NLP Papers

A list of notes about NLP papers we've read at NLP Town (www.nlp.town). Our selection is heavily influenced by our specific interests and the projects we're involved in.

Text Generation

• Learning to generate one-sentence biographies from Wikidata
• Neural Text Generation from Structured Data with Application to the Biography Domain.
• Generating Wikipedia by Summarizing Long Sentences
• Learning to Ask: Neural Question Generation for Reading Comprehension

Named Entity Recognition

• Polyglot-NER Massive Multilingual Named Entity Recognition

Sequence-to-Sequence Learning

• Pointer Networks
• Incorporating Copying Mechanism in Sequence-to-Sequence Learning
• Seq2SQL: Generating Structured Queries from Natural Language Using Reinforcement Learning
• Morphological Inflection Generation with Hard Monotonic Attention
• Delete, Retrieve, Generate: A Simple Approach to Sentiment and Style Transfer

Sequence modelling

• An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelling
• The Importance of Being Recurrent for Modeling Hierarchical Structure
• Comparative Study of CNN and RNN for Natural Language Processing

Machine Translation

• An Exploration of Neural Sequence-to-Sequence Architectures for Automatic Post-Editing
• Achieving Human Parity on Automatic Chinese to English News Translation

Negation Detection

• A Simple Algorithm for Identifying Negated Findings and Diseases in Discharge Summaries
• Neural Networks For Negation Scope Detection

Word Embeddings

• Diachronic Word Embeddings Reveal Statistical Laws of Semantic Change
• Dynamic Word Embeddings for Evolving Semantic Discovery
• Learned in Translation: Contextualized Word Vectors
• Deep contextualized word representations

Multilingual word embeddings

• Offline Bilingual Word Vectors, Orthogonal Transformations and the Inverted Softmax
• Word Translation without Parallel Data

Sentence Embeddings

• A Simple but Tough-to-Beat Baseline for Sentence Embeddings
• Supervised Learning of Universal Sentence Representations from Natural Language Inference Data
• Universal Sentence Encoder

Cross-modal models

• Do Neural Network Cross-Modal Mappings Really Bridge Modalities?

Transfer Learning

• Universal Language Model Fine-tuning for Text Classification

Multitask Learning

• The Natural Language Decathlon: Multitask Learning as Question Answering

Quality

• Semantically Equivalent Adversarial Rules for Debugging NLP models