传统模型
- TF-IDF: term frequency–inverse document frequency
- LDA: Latent Dirichlet Allocation, JMLR, 2003
词向量
- Word2vec: Efficient Estimation of Word Representations in Vector Space, ICLR, 2013
- GloVe: GloVe: Global Vectors for Word Representation, EMNLP, 2014
- FastText: Bag of Tricks for Efficient Text Classification, EACL, 2017
句向量
- STV: Skip-Thought Vector, NIPS, 2015 (word2vec相同思想)
- QTV (Quick-Thought Vectors): An efficient framework for learning sentence representations, LCLR, 2018 (改进STV为一个分类任务)
RNN
- RNN: Finding structure in time, Cognitive science, 1990
- LSTM: Long short-term memory, Neural computation, 1997
- GRU: Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling, NIPS, 2014
- SRU: Simple Recurrent Units for Highly Parallelizable Recurrence, ENNLP, 2018
- IndRNN: Independently Recurrent Neural Network (IndRNN): Building a Longer and Deeper RNN, CVPR, 2018
Deep Model
- Text CNN: Convolutional Neural Networks for Sentence Classification, ENNLP, 2014
- Char-CNN: Character-level Convolutional Networks for Text Classification, NIPS, 2015
- C-LSTM: A C-LSTM Neural Network for Text Classification, arXiv, 2015
- Text RNN: Recurrent Neural Network for Text Classification with Multi-Task Learning, IJCAI 2016
- Text CNN 2: A Sensitivity Analysis of (and Practitioners’ Guide to) Convolutional Neural Networks for Sentence Classification, IJCNLP, 2017
- Attention: Attention Is All You Need, NIPS, 2017