# Unsupervised Contrastive Learning

This is an intermediate version of a multilingual Mixture of Experts (MoE) text embedding model, obtained through multi-stage contrastive training

LLM2Vec is a solution for converting decoder-only large language models into text encoders, achieved by enabling bidirectional attention, masked next-word prediction, and unsupervised contrastive learning.

LLM2Vec is a simple method to convert decoder-only large language models into text encoders by enabling bidirectional attention, masked next-token prediction, and unsupervised contrastive learning.

LLM2Vec is a simple solution for converting decoder-only large language models into text encoders, achieved by enabling bidirectional attention mechanisms, masked next-token prediction, and unsupervised contrastive learning.

LLM2Vec is a simple solution for transforming decoder-only large language models into text encoders, achieved by enabling bidirectional attention, masked next-token prediction, and unsupervised contrastive learning.

A Czech-English semantic embedding model fine-tuned with SimCSE objective, based on Seznam/dist-mpnet-paracrawl-cs-en

A Czech sentence similarity model fine-tuned with SimCSE objective based on Seznam/small-e-czech model

This dataset is used for training and evaluating the SimCSE (Simple Contrastive Learning of Sentence Embeddings) model, supporting sentence similarity tasks.

A Chinese sentence vector representation model based on the unsupervised version of SimCSE, trained with supervised contrastive learning using Chinese NLI data

SciBERT-based scientific text sentence encoder, trained on 2 million scientific papers through self-supervised learning