域CTR預估的自适應推薦】
Efficient Learning to Learn a Robust CTR Model for Web-scale Online Sponsored Search Advertising【applied paper，用于線上搜尋廣告的CTR模型】
AutoIAS: Automatic Integrated Architecture Searcher for Click-Trough Rate Prediction【CTR預估的自動內建搜尋架構】
Click-Through Rate Prediction with Multi-Modal Hypergraphs【使用多模态超圖的點選率預測】
Open Benchmarking for Click-Through Rate Prediction【開源CTR預估Benchmark】
Disentangled Self-Attentive Neural Networks for Click-Through Rate Prediction【short paper，用于CTR預估的自注意力網絡】
AutoHERI: Automated Hierarchical Representation Integration for Post-Click Conversion Rate Estimation【short paper，用于點選後轉換率估計的分層表示學習】
1.2 Collaborative Filtering
SimpleX: A Simple and Strong Baseline for Collaborative Filtering【将Cosine Contrastive Loss引入協同過濾】
Incremental Graph Convolutional Network for Collaborative Filtering【增量圖卷積神經網絡用于協同過濾】
LT-OCF: Learnable-Time ODE-based Collaborative Filtering【Learnable-Time CF】
CausCF: Causal Collaborative Filtering for Recommendation Effect Estimation【applied paper，因果關系協同過濾用于推薦效果評估】
Vector-Quantized Autoencoder With Copula for Collaborative Filtering【short paper，用于協同過濾的矢量量化自動編碼器】
Anchor-based Collaborative Filtering for Recommender Systems【short paper，Anchor-based推薦系統協同過濾】
1.3 Sequential/Session-based Recommendation
Seq2Bubbles: Region-Based Embedding Learning for User Behaviors in Sequential Recommenders【序列推薦中基于區域的使用者行為Embedding學習】
Enhancing User Interest Modeling with Knowledge-Enriched Itemsets for Sequential Recommendation【序列推薦中使用物品集增強使用者興趣模組化】
Continuous-Time Sequential Recommendation with Temporal Graph Collaborative Transformer【将時序圖協同Transformer用于連續時間序列推薦】
Extracting Attentive Social Temporal Excitation for Sequential Recommendation【提取時序激勵用于序列推薦】
Hyperbolic Hypergraphs for Sequential Recommendation【使用雙曲超圖進行序列推薦】
Learning Dual Dynamic Representations on Time-Sliced User-Item Interaction Graphs for Sequential Recommendation【用于序列推薦的在時間片使用者物品互動圖上的對偶動态表示】
Lightweight Self-Attentive Sequential Recommendation【使用CNN捕獲局部特征，使用Self-Attention捕獲全局特征】
What is Next when Sequential Prediction Meets Implicitly Hard Interaction?【序列預測與互動】
Modeling Sequences as Distributions with Uncertainty for Sequential Recommendation【short paper，序列模組化】
Locker: Locally Constrained Self-Attentive Sequential Recommendation【short paper，局部限制的自注意力序列推薦】
CBML: A Cluster-based Meta-learning Model for Session-based Recommendation【用于會話推薦的基于聚類的元學習】
Self-Supervised Graph Co-Training for Session-based Recommendation【用于會話推薦的自監督圖協同訓練】
1.4 Knowledge-Aware Recommendation
A Knowledge-Aware Recommender with Attention-Enhanced Dynamic Convolutional Network【動态卷積用于知識感覺的推薦】
Entity-aware Collaborative Relation Network with Knowledge Graph for Recommendation【short paper，KG+RS】
Conditional Graph Attention Networks for Distilling and Refining Knowledge Graphs in Recommendation【GNN+KG+RS】
1.5 Social Recommendation
Social Recommendation with Self-Supervised Metagraph Informax Network【使用自監督元圖網絡的社交推薦】
1.6 News Recommendation
WG4Rec: Modeling Textual Content with Word Graph for News Recommendation【使用Word Graph為新聞推薦模組化文本内容】
Popularity-Enhanced News Recommendation with Multi-View Interest Representation【多視角興趣學習的流行度增強的新聞推薦】
Prioritizing Original News on Facebook【applied paper，原創新聞優先級排序】
1.7 Text-Aware Recommendation
Counterfactual Review-based Recommendation【基于評論的反事實推薦】
