日拱一卒

平时的阅读列表，一般会附上下载链接和关键词。日拱一卒，与诸君共勉。

2025

• M4
  • [Promotion, Marketing, Coupon Allocation, Uplift, Mercari, KDD2024] Optimizing Item-based Marketing Promotion Efficiency in C2C Marketplace with Dynamic Sequential Coupon Allocation Framework
  • [Promotion, Marketing, Coupon Allocation, Reinforcement Learning, AntGroup, Alibaba, CIKM2019] Model-based Constrained MDP for Budget Allocation in Sequential Incentive Marketing
  • [Advertising, User Engagement, Ads Allocation, Uplift, Meta, CIKM2024] Ads Supply Personalization via Doubly Robust Learning
  • [Advertising, User Engagement, Ads Allocation, Reinforcement Learning, Meta, KDD2025(pre)] Session-Level Dynamic Ad Load Optimization using Offline Robust Reinforcement Learning
• M3
  • [Advertising, Pacing, Preloaded Ads, Offline Reinforcement Learning, Alibaba, KDD2023] RLTP- Reinforcement Learning to Pace for Delayed Impression Modeling in Preloaded Ads
  • [Marketing, Promotion, Reinforcement Learning, Alibaba] BCRLSP: An Offline Reinforcement Learning Framework for Sequential Targeted Promotion
  • [Recommendation, Offline Reinforcement Learing, AntGroup, SIGIR2023] Model-free Reinforcement Learning with Stochastic Reward Stabilization for Recommender Systems
  • [Recommendation, Offline Reinforcement Learning, LTV, User Engagement, Kuaishou, KDD2023] PrefRec: Recommender Systems with Human Preferences for Reinforcing Long-term User Engagement
• M2
  • [Offline Reinforcement Learning] Offline Reinforcement Learning with Implicit Q-Learning
  • [Advertising, Bidding, Offline Reinforcement Learning, Meta, KDD2024] Offline Reinforcement Learning for Optimizing Production Bidding Policies
  • [Advertising, Offline Reinforcement Learning, User Engagement, ManagementScience2023] Optimizing User Engagement Through Adaptive Ad Sequencing
• M1
  • [Recommendation, Participation, Short-Form Video Recommendation, Google, RecSys2024] Optimizing for Participation in Recommender System
  • [Recommendation, Sequence Model, Google, RecSys2024] Short-form Video Needs Long-term Interests: An Industrial Solution for Serving Large User Sequence Models

2024

• M12
  • [Recommendation System, Multi-Task, Short-Form Video Recommendation, Google, RecSys2024] Co-optimize Content Generation and Consumption in a Large Scale Video Recommendation System
• M11
  • [Recommendation System, Reinforcement Learning, Off-Policy Selection, KDD2024] Off-Policy Selection for Optimizing Ad Display Timing in Mobile Games (Samsung Instant Plays)
• M10（碌碌无为）
• M9
  • [Recommendation System, CTR, Machine Learning, LinkedIn, KDD 2024] [PaperNote] LiRank: Industrial Large Scale Ranking Models at LinkedIn
  • [Causal Inference, Uplift, Marketing, Decision, Meituan, KDD2024] Decision Focused Causal Learning for Direct Counterfactual Marketing Optimization
• M8
  • [Causal Inference, Uplift, Marketing, Uber, KDD 2024 Workshop-Causal ML] Practical Marketplace Optimization at Uber Using Causally-Informed Machine Learning
  • [Causal Inference, Uplift, Marketing, Tencent Fit, RecSys 2024 ] End-to-End Cost-Effective Incentive Recommendation under Budget Constraint with Uplift Modeling
• M7
  • [Causal Inference, Survey] Causal Inference with Complex Treatments- A Survey
  • [Causal Inference, Uplift, Marketing, Mettuan, AAAI 2023] Direct Heterogeneous Causal Learning for Resource Allocation Problems in Marketing
  • [Advertising, Ads Allocation, ShareChat, WSDM 2024] Ad-load Balancing via Off-policy Learning in a Content Marketplace
• M6（碌碌无为）
• M5（碌碌无为）
• M4
  • [Advertising, Pre-Rank, Ranking Consistency, Alibaba, CIKM 2023] COPR: Consistency-Oriented Pre-Ranking for Online Advertising
• M3
  • [Recommendation, Fresh Content, Uncertainty, Google, WSDM2024] Long-Term Value of Exploration: Measurements, Findings and Algorithms
  • [Advertising, Budget Pacing, eBay, AdKdd 2023] Practical Budget Pacing Algorithms and Simulation Test Bed for eBay Marketplace Sponsored Search
• M2
  • [Recommendation, Pre-Rank, Ranking Consistency, Meta, AdKdd 2023] Towards the Better Ranking Consistency- A Multi-task Learning Framework for Early Stage Ads Ranking
  • [Recommendation, Rank, MTL, Debias, Imblance Learning, Short-Form Video Recommendation, Google, CIKM2023] Multitask Ranking System for Immersive Feed and No More Clicks: A Case Study of Short-Form Video Recommendation
• M1（碌碌无为）

