KoE5

About

What KoE5 is

🔎 KoE5 Introducing KoE5, a model with advanced retrieval abilities. It has shown remarkable performance in Korean text retrieval. For details, visit the KURE repository Model Versions Model Description This is the model card of a 🤗 transformers model that has been pushed on the Hub. - Developed by: NLP&AI Lab - Language(s) (NLP): Korean, English - License: MIT - Finetuned from model: intfloat/multilingual-e5-large - Finetuned dataset: ko-triplet-v1.0 Example code Install Dependencies First install the Sentence Transformers library: Python code Then you can load this model and run inference. Training Details Training Data - ko-triplet-v1.0 - Korean query-document-hardnegative data pair (open data) - About 700000+ examples used totally Training Procedure - loss: Used CachedMultipleNegativesRankingLoss by sentence-transformers - batch size: 512 - learning rate: 1e-05 - epochs: 1 Evaluation Metrics - Recall, Precision, NDCG, F1 Benchmark Datasets - Ko-StrategyQA: 한국어 ODQA multi-hop 검색 데이터셋 (StrategyQA 번역) - AutoRAGRetrieval: 금융, 공공, 의료, 법률, 커머스 5개 분야에 대해, pdf를 파싱하여 구성한 한국어 문서 검색 데이터셋 - MIRACLRetrieval): Wikipedia 기반의 한국어 문서 검색 데이터셋 - PublicHealthQA): 의료 및 공중보건 도메인에 대한 한국어 문서 검색 데이터셋 - BelebeleRetrieval): FLORES-200 기반의 한국어 문서 검색 데이터셋 - MrTidyRetrieval: Wikipedia 기반의 한국어 문서 검색 데이터셋 - MultiLongDocRetrieval: 다양한 도메인의 한국어 장문 검색 데이터셋 - XPQARetrieval: 다양한 도메인의 한국어 문서 검색 데이터셋 Results 아래는 모든 모델의, 모든 벤치마크 데이터셋에 대한 평균 결과입니다. 자세한 결과는 KURE Github에서 확인하실 수 있습니다. Top-k 1 Top-k 3 Top-k 5 Top-k 10 FAQ - Do I need to add the prefix "query: " and "passage: " to input texts? Yes, this is how the model is trained, otherwise you will see a performance degradation. Here are some rules of thumb: - Use "query: " and "passage: " correspondingly for asymmetric tasks such as passage retrieval in open QA, ad-hoc information retrieval. - Use "query: " prefix for symmetric tasks such as semantic similarity, bitext mining, paraphrase retrieval. - Use "query: " prefix if you want to use embeddings as features, such as linear probing classification, clustering. Citation If you find our paper or models helpful, please consider cite as follows: Limitations Long texts will be truncated to at most 512

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

How it works

How embedding models work

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys koe5 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (koe5 below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/embeddings \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"koe5","input":"text to embed"}'

Details

Languages, data & research

Licence

Open weights

Sources

Weights & code

Explore

More embedding models