Review-Aware Neural Recommendation with Cross-Modality Mutual Attention【short paper，文本+RS+跨模态】
1.8 Conversational Recommender System
Popcorn: Human-in-the-loop Popularity Debiasing in Conversational Recommender Systems【采用人在回路方式進行對話推薦系統的流行度去偏】
A Neural Conversation Generation Model via Equivalent Shared Memory Investigation【對話生成】
1.9 Cross-domain Recommendation
Expanding Relationship for Cross Domain Recommendation【擴充跨領域推薦的關系】
Learning Representations of Inactive Users: A Cross Domain Approach with Graph Neural Networks【short paper，跨領域方法結合圖神經網絡用于學習非活躍使用者表示】
Low-dimensional Alignment for Cross-Domain Recommendation【short paper，跨領域推薦的低維對齊】
1.10 Point-of-Interest
Answering POI-recommendation Questions using Tourism Reviews【使用旅遊者評論回答POI問題】
SNPR: A Serendipity-Oriented Next POI Recommendation Model【面向偶然性的POI推薦】
ST-PIL: Spatial-Temporal Periodic Interest Learning for Next Point-of-Interest Recommendation【short paper，用于POI推薦的時空周期興趣學習】
1.11 Online Recommendation
Generative Inverse Deep Reinforcement Learning for Online Recommendation【用于線上推薦的生成式逆強化學習】
1.12 Group Recommendation
Double-Scale Self-Supervised Hypergraph Learning for Group Recommendation【用于群組推薦的自監督超圖學習】
DeepGroup: Group Recommendation with Implicit Feedback【short paper，隐式回報的群組推薦】
1.13 Other Tasks
Learning An End-to-End Structure for Retrieval in Large-Scale Recommendations【在大規模推薦中學習一個端到端的結構用于檢索】
USER: A Unified Information Search and Recommendation Model based on Integrated Behavior Sequence【基于內建行為序列的統一搜尋與推薦模型】
Cross-Market Product Recommendation【跨市場産品推薦】
Multi-hop Reading on Memory Neural Network with Selective Coverage for Medication Recommendation【藥物推薦】
Concept-Aware Denoising Graph Neural Network for Micro-Video Recommendation【用于微視訊推薦的去噪GNN】
2. 按推薦的研究話題劃分
2.1 Debias in Recommender System
CauSeR: Causal Session-based Recommendations for Handling Popularity Bias【short paper，用于流行度去偏的因果關系序列推薦】
Mixture-Based Correction for Position and Trust Bias in Counterfactual Learning to Rank【位置和信任偏差】
Unbiased Filtering of Accidental Clicks in Verizon Media Native Advertising【applied paper，廣告意外點選的無偏過濾】
2.2 Fairness in Recommender System
SAR-Net: A Scenario-Aware Ranking Network for Personalized Fair Recommendation in Hundreds of Travel Scenarios【applied paper，用于個性化公平推薦的場景感覺排名網絡】
2.3 Explanation in Recommender System
Counterfactual Explainable Recommendation【反事實可解釋推薦】
On the Diversity and Explainability of Recommender Systems: A Practical Framework for Enterprise App Recommendation【applied paper，推薦系統的多樣性和可解釋性】
You Are What and Where You Are: Graph Enhanced Attention Network for Explainable POI Recommendation【applied paper，Attention圖神經網絡用于可解釋推薦】
XPL-CF: Explainable Embeddings for Feature-based Collaborative Filtering【short paper，可解釋Embedding用于基于特征的協同過濾】
Grad-SAM: Explaining Transformers via Gradient Self-Attention Maps【short paper，通過梯度Self-Attention解釋Transformer】
2.4 Cold-start in Recommender System
CMML: Contextual Modulation Meta Learning for Cold-Start Recommendation【元學習+冷啟動】
Reinforcement Learning to Optimize Lifetime Value in Cold-Start Recommendation【增強學習+冷啟動】
2.5 Ranking in Recommender System
Top-N Recommendation with Counterfactual User Preference Simulation【反事實使用者偏好模拟的Top-N推薦】
2.6 Evaluation
Evaluating Human-AI Hybrid Conversational Systems with Chatbot Message Suggestions【人機混合對話系統評估】
