Multi-task learning for few-shot biomedical relation extraction

被引:8
|
作者
Moscato, Vincenzo [1 ,2 ]
Napolano, Giuseppe [1 ]
Postiglione, Marco [1 ]
Sperli, Giancarlo [1 ,2 ]
机构
[1] Univ Naples Federico II, Dept Elect & Informat Technol, Via Claudio 21, I-80125 Naples, Italy
[2] CINI Consorzio Interuniv Nazl Informat, Rome, Italy
关键词
Biomedical NLP; Data-centric AI; Deep learning; Low-resource learning; Transformers; CORPUS;
D O I
10.1007/s10462-023-10484-6
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Artificial intelligence (AI) has advanced rapidly, but it has limited impact on biomedical text understanding due to a lack of annotated datasets (a.k.a. few-shot learning). Multi-task learning, which uses data from multiple datasets and tasks with related syntax and semantics, has potential to address this issue. However, the effectiveness of this approach heavily relies on the quality of the available data and its transferability between tasks. In this paper, we propose a framework, built upon a state-of-the-art multi-task method (i.e. MT-DNN), that leverages different publicly available biomedical datasets to enhance relation extraction performance. Our model employs a transformer-based architecture with shared encoding layers across multiple tasks, and task-specific classification layers to generate task-specific representations. To further improve performance, we utilize a knowledge distillation technique. In our experiments, we assess the impact of incorporating biomedical datasets in a multi-task learning setting and demonstrate that it consistently outperforms state-of-the-art few-shot learning methods in cases of limited data. This results in significant improvement across most datasets and few-shot scenarios, particularly in terms of recall scores.
引用
收藏
页码:13743 / 13763
页数:21
相关论文
共 50 条
  • [1] Multi-task learning for few-shot biomedical relation extraction
    Vincenzo Moscato
    Giuseppe Napolano
    Marco Postiglione
    Giancarlo Sperlì
    Artificial Intelligence Review, 2023, 56 : 13743 - 13763
  • [2] Multi-task classification network for few-shot learning
    Ji, Zhong
    Liu, Yuanheng
    Wang, Xuan
    Liu, Jingren
    Cao, Jiale
    Yu, Yunlong
    INTERNATIONAL JOURNAL OF MULTIMEDIA INFORMATION RETRIEVAL, 2025, 14 (01)
  • [3] Adaptive Multi-task Learning for Few-Shot Object Detection
    Ren, Yan
    Li, Yanling
    Kong, Adams Wai-Kin
    COMPUTER VISION-ECCV 2024, PT VII, 2025, 15065 : 297 - 314
  • [4] Few-Shot KBQA Method Based on Multi-Task Learning
    Ren, Yuan
    Li, Xutong
    Liu, Xudong
    Zhang, Richong
    2024 IEEE INTERNATIONAL CONFERENCE ON BIG DATA AND SMART COMPUTING, IEEE BIGCOMP 2024, 2024, : 226 - 233
  • [5] Hierarchical Prompt Tuning for Few-Shot Multi-Task Learning
    Liu, Jingping
    Chen, Tao
    Liang, Zujie
    Jiang, Haiyun
    Xiao, Yanghua
    Wei, Feng
    Qian, Yuxi
    Hao, Zhenghong
    Han, Bing
    PROCEEDINGS OF THE 32ND ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, CIKM 2023, 2023, : 1556 - 1565
  • [6] Multi-task Based Few-Shot Learning for Disease Similarity Measurement
    Gao, Jianliang
    Tian, Ling
    Liu, Yuxin
    Wang, Jianxin
    Li, Zhao
    Hu, Xiaohua
    2020 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE, 2020, : 263 - 268
  • [7] Improving Few-Shot Learning Through Multi-task Representation Learning Theory
    Bouniot, Quentin
    Redko, Ievgen
    Audigier, Romaric
    Loesch, Angelique
    Habrard, Amaury
    COMPUTER VISION, ECCV 2022, PT XX, 2022, 13680 : 435 - 452
  • [8] Exploring Task Difficulty for Few-Shot Relation Extraction
    Han, Jiale
    Cheng, Bo
    Lu, Wei
    2021 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP 2021), 2021, : 2605 - 2616
  • [9] Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues
    Cordier, Thibault
    Urvoy, Tanguy
    Lefevre, Fabrice
    Rojas-Barahona, Lina M.
    17TH CONFERENCE OF THE EUROPEAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, EACL 2023, 2023, : 432 - 441
  • [10] Multi-task few-shot learning with composed data augmentation for image classification
    Zhang, Rui
    Yang, Yixin
    Li, Yang
    Wang, Jiabao
    Li, Hang
    Miao, Zhuang
    IET COMPUTER VISION, 2023, 17 (02) : 211 - 221