🚀 KF-DeBERTa
KF-DeBERTa是由 KakaoBank 和 FNGuide 联合训练的金融领域专用语言模型。该模型结合了通用领域和金融领域的语料进行训练,在多个下游任务中展现出卓越性能。
✨ 主要特性
- 多领域融合训练:KF-DeBERTa 同时使用通用领域语料库和金融领域语料库进行训练,在通用领域和金融领域的下游任务中均表现出色。
- 架构选择:模型架构基于 DeBERTa-v2 进行训练。由于观察到使用 ELECTRA 的 RTD 作为训练目标的 DeBERTa-v3 在部分任务(如 KLUE-RE、WoS、Retrieval)中性能较低,最终选择了 DeBERTa-v2 作为架构。
- 优异性能:在通用领域和金融领域的下游任务中均验证了其卓越性能。特别是在金融领域的下游任务中,通过多种数据集进行了全面的性能验证,在 KLUE Benchmark 中表现优于 RoBERTa-Large。
📦 安装指南
文档未提供安装步骤,如需使用该模型,可参考 Hugging Face 的 Transformers 库的安装方法。
💻 使用示例
基础用法
from transformers import AutoModel, AutoTokenizer
model = AutoModel.from_pretrained("kakaobank/kf-deberta-base")
tokenizer = AutoTokenizer.from_pretrained("kakaobank/kf-deberta-base")
text = "카카오뱅크와 에프엔가이드가 금융특화 언어모델을 공개합니다."
tokens = tokenizer.tokenize(text)
print(tokens)
inputs = tokenizer(text, return_tensors="pt")
model_output = model(**inputs)
print(model_output)
📚 详细文档
基准测试
所有任务仅进行了以下基本的超参数搜索:
- 批量大小(batch size): {16, 32}
- 学习率(learning_rate): {1e-5, 3e-5, 5e-5}
- 权重衰减(weight_decay): {0, 0.01}
- 预热比例(warmup_proportion): {0, 0.1}
KLUE 基准测试
模型 |
YNAT(F1) |
KLUE-ST(Pearsonr/F1) |
KLUE-NLI(ACC) |
KLUE-NER(F1-Entity/F1-Char) |
KLUE-RE(F1-micro/AUC) |
KLUE-DP(UAS/LAS) |
KLUE-MRC(EM/ROUGE) |
WoS(JGA/F1-S) |
平均 |
mBERT (Base) |
82.64 |
82.97/75.93 |
72.90 |
75.56/88.81 |
58.39/56.41 |
88.53/86.04 |
49.96/55.57 |
35.27/88.60 |
71.26 |
XLM-R (Base) |
84.52 |
88.88/81.20 |
78.23 |
80.48/92.14 |
57.62/57.05 |
93.12/87.23 |
26.76/53.36 |
41.54/89.81 |
72.28 |
XLM-R (Large) |
87.30 |
93.08/87.17 |
86.40 |
82.18/93.20 |
58.75/63.53 |
92.87/87.82 |
35.23/66.55 |
42.44/89.88 |
76.17 |
KR-BERT (Base) |
85.36 |
87.50/77.92 |
77.10 |
74.97/90.46 |
62.83/65.42 |
92.87/87.13 |
48.95/58.38 |
45.60/90.82 |
74.67 |
KoELECTRA (Base) |
85.99 |
93.14/85.89 |
86.87 |
86.06/92.75 |
62.67/57.46 |
90.93/87.07 |
59.54/65.64 |
39.83/88.91 |
77.34 |
KLUE-BERT (Base) |
86.95 |
91.01/83.44 |
79.87 |
83.71/91.17 |
65.58/68.11 |
93.07/87.25 |
62.42/68.15 |
46.72/91.59 |
78.50 |
KLUE-RoBERTa (Small) |
85.95 |
91.70/85.42 |
81.00 |
83.55/91.20 |
61.26/60.89 |
93.47/87.50 |
58.25/63.56 |
46.65/91.50 |
77.28 |
KLUE-RoBERTa (Base) |
86.19 |
92.91/86.78 |
86.30 |
83.81/91.09 |
66.73/68.11 |
93.75/87.77 |
69.56/74.64 |
47.41/91.60 |
80.48 |
KLUE-RoBERTa (Large) |
85.88 |
93.20/86.13 |
89.50 |
84.54/91.45 |
71.06/73.33 |
93.84/87.93 |
75.26/80.30 |
49.39/92.19 |
