SimCSE and RankCSE

SimCSE is a foundational contrastive learning approach for sentence embeddings that RankCSE builds upon and extends by incorporating ranking-based objectives to improve representation quality.

SimCSE
58
Established
RankCSE
37
Emerging
Maintenance 0/25
Adoption 11/25
Maturity 25/25
Community 22/25
Maintenance 0/25
Adoption 8/25
Maturity 16/25
Community 13/25
Stars: 3,644
Forks: 534
Downloads:
Commits (30d): 0
Language: Python
License: MIT
Stars: 48
Forks: 7
Downloads:
Commits (30d): 0
Language: Python
License: MIT
Stale 6m
Stale 6m No Package No Dependents

About SimCSE

princeton-nlp/SimCSE

[EMNLP 2021] SimCSE: Simple Contrastive Learning of Sentence Embeddings https://arxiv.org/abs/2104.08821

This tool helps you understand how semantically similar different pieces of text are, even if they use different words. You input sentences or short phrases, and it outputs numerical representations (embeddings) and similarity scores. This is useful for anyone who needs to automatically group, retrieve, or compare text, such as researchers analyzing surveys or businesses categorizing customer feedback.

text-analysis information-retrieval customer-feedback content-categorization research-analysis

About RankCSE

perceptiveshawty/RankCSE

Implementation of "RankCSE: Unsupervised Sentence Representation Learning via Learning to Rank" (ACL 2023)

This project helps researchers in natural language processing (NLP) to create better text embeddings. It takes large text datasets, like Wikipedia articles, and processes them to generate numerical representations of sentences. These representations can then be used in various downstream NLP tasks to measure how similar different sentences are. This is primarily for NLP researchers and machine learning engineers looking to advance sentence understanding.

natural-language-processing text-embeddings semantic-similarity unsupervised-learning language-modeling

Scores updated daily from GitHub, PyPI, and npm data. How scores work