NLP-Progress records the latest NLP data sets, papers and codes: to help you keep up with the frontiers of NLP

NLP-Progress records the latest NLP data sets, papers and codes: to help you keep up with the frontiers of NLP

Blessed are students of natural language processing. In order to track the progress of natural language processing (NLP), a large number of people with lofty ideals maintain a library called NLP-Progress on Github. It records the baseline and standard data sets of almost all NLP tasks, as well as the state-of-the-art of these issues.

  Official website
  Organizing reports

NLP-Progress also covers traditional NLP tasks, such as dependency parsing and part-of-speech tagging, and some new tasks, such as reading comprehension and natural language reasoning. It not only provides readers with baselines and standard data sets for these tasks, but also records the state-of-the-art of these issues.

The following editor briefly lists several tasks recorded by NLP-Progress:

  Coreference resolution co-referential resolution
  Dependency parsing
  Dialogue dialogue
  Domain Adaption domain migration
  Entity Linking
  Information extraction
  Language modeling language model
  Machine translation Machine translation
  Multi-task learning Multi-task learning
  Multi- modal Multimodal
  Named entity recognition
  Natural language inference
  Part-of-speech tagging Part-of-speech tagging
  Question answering
  Relation prediction
  Relationship extraction
  Semantic textual similarity Semantic textual similarity
  Semantic parsing
  Semantic role labeling
  Sentiment analysis
  Summarization Summarization
  Taxonomy learning Taxonomy learning
  Temporal processing
  Time series analysis Text classification
  Word sense disambiguation
  . . .
  . . .

For each task, NLP-Progress will briefly introduce what the task does, and list the public standard data set in detail, as well as the current ranking of each model on the data set. For example, the more popular Question answering question answering system task, its organization is as follows:

Specific to a certain open data set, such as Quasar, the contributor will briefly introduce the composition of the data set, and then list the paper rankings, each line of which includes: model, effect, article name and link, and code link.

What are you waiting for? Go to Star quickly.

The original publication time is: 2018-11-15

This article comes from the " Professional Knowledge " of a partner in the Yunqi Community . You can follow the " Professional Knowledge " for relevant information .