paper-with-me

Papers

Quadapter: Adapter for GPT-2 Quantization

2022-11-30 · Minseop Park, Jaeseong You, Markus Nagel, Simyung Chang

Transformer language models such as GPT-2 are difficult to quantize because of outliers in activations leading to a large quantization error. To adapt to the error, one must use quantization-aware training, which entails a fine-tuning process based on the dataset and the training pipeline identical to those for the original model. Pretrained language models, however, often do not grant access to their datasets and training pipelines, forcing us to rely on arbitrary ones for fine-tuning. In that case, it is observed that quantization-aware training overfits the model to the fine-tuning data. For quantization without overfitting, we introduce a quantization adapter (Quadapter), a small set of parameters that are learned to make activations quantization-friendly by scaling them channel-wise. It keeps the model parameters unchanged. By applying our method to the challenging task of quantizing GPT-2, we demonstrate that it effectively prevents the overfitting and improves the quantization performance.

📄 PDF Abstract BibTeX arXiv:2211.16912

Code (0)

등록된 구현이 없습니다.

Tasks

Quantization

Methods 이 논문이 사용한 방법론

Attention 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Weight Decay 설명 없음
Multi-Head Attention 설명 없음
Discriminative Fine-Tuning Discriminative Fine-Tuning is a fine-tuning strategy that is used for ULMFiT type models. Instead of using the same learning rate…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…

Similar Papers 제목 키워드 기반

QWHA: Quantization-Aware Walsh-Hadamard Adaptation for Parameter-Efficient Fine-Tuning on Large Language Models

2025-09-22 · Hyesung Jeon, Seojune Lee, Beomseok Kang, Yulhwa Kim 외 arxiv

The demand for efficient deployment of large language models (LLMs) has driven interest in quantization, which reduces inference cost, and parameter-efficient fine-tuning (PEFT), which lowers training overhead. This moti…

parameter-efficient fine-tuning

Revisiting the Parameter Efficiency of Adapters from the Perspective of Precision Redundancy

2023-07-31 · ICCV 2023 1 · Shibo Jie, Haoqing Wang, Zhi-Hong Deng

Current state-of-the-art results in computer vision depend in part on fine-tuning large pre-trained vision models. However, with the exponential growth of model sizes, the conventional full fine-tuning, which needs to st…

Quantization

LoRAQuant: Mixed-Precision Quantization of LoRA to Ultra-Low Bits

2025-10-30 · Amir Reza Mirzaei, Yuqiao Wen, Yanshuai Cao, Lili Mou arxiv

Low-Rank Adaptation (LoRA) has become a popular technique for parameter-efficient fine-tuning of large language models (LLMs). In many real-world scenarios, multiple adapters are loaded simultaneously to enable LLM custo…

parameter-efficient fine-tuningMathematical Reasoning

On-the-Fly Adaptation to Quantization: Configuration-Aware LoRA for Efficient Fine-Tuning of Quantized LLMs

2025-09-22 · Rongguang Ye, Ming Tang, Edith C. H. Ngai arxiv

As increasingly large pre-trained models are released, deploying them on edge devices for privacy-preserving applications requires effective compression. Recent works combine quantization with the fine-tuning of high-pre…

QuEPT: Quantized Elastic Precision Transformers with One-Shot Calibration for Multi-Bit Switching

2026-02-13 · Ke Xu, Yixin Wang, Zhongcheng Li, Hao Cui 외 arxiv

Elastic precision quantization enables multi-bit deployment via a single optimization pass, fitting diverse quantization scenarios.Yet, the high storage and optimization costs associated with the Transformer architecture…