---

2023

• M12（碌碌无为）
• M11
  • [Recommendation, Retrieval, Match, CIKM2023] Batch-Mix Negative Sampling for Learning Recommendation Retrievers
  • [Causal Inference, Uplift, Quantile Regression, ByteDance, KDD2023] DNet- Distributional Network for Distributional Individualized Treatment Effects
  • [Causal Inference, Uplift, Marketing, Optimization, PID, Meituan, KDD2023] A Multi-stage Framework for Online Bonus Allocation Based on Constrained User Intent Detection
• M10
  • [Recommendation, Retrieval, Match, Youtube, Google, RecSys2019] Sampling-Bias-Corrected Neural Modeling for Large Corpus Item Recommendations
• M9
  • [Recommendation, Retrieval, Match, Twitter, AdKdd 2023] TwERC: High Performance Ensembled Candidate Generation for Ads Recommendation at Twitter
  • [Recommendation, Fresh Content, Spotify, RecSys2023] Accelerating Creator Audience Building through Centralized Exploration
  • [Recommendation, Retrieval, Match, Youtube, Google, WWW2020] Mixed Negative Sampling for Learning Two-tower Neural Networks in Recommendations
• M8
  • [Machine Learning, Recommendation System, CTR, ODL, WeChat] [PaperNote] Streaming CTR Prediction- Rethinking Recommendation Task for Real-World Streaming Data
  • [Marketing, Uplift, Causal Inference, Multi-Task, KDD 2023] Who Should Be Given Incentives? Counterfactual Optimal Treatment Regimes Learning for Recommendation
  • [Recommendation, Retrieval, Match, Debias, Pinterest, KDD 2023] [PaperNote] An Empirical Study of Selection Bias in Pinterest Ads Retrieval
• M7
  • [Marketing, Uplift, Subsidy, Didi, SIGIR 2023] A Consumer Compensation System in Ride-hailing Service
• M6
  • [Advertising, CTR, Debias, Alibaba] Rec4Ad: A Free Lunch to Mitigate Sample Selection Bias for Ads CTR Prediction in Taobao
  • [Recommendation, Fresh Content, Youtube, KDD 2023] [PaperNote] Fresh Content Needs More Attention: Multi-funnel Fresh Content Recommendation
  • [Recommendation, Debias, Kuaishou, KDD 2022] Deconfounding Duration Bias in Watch-time Prediction for Video Recommendation
  • [Recommendation, Debias, Long-tail, Google, KDD 2023] Empowering Long-tail Item Recommendation through Cross Decoupling Network (CDN)
• M5
  • [NLP, Deep Learning, Google, NeurIPS 2017] [PaperNote] Attention is All You Need
• M4
  • [Advertising, Bidding, LinkedIn, AdKdd 2022] [TODO PaperNote] Bidding Agent Design in the LinkedIn Ad Marketplace
  • [Marketing, Causal Inference, Optimization, Kuaishou] An End-to-End Framework for Marketing Effectiveness Optimization under Budget Constraint
  • [Experimentation, Policy Optimization, Meta] Practical Policy Optimization with Personalized Experimentation

PS.

