KoSimCSE-roberta is a RoBERTa-based model trained with SimCSE for Korean sentence representation learning. It produces high-quality embeddings for semantic textual similarity, retrieval, and clustering tasks in Korean. The model is compatible with Hugging Face Transformers and text-embeddings-inference.
KoSimCSE Roberta sits in PulseGate's Embeddings & retrieval category. It focuses on generating high-quality Korean sentence embeddings for semantic search and similarity tasks. It is built as an open-source project for NLP researchers and Korean language developers. KoSimCSE Roberta is open source under the Open Source license. It ships for the web.
It is developed by BM-K (South Korea), and it first shipped in 2022. Among its 3 catalogued features are Sentence Embeddings, Semantic Similarity, and Korean NLP. It exposes integrations via a public API.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do