POSSCORE: A Simple Yet Effective Evaluation of Conversational Search with Part of Speech Labelling【使用部分語音标簽對會話搜尋進行簡單有效的評估】
2.7 Others
DSKReG: Differentiable Sampling on Knowledge Graph for Recommendation with Relational GNN【short paper，用于推薦的知識圖譜采樣】
Disentangling Preference Representations for Recommendation Critiquing with ?-VAE【用于推薦的VAE偏好表示】
3. 熱門技術在推薦中的應用
3.1 Graph Neural Network in Recommender System
UltraGCN: Ultra Simplification of Graph Convolutional Networks for Recommendation【GNN+RS】
How Powerful is Graph Convolution for Recommendation?【GNN+RS】
3.2 Contrastive Learning in Recommender System
Contrastive Curriculum Learning for Sequential User Behavior Modeling via Data Augmentation【applied paper，通過資料增強進行序列使用者行為模組化的對比課程學習】
Graph Structure Aware Contrastive Knowledge Distillation for Incremental Learning in Recommender Systems【short paper，推薦系統中用于增量學習的圖結構感覺的對比知識蒸餾】
3.3 Reinforcement Learning in Recommender System
Explore, Filter and Distill: Distilled Reinforcement Learning in Recommendation【applied paper，推薦中的蒸餾強化學習】
3.4 Variational Autoencoder in Recommender System
Semi-deterministic and Contrastive Variational Graph Autoencoder for Recommendation【用于推薦的半确定性和對比變分圖自動編碼器】
3.5 Zero-Shot Learning in Recommender System
Zero Shot on the Cold-Start Problem: Model-Agnostic Interest Learning for Recommender Systems【零樣本學習+冷啟動】
4. 其他研究方向
4.1 Pre-training
Pre-training for Ad-hoc Retrieval: Hyperlink is Also You Need【Ad-hoc檢索預訓練】
Pulling Up by the Causal Bootstraps: Causal Data Augmentation for Pre-training Debiasing【用于預訓練去偏的因果關系資料增強】
Contrastive Pre-Training of GNNs on Heterogeneous Graphs【圖神經網絡的對比預訓練】
HORNET: Enriching Pre-trained Language Representations with Heterogeneous Knowledge Sources【異構知識來源的預訓練】
WebKE: Knowledge Extraction from Semi-structured Web with Pre-trained Markup Language Model【知識抽取+預訓練】
Natural Language Understanding with Privacy-Preserving BERT【NLU+BERT】
K-AID: Enhancing Pre-trained Language Models with Domain Knowledge for Question Answering【applied paper，QA+領域知識+預訓練】
DialogueBERT: A Self-Supervised Learning based Dialogue Pre-training Encoder【short paper，自監督對話預訓練】
BERT-QPP: Contextualized Pre-trained transformers for Query Performance Prediction【short paper，用于查詢性能預測的上下文預訓練】
CANCN-BERT: A Joint Pre-Trained Language Model for Classical and Modern Chinese【short paper，古典和現代中文的聯合預訓練】
Distilling Knowledge from BERT into Simple Fully Connected Neural Networks for Efficient Vertical Retrieval【applied paper，知識蒸餾+預訓練+檢索】
Adversarial Reprogramming of Pretrained Neural Networks for Fraud Detection【short paper，用于欺詐檢測的預訓練對抗再程式設計】
Adversarial Domain Adaptation for Cross-lingual Information Retrieval with Multilingual BERT【short paper，使用多語言 BERT 進行跨語言資訊檢索的對抗域自适應】
Multi-modal Dictionary BERT for Cross-modal Video Search in Baidu Advertising【applied paper，百度廣告中用于跨模态視訊搜尋的多模态詞典BERT】
RABERT: Relation-Aware BERT for Target-Oriented Opinion Words Extraction【short paper，用于詞提取的關系感覺BERT】
4.2 Transformer
LiteGT: Efficient and Lightweight Graph Transformers【高效輕量化圖Transformer】
Block Access Pattern Discovery via Compressed Full Tensor Transformer【Transformer壓縮】
Mixed Attention Transformer for Leveraging Word-Level Knowledge to Neural Cross-Lingual Information Retrieval【用于跨語言資訊檢索的混合注意力Transformer】
Match-Ignition: Plugging PageRank into Transformer for Long-form Text Matching【PageRank+Transformer】