过去一段时间看的文章比较少，主要是因为老板的一些建议，少花些精力在技术上，多关注一下身边事和人。

个人还是非常认可老板的建议的，只会做技术，职业生涯怎么往前走呢。

以后会多写总结、多输出观点，少写翻译和技术学习。

2022

• M11

  • [Machine Learning, Deep Learning, Google] A Review of Sparse Expert Models in Deep Learning
  • [Recommendation System, CTR, Deep Learning, Google, WWW 2021] [TODO PaperNote] DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems

• M10

  • [Advertising, CTR, Machine Learning, Deep Learning, Google, RecSys 2022] [PaperNote] On the Factory Floor: ML Engineering for Industrial-Scale Ads Recommendation Models

• M9

  • [Advertising, Budget Control, Pacing, LinkedIn, CIKM 2020] [PaperNote] Impression Pacing for Jobs Marketplace at LinkedIn

• M8

  • [Causal Inference, Dynamic Pricing, ITE, Amazon, KDD2022] [TODO PaperNote] ASPIRE: Air Shipping Recommendation for E-commerce Products via Causal Inference Framework
  • [Recommendation System, Advertising, Multi-Task Learning, Google, KDD2022] An Online Multi-task Learning Framework for Google Feed Ads Auction Models

• M7

  • [Advertising, Bidding, Alibaba, CIKM2018] Budget Constrained Bidding by Model-free Reinforcement Learning in Display Advertising
  • [Causal Inference, Individual Treatment Effect] [PaperNote] What Makes Forest-Based Heterogeneous Treatment Effect Estimators Work?

• M6

  • [Advertising, Bidding, Alibaba, KDD2021] [PaperNote] A Unified Solution to Constrained Bidding in Online Display Advertising
  • [Advertising, Ads Allocation, LinkedIn, KDD2020] [PaperNote] Ads Allocation in Feed via Constrained Optimization

• M5

  • [Advertising, Calibration, Tencent, WWW2020] [PaperNote] Field-aware Calibration- A Simple and Empirically Strong Method for Reliable Probabilistic Predictions
  • [Advertising, Calibration, Alibaba, WWW2022] MBCT- Tree-Based Feature-Aware Binning for Individual Uncertainty Calibration
  • [Advertisinig, Bidding, Alibaba, KDD2017] [PaperNote] Optimized Cost per Click in Taobao Display Advertising
  • [Advertising, Bidding, Alibaba, KDD2019] [PaperNote] Bid Optimization by Multivariable Control in Display Advertising
  • [Advertising, Bidding, Alibaba, PMLR2020] Dynamic Knapsack Optimization Towards Efficient Multi-Channel Sequential Advertising
  • [Advertising, Budget Control, Pacing, LinkedIn, KDD2014] [PaperNote] Budget Pacing for Targeted Online Advertisements at LinkedIn
  • [Advertising, Budget Control, Pacing, Yahoo, KDD2015] [PaperNote] Smart Pacing for Effective Online Ad Campaign Optimization

• M4

  • [Survival Analysis, Push, LinkedIn, WSDM2019] A State Transition Model for Mobile Notifications via Survival Analysis
  • [ETA, Deep Learning, Meituan, KDD2021] A Deep Learning Method for Route and Time Prediction in Food Delivery Service
  • [LTV, Tencent(PCG), KDD2021] Learning Reliable User Representations from Volatile and Sparse Data to Accurately Predict Customer Lifetime Value
  • [Churn Prediction, Game, User Growth, Survival Analysis,KDD2021] A Difficulty-Aware Framework for Churn Prediction and Intervention in Games
  • [Non-Personalized Recommendation, Google, KDD2021] Bootstrapping Recommendations at Chrome Web Store
  • [LTV, Google] [PaperNote] A Deep Probabilistic Model for Customer Lifetime Value Prediction
  • [LTV, Embedding, ASOS, KDD2017] Customer Lifetime Value Prediction Using Embeddings

• M3

  • [Marketing, Causal Inference, Causal Forest, Kuaishou, WWW2022] LBCF: A Large-Scale Budget-Constrained Causal Forest Algorithm
  • [LTV, Observational Causal Inference, Pricing, Netfilx, WWW2022] Beyond Customer Lifetime Valuation- Measuring the Value of Acquisition and Retention for Subscription Services

• M2

  • [Recommendation System, Multi-task Learning, PLE, Tencent, RecSys2020] Progressive Layered Extraction (PLE) - A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations
  • [Recommendation System, Reinforcement Learning, Off-Policy, REINFORCE, Google, WSDM2019] Top-K Off-Policy Correction for a REINFORCE Recommender System
  • [Causal Inference, Causal Forest, Didichuxing] GCF: Generalized Causal Forest for Heterogeneous Treatment Effect Estimation Using Nonparametric Methods
  • [Uplift, Causal Tree, LinkedIn] Generalized Causal Tree for Uplift Modeling

