English

Shapley Head Pruning: Identifying and Removing Interference in Multilingual Transformers

Computation and Language 2023-05-08 v1 Machine Learning

Abstract

Multilingual transformer-based models demonstrate remarkable zero and few-shot transfer across languages by learning and reusing language-agnostic features. However, as a fixed-size model acquires more languages, its performance across all languages degrades, a phenomenon termed interference. Often attributed to limited model capacity, interference is commonly addressed by adding additional parameters despite evidence that transformer-based models are overparameterized. In this work, we show that it is possible to reduce interference by instead identifying and pruning language-specific parameters. First, we use Shapley Values, a credit allocation metric from coalitional game theory, to identify attention heads that introduce interference. Then, we show that removing identified attention heads from a fixed model improves performance for a target language on both sentence classification and structural prediction, seeing gains as large as 24.7\%. Finally, we provide insights on language-agnostic and language-specific attention heads using attention visualization.

Keywords

Cite

@article{arxiv.2210.05709,
  title  = {Shapley Head Pruning: Identifying and Removing Interference in Multilingual Transformers},
  author = {William Held and Diyi Yang},
  journal= {arXiv preprint arXiv:2210.05709},
  year   = {2023}
}

Comments

8 Pages, 9 Figures

R2 v1 2026-06-28T03:21:52.489Z