paper-with-me

홈 › Papers

QuantTune: Optimizing Model Quantization with Adaptive Outlier-Driven Fine Tuning

2024-03-11 · Jiun-Man Chen, Yu-Hsuan Chao, Yu-Jie Wang, Ming-Der Shieh, Chih-Chung Hsu, Wei-Fen Lin

Transformer-based models have gained widespread popularity in both the computer vision (CV) and natural language processing (NLP) fields. However, significant challenges arise during post-training linear quantization, leading to noticeable reductions in inference accuracy. Our study focuses on uncovering the underlying causes of these accuracy drops and proposing a quantization-friendly fine-tuning method, \textbf{QuantTune}. Firstly, our analysis revealed that, on average, 65\% of quantization errors result from the precision loss incurred by the dynamic range amplification effect of outliers across the target Transformer-based models. Secondly, \textbf{QuantTune} adjusts weights based on the deviation of outlier activations and effectively constrains the dynamic ranges of the problematic activations. As a result, it successfully mitigates the negative impact of outliers on the inference accuracy of quantized models. Lastly, \textbf{QuantTune} can be seamlessly integrated into the back-propagation pass in the fine-tuning process without requiring extra complexity in inference software and hardware design. Our approach showcases significant improvements in post-training quantization across a range of Transformer-based models, including ViT, Bert-base, and OPT. QuantTune reduces accuracy drops by 12.09\% at 8-bit quantization and 33.8\% at 7-bit compared to top calibration methods, outperforming state-of-the-art solutions by over 18.84\% across ViT models.

📄 PDF Abstract BibTeX arXiv:2403.06497

Code (0)

등록된 구현이 없습니다.

Tasks

Quantization

Methods 이 논문이 사용한 방법론

OPT OPT is a suite of decoder-only pre-trained transformers ranging from 125M to 175B parameters. The model uses an AdamW optimizer and weight decay of 0.1. It follows a linear…

Similar Papers 제목 키워드 기반

Breaking Modality Heterogeneity in Low-Bit Quantization for Large Vision-Language Models

2026-05-19 · Yi Zhong, Haotong Qin, Xindong Zhang, Lei Zhang 외 arxiv

Low-bit post-training quantization (PTQ) is a pivotal technique for deploying Vision-Language Models (VLMs) on resource-constrained devices. However, existing PTQ methods often degrade VLMs' accuracy due to the heterogen…

Outliers and Calibration Sets have Diminishing Effect on Quantization of Modern LLMs

2024-05-31 · Davide Paglieri, Saurabh Dash, Tim Rocktäschel, Jack Parker-Holder

Post-Training Quantization (PTQ) enhances the efficiency of Large Language Models (LLMs) by enabling faster operation and compatibility with more accessible hardware through reduced memory usage, at the cost of small per…

Quantization

Rethinking Channel Dimensions to Isolate Outliers for Low-bit Weight Quantization of Large Language Models

2023-09-27 · Jung Hwan Heo, Jeonghoon Kim, Beomseok Kwon, Byeongwook Kim 외

Large Language Models (LLMs) have recently demonstrated remarkable success across various tasks. However, efficiently serving LLMs has been a challenge due to the large memory bottleneck, specifically in small batch infe…

HumanEvalLanguage ModelingLanguage ModellingMMLU+1

A Unified View of Attention and Residual Sinks: Outlier-Driven Rescaling is Essential for Transformer Training

2026-01-30 · Zihan Qiu, Zeyu Huang, Kaiyue Wen, Peng Jin 외 arxiv

We investigate the functional role of emergent outliers in large language models, specifically attention sinks (a few tokens that consistently receive large attention logits) and residual sinks (a few fixed dimensions wi…

Value-Driven Mixed-Precision Quantization for Patch-Based Inference on Microcontrollers

2024-01-24 · Wei Tao, Shenglin He, Kai Lu, Xiaoyang Qu 외

Deploying neural networks on microcontroller units (MCUs) presents substantial challenges due to their constrained computation and memory resources. Previous researches have explored patch-based inference as a strategy t…

Quantization