Iterative Layer Pruning for Efficient Translation Inference
This addresses computational efficiency challenges for deploying LLMs in translation tasks, though it appears incremental as it applies existing pruning techniques to specific translation scenarios.
The paper tackles efficient deployment of large language models for machine translation by proposing iterative layer pruning guided by layer importance analysis, achieving substantial reductions in model size and inference time while maintaining baseline translation quality.
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.