Real-world business applications require a trade-off between language model performance and size. We propose a new method for model compression that relies on vocabulary transfer. We evaluate the method on various vertical domains and downstream tasks. Our results indicate that vocabulary transfer can be effectively used in combination with other compression techniques, yielding a significant reduction in model size and inference time while marginally compromising on performance.
@article{arxiv.2402.09977,
title = {Fast Vocabulary Transfer for Language Model Compression},
author = {Leonidas Gee and Andrea Zugarini and Leonardo Rigutini and Paolo Torroni},
journal= {arXiv preprint arXiv:2402.09977},
year = {2024}
}
Comments
The 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022)