paper-with-me

Papers

Semantics Prompting Data-Free Quantization for Low-Bit Vision Transformers

2024-12-21 · Yunshan Zhong, Yuyao Zhou, Yuxin Zhang, Shen Li, Yong Li, Fei Chao, Zhanpeng Zeng, Rongrong Ji

Data-free quantization (DFQ), which facilitates model quantization without real data to address increasing concerns about data security, has garnered significant attention within the model compression community. Recently, the unique architecture of vision transformers (ViTs) has driven the development of specialized DFQ techniques. However, we observe that the synthetic images from existing methods suffer from the deficient semantics issue compared to real images, thereby compromising performance. Motivated by this, we propose SPDFQ, a Semantics Prompting Data-Free Quantization method for ViTs. First, SPDFQ incorporates Attention Priors Alignment (APA), which uses randomly generated attention priors to enhance the semantics of synthetic images. Second, SPDFQ introduces Multi-Semantic Reinforcement (MSR), which utilizes localized patch optimization to prompt efficient parameterization and diverse semantics in synthetic images. Finally, SPDFQ employs Softlabel Learning (SL), where soft learning targets are adapted to encourage more complex semantics and accommodate images augmented by MSR. Experimental results demonstrate that SPDFQ significantly outperforms existing methods. For instance, SPDFQ achieves a 15.52% increase in top-1 accuracy on ImageNet for W4A4 ViT-B

📄 PDF Abstract BibTeX arXiv:2412.16553

Code (0)

등록된 구현이 없습니다.

Tasks

Data Free QuantizationModel CompressionQuantization

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Point Cloud Quantization through Multimodal Prompting for 3D Understanding

2025-11-15 · Hongxuan Li, Wencheng Zhu, Huiying Xu, Xinzhong Zhu 외 arxiv

Vector quantization has emerged as a powerful tool in large-scale multimodal models, unifying heterogeneous representations through discrete token encoding. However, its effectiveness hinges on robust codebook design. Cu…

Enhancing Generalization in Data-free Quantization via Mixup-class Prompting

2025-07-29 · Jiwoong Park, Chaeun Lee, Yongseok Choi, Sein Park 외 arxiv

Post-training quantization (PTQ) improves efficiency but struggles with limited calibration data, especially under privacy constraints. Data-free quantization (DFQ) mitigates this by generating synthetic images using gen…

Prompt Engineering

Robustness-Guided Image Synthesis for Data-Free Quantization

2023-10-05 · Jianhong Bai, Yuchen Yang, Huanpeng Chu, Hualiang Wang 외

Quantization has emerged as a promising direction for model compression. Recently, data-free quantization has been widely studied as a promising method to avoid privacy concerns, which synthesizes images as an alternativ…

Data Free QuantizationDiversityImage GenerationModel Compression+1

Unlocking Vision-Language Models for Video Anomaly Detection via Fine-Grained Prompting

2025-10-02 · Shu Zou, Xinyu Tian, Lukas Wesemann, Fabian Waschkowski 외 arxiv

Prompting has emerged as a practical way to adapt frozen vision-language models (VLMs) for video anomaly detection (VAD). Yet, existing prompts are often overly abstract, overlooking the fine-grained human-object interac…

Video Anomaly Detection

Selective Coupling of Decoupled Informative Regions: Masked Attention Alignment for Data-Free Quantization of Vision Transformers

2026-06-03 · Biao Qian, Yang Wang, Yong Wu, Jungong Han arxiv

Data-Free Quantization (DFQ) addresses data security concerns by synthesizing samples, without accessing real data. It has garnered increasing attention in the context of Vision Transformers (ViTs), owing to the superior…