On the Way to Lossless Compression of Language Transformers: Exploring Cross-Domain Properties of Quantization
Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
Abstract
Modern Transformers achieved impressive results on various Natural Language Processing tasks over the last few years. The one downside of this success is the size of these models. Huge capacity, which sometimes surpasses billions of parameters, improves generalization abilities, but makes it difficult to employ. Developing field of model compression seeks to reduce the model size and inference latency. This research focuses on one of the compression techniques — Post-Training Quantization. We present a methodology to effectively quantize at least 95% of Transformer weights and corresponding activations to INT8 without any access to task-specific data so the drop in performance does not exceed 0.02%. Furthermore, we provide intriguing observations that reflect cross-domain nature of some of the quantization properties.