转自: 圆圆的算法笔记

链接: mp.weixin.qq.com/s/aWdPk_zKwJX7gnsahQNcSw

本文汇总了2024年KDD、WWW、WSDM等顶会中，阿里、腾讯、亚马逊等头部企业发表的推荐系统代表性工作，共 49篇 论文，涉及大模型、冷启动、消偏、多场景、多任务、多模态、兴趣建模、模型结构优化、CVR预估等 11个 推荐系统主要优化方向。

相关领域的详细工作解析和整理，后续也会逐渐在公众号中更新，感兴趣的同学欢迎持续关注。

RecSys——华为——大模型：FLIP: Fine-grained Alignment between ID-based Models and Pretrained Language Models for CTR Prediction

RecSys——美团——大模型：LARR: Large Language Model Aided Real-time Scene Recommendation with Semantic Understanding

RecSys——腾讯——大模型：Scaling Law of Large Sequential Recommendation Models

RecSys——腾讯——大模型：The Elephant in the Room: Rethinking the Usage of Pre-trained Language Model in Sequential Recommendation

RecSys——蚂蚁——大模型：ReLand: Integrating Large Language Models’ Insights into Industrial Recommenders via a Controllable Reasoning Pool

WWW——领英——大模型：Collaborative Large Language Model for Recommender Systems

WWW——华为——大模型：ClickPrompt: CTR Models are Strong Prompt Generators for Adapting Language Models to CTR Prediction

WWW——百度——大模型：Representation Learning with Large Language Models for Recommendation

KDD——微软——大模型：RecExplainer: Aligning Large Language Models for Explaining Recommendation Models

ICDE——腾讯——大模型：Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation

SIGIR——蚂蚁——大模型：Enhancing Sequential Recommenders with Augmented Knowledge from Aligned Large Language Models

NeurIPS——华为——大模型：LLM-ESR: Large Language Models Enhancement for Long-tailed Sequential Recommendation

AAAI——蚂蚁——大模型：Fine-Tuning Large Language Model Based Explainable Recommendation with Explainable Quality Reward

RecSys——快手——多模态：A Multi-modal Modeling Framework for Cold-start Short-video Recommendation

KDD——腾讯——多模态：Improving Multi-modal Recommender Systems by Denoising and Aligning Multi-modal Content and User Feedback

RecSys——亚马逊——冷启动：MARec: Metadata Alignment for cold-start Recommendation

RecSys——快手——冷启动：Prompt Tuning for Item Cold-start Recommendation

KDD——百度——冷启动：Warming Up Cold-Start CTR Prediction by Learning Item-Specific Feature Interactions

ICDE——腾讯——冷启动：Firzen: Firing Strict Cold-Start Items with Frozen Heterogeneous and Homogeneous Graphs for Recommendation

RecSys——阿里——多场景：MLoRA: Multi-Domain Low-Rank Adaptive Network for CTR Prediction

RecSys——美团——多场景：Scene-wise Adaptive Network for Dynamic Cold-start Scenes Optimization in CTR Prediction

WWW——蚂蚁——多场景：Rethinking Cross-Domain Sequential Recommendation under Open-World Assumptions

WWW——阿里——多场景：Not All Embeddings are Created Equal: Towards Robust Cross-domain Recommendation via Contrastive Learning

SIGIR——腾讯——多场景：Aiming at the Target: Filter Collaborative Information for Cross-Domain Recommendation

SIGIR——阿里——多场景：Scenario-Adaptive Fine-Grained Personalization Network: Tailoring User Behavior Representation to the Scenario Context

RecSys——腾讯——多任务：Touch the Core: Exploring Task Dependence Among Hybrid Targets for Recommendation

AAAI——腾讯——多任务：STEM: Unleashing the Power of Embeddings for Multi-task Recommendation

SIGIR——快手——多任务：MDMTRec: An Adaptive Multi-Task Multi-Domain Recommendation Framework

RecSys——快手——表示学习：MMGCL: Meta Knowledge-Enhanced Multi-view Graph Contrastive Learning for Recommendations

WWW——腾讯——表示学习：GraphPro: Graph Pre-training and Prompt Learning for Recommendation

KDD——亚马逊——表示学习：Shopping Trajectory Representation Learning with Pre-training for E-commerce Customer Understanding and Recommendation

RecSys——华为——CVR预估：Ranking-Aware Unbiased Post-Click Conversion Rate Estimation via AUC Optimization on Entire Exposure Space

RecSys——腾讯——CVR预估：Utilizing Non-click Samples via Semi-supervised Learning for Conversion Rate Prediction

WWW——腾讯——消偏：Causally Debiased Time-aware Recommendation

KDD——消偏——Counteracting Duration Bias in Video Recommendation via Counterfactual Watch Time

RecSys——华为——消偏：AIE: Auction Information Enhanced Framework for CTR Prediction in Online Advertising

RecSys——雅虎——特征优化：Low Rank Field-Weighted Factorization Machines for Low Latency Item Recommendation

AAAI——快手——时长预估：CREAD: A Classification-Restoration Framework with Error Adaptive Discretization for Watch Time Prediction in Video Recommender Systems

SIGIR——腾讯——CTR预估：Deep Pattern Network for Click-Through Rate Prediction

AAAI——京东——优化算法：Generalize for Future: Slow and Fast Trajectory Learning for CTR Prediction

WSDM——亚马逊——优化算法：To Copy, or not to Copy; That is a Critical Issue of the Output Softmax Layer in Neural Sequential Recommenders

WSDM——华为——兴趣建模：User Behavior Enriched Temporal Knowledge Graph for Sequential Recommendation

WSDM——快手——兴趣建模：Collaboration and Transition: Distilling Item Transitions into Multi-Query Self-Attention for Sequential Recommendation

SIGIR——快手——疲劳度建模：Modeling User Fatigue for Sequential Recommendation

上述工作的详细解析，也会逐渐更新到公众号中，感兴趣的同学可以持续关注。

欢迎 干货投稿 \ 论文宣传 \ 合作交流

推荐阅读