AMU-EURANOVA at CASE 2021 Task 1: Assessing the stability of multilingual BERT
arXiv:2106.14625v2712 citations
Originality Synthesis-oriented
AI Analysis
This work addresses a stability issue in a specific NLP task, but it is incremental as it applies an existing method to a new dataset.
The paper tackled the instability problem in fine-tuning multilingual BERT for event information extraction from news, achieving results on a small dataset but without reporting specific performance numbers.
This paper explains our participation in task 1 of the CASE 2021 shared task. This task is about multilingual event extraction from news. We focused on sub-task 4, event information extraction. This sub-task has a small training dataset and we fine-tuned a multilingual BERT to solve this sub-task. We studied the instability problem on the dataset and tried to mitigate it.