EXAONE 3.5: Series of Large Language Models for Real-world Use Cases
Paper • 2412.04862 • Published • 52
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
EXAONE Korean Response Scoring SFT
EXAONE-3.5-7.8B-Instruct model fine-tuned for Korean learner response scoring.
Evaluation
Evaluation was performed on a held-out Korean response scoring test set using quadratic weighted kappa (QWK).
| Metric | QWK |
|---|---|
| Language | 0.8252 |
| Content | 0.8283 |
| Mean | 0.8267 |
Best validation checkpoint:
| Metric | Value |
|---|---|
| Validation mean QWK | 0.8400 |
| Step | 560 |
License
This model is derived from EXAONE-3.5-7.8B-Instruct and is subject to the EXAONE AI Model License Agreement 1.1 - NC.
Citation
@article{exaone-3.5,
title={EXAONE 3.5: Series of Large Language Models for Real-world Use Cases},
author={LG AI Research},
journal={arXiv preprint arXiv:https://arxiv.org/abs/2412.04862},
year={2024}
}
- Downloads last month
- 17
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support