10 3 21

Youngjoon Jang

yjoonjang

yjoonjang

AI & ML interests

None yet

Recent Activity

reacted to tomaarsen's post with 🔥 6 days ago

I just released Sentence Transformers v4.1; featuring ONNX and OpenVINO backends for rerankers offering 2-3x speedups and improved hard negatives mining which helps prepare stronger training datasets. Details: 🏎️ ONNX, OpenVINO, Optimization, Quantization - I've added ONNX and OpenVINO support with just one extra argument: "backend" when loading the CrossEncoder reranker, e.g.: `CrossEncoder("cross-encoder/ms-marco-MiniLM-L6-v2", backend="onnx")` - The `export_optimized_onnx_model`, `export_dynamic_quantized_onnx_model`, and `export_static_quantized_openvino_model` functions now work with CrossEncoder rerankers, allowing you to optimize (e.g. fusions, gelu approximations, etc.) or quantize (int8 weights) rerankers. - I've uploaded ~340 ONNX & OpenVINO models for all existing models under the cross-encoder Hugging Face organization. You can use these without having to export when loading. ⛏ Improved Hard Negatives Mining - Added 'absolute_margin' and 'relative_margin' arguments to `mine_hard_negatives`. - `absolute_margin` ensures that `sim(query, negative) < sim(query, positive) - absolute_margin`, i.e. an absolute margin between the negative & positive similarities. - `relative_margin` ensures that `sim(query, negative) < sim(query, positive) * (1 - relative_margin)`, i.e. a relative margin between the negative & positive similarities. - Inspired by the excellent NV-Retriever paper from NVIDIA. And several other small improvements. Check out the full release notes here: https://github.com/UKPLab/sentence-transformers/releases/tag/v4.1.0 With this release, I introduce near-feature parity between the SentenceTransformer embedding & CrossEncoder reranker models, which I've wanted to do for quite some time! With rerankers very strongly supported now, it's time to look forward to other useful architectures!

upvoted a paper 6 days ago

MIEB: Massive Image Embedding Benchmark

published a model 12 days ago

yjoonjang/preranker-v1

View all activity

Organizations

yjoonjang's activity

liked a dataset about 1 month ago

Salesforce/xlam-function-calling-60k

Viewer • Updated Jan 24 • 60k • 3.09k • 439

liked a model about 1 month ago

trillionlabs/Trillion-7B-preview

Text Generation • Updated 28 days ago • 7.75k • 77

liked a dataset about 1 month ago

mangopy/ToolRet-Tools

Viewer • Updated Mar 8 • 44.5k • 225 • 3

liked 3 models about 1 month ago

LGAI-EXAONE/EXAONE-Deep-32B

Text Generation • Updated Mar 19 • 41.8k • 288

mixedbread-ai/mxbai-rerank-large-v2

Text Ranking • Updated 19 days ago • 141k • 79

dragonkue/snowflake-arctic-embed-l-v2.0-ko

liked 4 models about 2 months ago

facebook/drama-base

facebook/drama-large

facebook/drama-1b

kakaocorp/kanana-nano-2.1b-embedding

Text Generation • Updated Feb 27 • 11.5k • 20

liked 2 models 2 months ago

perplexity-ai/r1-1776

Text Generation • Updated Feb 26 • 8.67k • • 2.23k

yibinlei/LENS-d8000

Feature Extraction • Updated Jan 22 • 868 • 6

liked a dataset 3 months ago

sentence-transformers/embedding-training-data

Updated Sep 11, 2024 • 2.12k • 123

liked a model 4 months ago

nlpai-lab/KURE-v1

Feature Extraction • Updated Dec 23, 2024 • 30.5k • 35

liked a dataset 5 months ago

channelcorp/KoMagpie-raw

Viewer • Updated Sep 23, 2024 • 2.57M • 156 • 30

liked a dataset 6 months ago

Shitao/MLDR

Updated Feb 6, 2024 • 1.59k • 65

liked 2 datasets 7 months ago

nlpai-lab/ko-triplet-v1.0

Viewer • Updated Nov 29, 2024 • 745k • 92 • 22

google/frames-benchmark

Viewer • Updated Oct 15, 2024 • 824 • 2.05k • 194

liked a model 7 months ago

nlpai-lab/KoE5

Feature Extraction • Updated Dec 23, 2024 • 7.14k • 41

liked a dataset 8 months ago

nayohan/Magpie-Pro-MT-300K-v0.1-ko

Viewer • Updated Jul 13, 2024 • 300k • 37 • 25