SNU CS & AI 4190.678: Natural Language Processing

reference model : MemSum

Scrapping ToS;DR dataset from the ToS;DR website.

• ToS;DR website : https://tosdr.org/

Improve the LSE module of MemSum by using the word embedding specialized for legal documents.

• download word embedding : https://osf.io/qvg8s/wiki/home/

Improve the LSE module of MemSum by using SBERT sentence embedding.

• SBERT model : "sentence-transformers/all-mpnet-base-v2"