Downloads · 30 days
4
11% of all-time downloads
time-embed-bge-m3-lms-temporal-v1-7
kev-KOH/time-embed-bge-m3-lms-temporal-v1-7
time-embed-bge-m3-lms-temporal-v1-7 is a machine learning model from kev-KOH. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for FlagEmbedding.
BGE-M3 embedding model fine-tuned for Korean LMS temporal retrieval.
All-time downloads
37
Public
Parameters
568M
2.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors2.3 GB · 99%
At a glance
- Library
- FlagEmbedding
- Model type
- xlm-roberta
- Access
- Public
- Created
- May 30, 2026
- Updated
- May 30, 2026
- SHA
- b005fb0f
Base models
Model card
From the Hugging Face model README
time-embed-bge-m3-lms-temporal-v1-7
BGE-M3 embedding model fine-tuned for Korean LMS temporal retrieval.
Purpose
This model focuses on Korean temporal equivalence and hard-negative separation for LMS item queries, including cases such as:
일주일 전=7일 전=1주 전지난주!=7일 전최근 7일!=7일 전- calendar week/month/year expressions vs fixed duration offsets
Training Run
- Run directory:
~/Time-Embed/runs/time-embed-bge-m3-v1_7/20260530-025708 - Base model:
kev-KOH/time-embed-bge-m3-lms-temporal-v1-6-1 - Dataset family:
kev-KOH/time-embed-bge-m3, v1.7 LMS temporal/calendar data - Training objective: FlagEmbedding BGE-M3 contrastive fine-tuning
Intended Use
Use with FlagEmbedding.BGEM3FlagModel for Korean LMS search/retrieval query embeddings.
- Library
- FlagEmbedding
- Type
- xlm-roberta
- Languages
- ko
- Created
- May 30, 2026
- Updated
- May 30, 2026