• M1

  • [Causal Inference, Long Term Effect, Representation Learning, WSDM2021] Long-Term Effect Estimation with Surrogate Representation
  • [Recommendation Sysytem, Causal Inference, User Retention, IJCAI2021] User Retention- A Causal Approach with Triple Task Modeling
  • [Recommendation System, Causal Inference, Doubly Robust, WWW2020] Large-scale Causal Approaches to Debiasing Post-click Conversion Rate Estimation with Multi-task Learning
  • [Recommendation System, Reinforcement Learning, Survey2021] A Survey of Deep Reinforcement Learning in Recommender Systems- A Systematic Review and Future Directions
  • [Recommendation System, Reinforcement Learning, Tencent] Value Penalized Q-Learning for Recommender Systems
  • [Recommendation System, Reinforcement Learning, Google, WWW2020] Off-policy Learning in Two-stage Recommender Systems
  • [Recommendation System, Reinforcement Learning, Google, SIGIR2020] Self-Supervised Reinforcement Learning for Recommender Systems

---

2021

• M12

  • [Recommendation System, Reinforcement Learning, Google, WSDM 2021] User Response Models to Improve a REINFORCE Recommender System

• M11

  • [Marketing, Uplift, Adversarial Learning, Alibaba, CIKM2021] Adversarial Learning for Incentive Optimization in Mobile Payment Marketing（这篇文章简直辣眼睛）
  • [Recommendation System, Multi-task Learning, MMoE, Google, KDD2018] Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts
  • [Recommendation System, Multi-task Learning, Meituan] Modeling the Sequential Dependence among Audience Multi-step Conversions with Multi-task Learning in Targeted Display Advertising
  • [Recommendation System, Multi-task Learning, Google, RecSys2019] Recommending What Video to Watch Next- A Multitask Ranking System
  • [Causal Inference, ITE, Optimization, LinkedIn, WWW2021] [ PaperNote ] Personalized Treatment Selection using Causal Heterogeneity

• M10

  • [Advertising, Causal Inference, Huawei, KDD2021] Estimating True Post-Click Conversion via Group-stratified Counterfactual Inference

• M9

  • [Recommendation System, Reinforcement Learning, User Engagement, Alibaba, KDD2020] Maximizing Cumulative User Engagement in Sequential Recommendation- An Online Optimization Perspective
  • [Recommendation System, Ranking, Causal Inference] Top-N Recommendation with Counterfactual User Preference Simulation
  • [Dyanmic Pricing, Reinforcement Learning, Thesis] Dynamic Pricing using Reinforcement Learning for the Amazon marketplace
  • [Causal Inference, Individual Treatment Effect, Criteo, KDD2021] Individual Treatment Prescription Effect Estimation in a Low Compliance Setting
  • [Marketing, Promotion, Uplift, Optimization, Booking] E-Commerce Promotions Personalization via Online Multiple-Choice Knapsack with Uplift Modeling
  • [Marketing, Promotion, Uplift, Counterfactual Prediction, Alibaba] A Framework for Massive Scale Personalized Promotion
  • [Recommendation System, Cold Start, Deep Learning, KDD2021] A Semi-Personalized System for User Cold Start Recommendation on Music Streaming Apps

• M8

  • [Marketing, Machine Learning, Optimization] [ PaperNote ] Large-Scale Data-Driven Airline Market Influence Maximization
  • [Reinforcement Learning, Dueling DQN] Dueling Network Architectures for Deep Reinforcement Learning
  • [Reinforcement Learning, Double DQN] Deep Reinforcement Learning with Double Q-learning
  • [Reinforcement Learning, DQN] Human-level control through deep reinforcement learning

• M7

  • [Recommendation System, Popular Bias, Causal Inference, SIGIR2021] Causal Intervention for Leveraging Popularity Bias in Recommendation
  • [Causal Inference, Recommendation System, KDD2021] Deconfounded Recommendation for Alleviating Bias Amplification
  • [Causal Inference, ITE, Advertising, Criteo, KDD2021] [ PaperNote ] Causal Models for Real Time Bidding with Repeated User Interactions

• M6

  • [Causal Inference, ITE, Advertising, Alibaba] Estimating Individual Advertising Effect in E-Commerce
  • [Dynamic Pricing, Optimization, Causal Inference, Alibaba, KDD2021] [ PaperNote ] Markdowns in E-Commerce Fresh Retail- A Counterfactual Prediction and Multi-Period Optimization Approach

---

2017-2018

• 很久之前的 Readling List