82.43 |
KF-DeBERTa (Base) |
87.51 |
93.24/87.73 |
88.37 |
89.17/93.30 |
69.70/75.07 |
94.05/87.97 |
72.59/78.08 |
50.21/92.59 |
82.83 |
注:粗体表示所有模型中的最高分数,下划线表示基础模型中的最高分数。
金融领域基准测试
模型 |
FN-Sentiment (v1)(ACC) |
FN-Sentiment (v2)(ACC) |
FN-Adnews(ACC) |
FN-NER(F1-micro) |
KorFPB(ACC) |
KorFiQA-SA(MSE) |
KorHeadline(Mean F1) |
平均(除 FiQA-SA) |
KLUE-RoBERTa (Base) |
98.26 |
91.21 |
96.34 |
90.31 |
90.97 |
0.0589 |
81.11 |
94.03 |
KoELECTRA (Base) |
98.26 |
90.56 |
96.98 |
89.81 |
92.36 |
0.0652 |
80.69 |
93.90 |
KF-DeBERTa (Base) |
99.36 |
92.29 |
97.63 |
91.80 |
93.47 |
0.0553 |
82.12 |
95.27 |
- FN-Sentiment:金融领域情感分析
- FN-Adnews:金融领域广告新闻分类
- FN-NER:金融领域命名实体识别
- KorFPB:FinancialPhraseBank 翻译数据集
- 引用:
Malo, Pekka, et al. "Good debt or bad debt: Detecting semantic orientations in economic texts." Journal of the Association for Information Science and Technology 65.4 (2014): 782-796.
- KorFiQA-SA:FiQA-SA 翻译数据集
- 引用:
Maia, Macedo & Handschuh, Siegfried & Freitas, Andre & Davis, Brian & McDermott, Ross & Zarrouk, Manel & Balahur, Alexandra. (2018). WWW'18 Open Challenge: Financial Opinion Mining and Question Answering. WWW '18: Companion Proceedings of the The Web Conference 2018. 1941-1942. 10.1145/3184558.3192301.
- KorHeadline:Gold Commodity News and Dimensions 翻译数据集
- 引用:
Sinha, A., & Khandait, T. (2021, April). Impact of News on the Commodity Market: Dataset and Results. In Future of Information and Communication Conference (pp. 589-601). Springer, Cham.
通用领域基准测试
模型 |
NSMC(ACC) |
PAWS(ACC) |
KorNLI(ACC) |
KorSTS(spearman) |
KorQuAD(EM/F1) |
平均(除 KorQuAD) |
KLUE-RoBERTa (Base) |
90.47 |
84.79 |
81.65 |
84.40 |
86.34/94.40 |
85.33 |
KoELECTRA (Base) |
90.63 |
84.45 |
82.24 |
85.53 |
84.83/93.45 |
85.71 |
KF-DeBERTa (Base) |
91.36 |
86.14 |
84.54 |
85.99 |
86.60/95.07 |
87.01 |
📄 许可证
KF-DeBERTa 的源代码和模型在 MIT 许可证下公开。许可证全文可在 MIT 文件 中查看。本公司对因使用该模型而产生的任何损失不承担责任。
🔗 引用
@proceedings{jeon-etal-2023-kfdeberta,
title = {KF-DeBERTa: Financial Domain-specific Pre-trained Language Model},
author = {Eunkwang Jeon, Jungdae Kim, Minsang Song, and Joohyun Ryu},
booktitle = {Proceedings of the 35th Annual Conference on Human and Cognitive Language Technology},
moth = {oct},
year = {2023},
publisher = {Korean Institute of Information Scientists and Engineers},
url = {http://www.hclt.kr/symp/?lnb=conference},
pages = {143--148},
}