Self-Distilled Quantization: Achieving High Compression Rates in Transformer-Based Language Models

Autor:	Neill, James O', Dutta, Sourav
Rok vydání:	2023
Předmět:	Computer Science - Computation and Language Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	We investigate the effects of post-training quantization and quantization-aware training on the generalization of Transformer language models. We present a new method called self-distilled quantization (SDQ) that minimizes accumulative quantization errors and outperforms baselines. We apply SDQ to multilingual models XLM-R-Base and InfoXLM-Base and demonstrate that both models can be reduced from 32-bit floating point weights to 8-bit integer weights while maintaining a high level of performance on the XGLUE benchmark. Our results also highlight the challenges of quantizing multilingual models, which must generalize to languages they were not fine-tuned on.
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2307.05972 Zobrazit plný text záznamu View this record from Arxiv