Iterative Layer Pruning for Efficient Translation Inference

TL;DR

This paper introduces an iterative layer pruning method guided by layer importance analysis to reduce the size and inference time of large language models for translation, maintaining quality.

Contribution

It presents a novel iterative layer pruning technique specifically designed for efficient translation inference in large language models.

Findings

Significant reduction in model size and inference time.

Maintained translation quality comparable to baseline models.

Effective across multiple language pairs.

Abstract

Large language models (LLMs) have transformed many areas of natural language processing, including machine translation. However, efficient deployment of LLMs remains challenging due to their intensive computational requirements. In this paper, we address this challenge and present our submissions to the Model Compression track at the Conference on Machine Translation (WMT 2025). In our experiments, we investigate iterative layer pruning guided by layer importance analysis. We evaluate this method using the Aya-Expanse-8B model for translation from Czech to German, and from English to Egyptian Arabic. Our approach achieves substantial reductions in model size and inference time, while maintaining the translation quality of the baseline models.