NLP & 推荐算法 论文+博客整理
一、经典论文
nlp:
推荐算法:
[Youtube-DNN] Deep Neural Networks for YouTube Recommendations(Google 2016,非常经典的论文)
[Pinterest] Graph Convolutional Neural Networks for Web-Scale Recommender Systems (Pinterest 2018)
[DL Recsys Intro] Deep Learning based Recommender System- A Survey and New Perspectives (UNSW 2018)
召回:
[DSSM双塔模型] Learning Deep Structured Semantic Models for Web Search using Clickthrough Data (UIUC 2013)
[TDM] Learning Tree-based Deep Model for Recommender Systems(Alibaba 2018)
排序:
[ESMM] Entire Space Multi-Task Model - An Effective Approach for Estimating Post-Click Conversion Rate (Alibaba 2018,多任务)
[MMOE] Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts(Google 2018,众多大厂都有用这个模型,多任务)
[DIEN] Deep Interest Evolution Network for Click-Through Rate Prediction (Alibaba 2019)
[DIN] Deep Interest Network for Click-Through Rate Prediction (Alibaba 2018)
其它:
[tutorial] Learning to Rank for Information Retrieval(Microsoft 2010,刘铁岩经典综述)
[DeepWalk] DeepWalk: Online Learning of Social Representations(2014)
[item2vec] Item2Vec-Neural Item Embedding for Collaborative Filtering (Microsoft 2016)
[node2vec] node2vec: Scalable Feature Learning for Networks(2016)
二、近年新论文
nlp:
描述 | 论文 |
小样本 |
Multi-Label Few-Shot Learning for Aspect Category Detection https://arxiv.org/abs/2105.1417 Few-Shot Text Ranking with Meta Adapted Synthetic Weak Supervision https://arxiv.org/abs/2012.1486 Generalizing from a Few Examples: A Survey on Few-Shot Learning(小样本学习综述) |
NER |
Named Entity Recognition with Small Strongly Labeled and Large Weakly Labeled Data https://arxiv.org/abs/2106.0897 Document-level Event Extraction via Heterogeneous Graph-based Interaction Model with a Tracker |
对话 |
Answering Ambiguous Questions through Generative Evidence Fusion and Round-Trip Prediction |
生成 |
Prefix-Tuning: Optimizing Continuous Prompts for Generation |
摘要 |
Cross-Lingual Abstractive Summarization with Limited Parallel Resources https://arxiv.org/abs/2105.1364 Long-Span Summarization via Local Attention and Content Selection |
预训练模型 |
NEZHA: Neural Contextualized Representation for Chinese Language Understanding ERNIE-Gram: Pre-Training with Explicitly N-Gram Masked Language Modeling for Natural Language ERNIE 2.0: A Continual Pre-training Framework for Language Understanding |
表征学习 |
DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representations https://arxiv.org/abs/2006.03659(对比学习) ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfer https://arxiv.org/abs/2105.1174 Self-Guided Contrastive Learning for BERT Sentence Representations |
知识图谱 |
Dynamic Knowledge Graph Construction for Zero-shot Commonsense Question Answering
Case-based Reasoning for Natural Language Queries over Knowledge Bases https://arxiv.org/pdf/2104.08762.pdf Dynamic Knowledge Graph Construction for Zero-shot Commonsense Question Answering |
推荐算法:
[Graph learning] Graph Learning Approaches to Recommender Systems: A Review(2021)
召回:
[JTM] Joint Optimization of Tree-based Index and Deep Model for Recommender Systems(Alibaba 2019)
[Deep Retrieval] Deep Retrieval: Learning A Retrievable Structure for Large-Scale Recommendations(字节 2021)
排序:
[PLE] Progressive Layered Extraction (PLE): A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations(腾讯 2020,近一年很多大厂有follow这项工作,多任务)
其它:
[KAFtt] Kalman Filtering Attention for User Behavior Modeling in CTR Prediction(京东 2020)
[SIM] Search-based User Interest Modeling with Lifelong Sequential Behavior Data for Click-Through Rate Prediction (Alibaba 2020)
[BST] Behavior Sequence Transformer for E-commerce Recommendation in Alibaba(Alibaba 2019)
[GIN] Graph Intention Network for Click-through Rate Prediction in Sponsored Search(Alibaba 2019)
三、知识点重难点解读
nlp:
Prompt Based Task Reformulation in NLP调研 | Thinkwee's Blog
https://github.com/km1994/nlp_paper_study