ymgw55/WSMD
Improving word mover’s distance by leveraging self-attention matrix (Published in EMNLP 2023 Findings)
This project helps researchers and data scientists more accurately compare the similarity between two pieces of text, even when they use similar words in different contexts. It takes two sentences or short texts as input and provides a numerical score indicating how semantically alike they are, considering both word meaning and sentence structure. This is designed for natural language processing specialists working on text analysis and understanding tasks.
Use this if you need a more nuanced and accurate way to measure the semantic similarity between sentences, especially when word order and context are crucial.
Not ideal if you are looking for a simple, off-the-shelf text similarity tool without needing to dive into underlying model configurations and datasets.
Stars
10
Forks
—
Language
Python
License
—
Category
Last pushed
Mar 10, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/ymgw55/WSMD"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
alvations/pywsd
Python Implementations of Word Sense Disambiguation (WSD) Technologies.
SapienzaNLP/ewiser
A Word Sense Disambiguation system integrating implicit and explicit external knowledge.
danlou/LMMS
Language Modelling Makes Sense - WSD (and more) with Contextual Embeddings
dustalov/watset
Watset: Automatic Induction of Synsets from a Graph of Synonyms
USC-NSL/sage
SAGE disambiguates protocol description in an IETF RFC document, then converts the disambiguated...