# Contrastive learning optimization

InstructCLIP is a model that automatically optimizes data through contrastive learning to enhance instruction-guided image editing.

Instruct-CLIP is a contrastive learning-based model designed to optimize instruction-guided image editing tasks.

InstructCLIP is an instruction-guided image editing model improved through contrastive learning-based automatic data optimization. It combines CLIP and Stable Diffusion technologies to edit images based on textual instructions.

General-purpose Japanese text embedding model, optimized for retrieval tasks with excellent performance on CPUs

MMLW is a neural text encoder for Polish, optimized for information retrieval tasks, capable of converting queries and passages into 1024-dimensional vectors

MMLW (I Must Get Better Messages) is a neural text encoder for Polish, optimized for information retrieval tasks, capable of converting queries and passages into 384-dimensional vectors.

CompCSE and SimCSE are contrastive learning-based sentence embedding models for calculating sentence similarity.

COCO-DR is a dense retrieval model based on BERT-base, which uses contrastive learning and distributionally robust learning methods to address the distribution shift problem in zero-shot dense retrieval.

A sentence embedding model fine-tuned on over 1 billion sentence pairs through self-supervised contrastive learning based on MiniLM-L12, capable of generating high-quality semantic vector representations

A sentence embedding model based on MiniLM architecture, trained on over 1 billion sentence pairs through self-supervised contrastive learning, capable of generating high-quality sentence vector representations

A RoBERTa-large based sentence embedding model trained on 1 billion sentence pairs through self-supervised contrastive learning for generating semantically rich sentence vector representations

A sentence embedding model based on MiniLM-L12 architecture, trained on over 1 billion sentence pairs through contrastive learning, capable of generating high-quality semantic vector representations

A sentence embedding model based on mpnet-base, trained on 1 billion sentence pairs through self-supervised contrastive learning, capable of generating high-quality semantic vector representations of sentences

A lightweight sentence embedding model based on MiniLM architecture, fine-tuned with contrastive learning on a 1-billion sentence pair dataset, suitable for semantic similarity calculation and information retrieval tasks