Skip to content

This project investigates the use of our dependency parser/BERT/coreference resolver pipeline with a Transformer classifier to understand the MultiRC dataset (Multi Sentence Reading Comprehension).

Notifications You must be signed in to change notification settings

MisterPNP/MultiRCTransformer

Repository files navigation

MultiRCTransformer

Final project of the COMP590: NLP class at UNC Spring 2022.

This project investigates the use of our dependency parser/BERT/coreference resolver pipeline with a Transformer classifier to understand the MultiRC dataset (Multi Sentence Reading Comprehension).

The pipeline involves:

  • a BERT encoder for input descriptions and answers.
  • A SpanBERT Coreference Resolver involving the description, question, and answer.
  • A dependency parser to create dependency relations from the description and questions.
  • A relation pooler to extract relations from the three inputs
  • A final structured correctness classifier to classify each answer to the question conditionally to the background question description.

About

This project investigates the use of our dependency parser/BERT/coreference resolver pipeline with a Transformer classifier to understand the MultiRC dataset (Multi Sentence Reading Comprehension).

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages