# Dual-encoder architecture

ReasonIR-8B is the first retrieval model specifically trained for general reasoning tasks, achieving state-of-the-art retrieval performance on the BRIGHT benchmark and significantly improving performance on MMLU and GPQA benchmarks in RAG applications.

Hypencoder is a dual-encoder model for information retrieval, consisting of a text encoder and a hypernetwork (Hypencoder), capable of converting text into small neural networks for computing relevance scores.

A dual-encoder retrieval model for the medical field, capable of processing Korean-English mixed medical records.

A BERT-based zero-shot text classification model trained on the UTCD dataset using explicit training methods

A zero-shot text classification model based on sentence-transformers, achieving text classification without labeled data through implicit training

BERT-based dual-encoder model, specifically designed for zero-shot text classification tasks, trained on the UTCD dataset

A model that maps abstract sentence descriptions to matching sentences, trained on Wikipedia using a dual-encoder architecture.

An efficient SPLADE model for paragraph retrieval, using a dual-model architecture (separate query model and document model)

Dual-encoder fine-tuned from German ELECTRA model for German passage retrieval tasks