SLICER: Sliced Fine-Tuning for Low-Resource Cross-Lingual Transfer for Named Entity Recognition

Fabian David Schmidt, Ivan Vulić, Goran Glavaš


Abstract
Large multilingual language models generally demonstrate impressive results in zero-shot cross-lingual transfer, yet often fail to successfully transfer to low-resource languages, even for token-level prediction tasks like named entity recognition (NER). In this work, we introduce a simple yet highly effective approach for improving zero-shot transfer for NER to low-resource languages. We observe that NER fine-tuning in the source language decontextualizes token representations, i.e., tokens increasingly attend to themselves. This increased reliance on token information itself, we hypothesize, triggers a type of overfitting to properties that NE tokens within the source languages share, but are generally not present in NE mentions of target languages. As a remedy, we propose a simple yet very effective sliced fine-tuning for NER (SLICER) that forces stronger token contextualization in the Transformer: we divide the transformed token representations and classifier into disjoint slices that are then independently classified during training. We evaluate SLICER on two standard benchmarks for NER that involve low-resource languages, WikiANN and MasakhaNER, and show that it (i) indeed reduces decontextualization (i.e., extent to which NE tokens attend to themselves), consequently (ii) yielding consistent transfer gains, especially prominent for low-resource target languages distant from the source language.
Anthology ID:
2022.emnlp-main.740
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
10775–10785
Language:
URL:
https://aclanthology.org/2022.emnlp-main.740
DOI:
10.18653/v1/2022.emnlp-main.740
Bibkey:
Cite (ACL):
Fabian David Schmidt, Ivan Vulić, and Goran Glavaš. 2022. SLICER: Sliced Fine-Tuning for Low-Resource Cross-Lingual Transfer for Named Entity Recognition. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 10775–10785, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
SLICER: Sliced Fine-Tuning for Low-Resource Cross-Lingual Transfer for Named Entity Recognition (Schmidt et al., EMNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.emnlp-main.740.pdf