DCAP: Deep Cross Attentional Product Network for User Response Prediction【用于使用者響應預測的交叉注意力産品網絡】
4.3 Knowledge Graph
Tracking Semantic Evolutionary Changes in Large-Scale Ontological Knowledge Bases【大規模本體知識庫中語義演化的跟蹤】
Cycle or Minkowski: Which is More Appropriate for Knowledge Gragh Embedding?【KG Embedding】
HopfE: Knowledge Graph Representation Learning using Inverse Hopf Fibrations【知識圖譜表示學習】
Automated Query Graph Generation for Querying Knowledge Graphs【用于查詢知識圖譜的自動查詢圖生成】
Differentially Private Federated Knowledge Graphs Embedding【差異化隐私聯邦KG Embedding】
A Lightweight Knowledge Graph Embedding Framework for Efficient Inference and Storage【輕量化KG Embedding】
Predicting Instance Type Assertions in Knowledge Graphs Using Stochastic Neural Networks【知識圖譜中的執行個體類型斷言預測】
When Hardness Makes a Difference: Multi-Hop Knowledge Graph Reasoning over Few-Shot Relations【小樣本關系上的知識圖譜多跳推理】
Query Reformulation for Descriptive Queries of Jargon Words Using a Knowledge Graph based on a Dictionary【使用基于字典的知識圖譜進行查詢重構】
Computing and Maintaining Provenance of Query Result Probabilities in Uncertain Knowledge Graphs【不确定知識圖譜中計算和維護查詢結果機率】
REFORM: Error-Aware Few-Shot Knowledge Graph Completion【錯誤感覺的小樣本知識圖譜補全】
DisenKGAT: Knowledge Graph Embedding with Disentangled Graph Attention Network【KG Embedding+GNN】
Complex Temporal Question Answering on Knowledge Graphs【QA+KG】
Mixed Attention Transformer for Leveraging Word-Level Knowledge to Neural Cross-Lingual Information Retrieval【Transformer+IR】
Knowledge Graph Representation Learning as Groupoid: Unifying TransE, RotatE, QuatE, ComplEx【知識圖譜表示學習】
DataType-Aware Knowledge Graph Representation Learning in Hyperbolic Space【雙曲空間中基于資料類型的知識圖譜表示學習】
Evidential Relational-Graph Convolutional Networks for Entity Classification in Knowledge Graphs【short paper，GNN+KG】
4.4 Multi-Modality
Student Can Also be a Good Teacher: Extracting Knowledge from Vision-and-Language Model for Cross-Modal Retrieval【short paper，用于跨模态檢索的知識提取】
Supervised Contrastive Learning for Multimodal Unreliable News Detection in COVID-19 Pandemic【short paper，用于多模态不可靠新聞檢測的有監督對比學習】
4.5 Data Augmentation
Influence-guided Data Augmentation for Neural Tensor Completion【用于張量補全的資料增強】
Learning to Augment Imbalanced Data for Re-ranking Models【用于再排序模型的資料增強】
Action Sequence Augmentation for Early Graph-based Anomaly Detection【用于異常檢測的動作序列增強】
4.6 Meta Learning
Multimodal Graph Meta Contrastive Learning【short paper，多模态元圖對比學習】
Meta-Learning Based Hyper-Relation Feature Modeling for Out-of-Knowledge-Base Embedding【基于元學習的超關系特征模組化】
HetMAML: Task-Heterogeneous Model-Agnostic Meta-Learning for Few-Shot Learning Across Modalities【Meta Learning+Few-Shot Learning】
Pruning Meta-Trained Networks for On-Device Adaptation【用于裝置自适應的元訓練網絡剪枝】
Meta Hyperparameter Optimization with Adversarial Proxy Subsets Sampling【元超參優化】
4.7 Few-Shot Learning
Behind the Scenes: An Exploration of Trigger Biases Problem in Few-Shot Event Classification【小樣本學習中偏差問題的探讨】
Learning Discriminative and Unbiased Representations for Few-Shot Relation Extraction【用于小樣本關系提取的無偏表示學習】
Multi-view Interaction Learning for Few-Shot Relation Classification【用于小樣本關系分類的多視角互動學習】
One-shot Transfer Learning for Population Mapping【單樣本遷移學習】
![使用try-with-resources優雅關閉資源[圖]](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACwAAAAAAQABAAACAkQBADs=)