logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Benchmarking context-based data augmentation for biomedical NER: A reproducible framework

delete2026-08-21
delete0
PRE
AI
A
Andrea Vignali *
M
Marco Postiglione
G
Giancarlo Sperlí
I
Ilaria Bartolini
V
Vincenzo Moscato
A
Andrea Seveso
S
Saurabh Kumar
DOI：10.1016/j.is.2026.102792delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Reproducibility framework for COSINER experiments, including code and instructions. • Replicated experiments show weak reproducibility and high correlation with originals. • Extends COSINER support to Windows and Linux operating systems. • Provides automated workflows for setup, execution, and processing of all experiments.
Keywords:
Named Entity Recognition
Data augmentation
Similarity learning
Few-shot learning
Reproducibility framework

Journal

I
Information Systems
IF:
3.4
Papers:
118
Citations:
0

Organization

U
university of milano-bicocca
Scholars:
2.0W
Papers: 1.5W
Citations: 22
I
indian institute of technology hyderabad
Scholars:
483
Papers: 205
Citations: 0
U
university of bologna
Scholars:
5.9K
Papers: 2.5K
Citations: 0
U
University of Naples Federico II
Scholars:
4.7W
Papers: 3.6W
Citations: 51
N
Northwestern University
Scholars:
6.1W
Papers: 5.3W
Citations: 3.9K
researcher View more organizations