paper-with-me

Papers

Q&C: When Quantization Meets Cache in Efficient Image Generation

2025-03-04 · Xin Ding, Xin Li, Haotong Qin, Zhibo Chen

Quantization and cache mechanisms are typically applied individually for efficient Diffusion Transformers (DiTs), each demonstrating notable potential for acceleration. However, the promoting effect of combining the two mechanisms on efficient generation remains under-explored. Through empirical investigation, we find that the combination of quantization and cache mechanisms for DiT is not straightforward, and two key challenges lead to severe catastrophic performance degradation: (i) the sample efficacy of calibration datasets in post-training quantization (PTQ) is significantly eliminated by cache operation; (ii) the combination of the above mechanisms introduces more severe exposure bias within sampling distribution, resulting in amplified error accumulation in the image generation process. In this work, we take advantage of these two acceleration mechanisms and propose a hybrid acceleration method by tackling the above challenges, aiming to further improve the efficiency of DiTs while maintaining excellent generation capability. Concretely, a temporal-aware parallel clustering (TAP) is designed to dynamically improve the sample selection efficacy for the calibration within PTQ for different diffusion steps. A variance compensation (VC) strategy is derived to correct the sampling distribution. It mitigates exposure bias through an adaptive correction factor generation. Extensive experiments have shown that our method has accelerated DiTs by 12.7x while preserving competitive generation capability. The code will be available at https://github.com/xinding-sys/Quant-Cache.

📄 PDF Abstract BibTeX arXiv:2503.02508

Code (0)

등록된 구현이 없습니다.

Tasks

Image GenerationQuantization

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

InnerQ: Hardware-Aware Tuning-Free Quantization of KV Cache for Large Language Models

2026-02-26 · Sayed Mohammadreza Tayaranian Hosseini, Amir Ardakani, Warren J. Gross arxiv

When transformer-based language models are deployed for text generation, most of the inference time is spent in the decoding stage, where output tokens are generated sequentially. Reducing the hardware cost of each decod…

Text Generation

ZipCache: Accurate and Efficient KV Cache Quantization with Salient Token Identification

2024-05-23 · Yefei He, Luoming Zhang, Weijia Wu, Jing Liu 외

KV cache stores key and value states from previous tokens to avoid re-computation, yet it demands substantial storage space, especially for long sequences. Adaptive KV cache compression seeks to discern the saliency of t…

GPUGSM8KQuantization

Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization

2026-02-03 · Haocheng Xi, Shuo Yang, Yilong Zhao, Muyang Li 외 arxiv

Despite rapid progress in autoregressive video diffusion, an emerging system algorithm bottleneck limits both deployability and generation capability: KV cache memory. In autoregressive video generation models, the KV ca…

Video Generation

Prefixing Attention Sinks can Mitigate Activation Outliers for Large Language Model Quantization

2024-06-17 · Seungwoo Son, Wonpyo Park, Woohyun Han, Kyuyeun Kim 외

Despite recent advances in LLM quantization, activation quantization remains to be challenging due to the activation outliers. Conventional remedies, e.g., mixing precisions for different channels, introduce extra overhe…

Language ModelingLanguage ModellingLarge Language ModelQuantization

AsymKV: Enabling 1-Bit Quantization of KV Cache with Layer-Wise Asymmetric Quantization Configurations

2024-10-17 · Qian Tao, Wenyuan Yu, Jingren Zhou

Large language models have shown exceptional capabilities in a wide range of tasks, such as text generation and video generation, among others. However, due to their massive parameter count, these models often require su…

DecoderQuantizationText GenerationVideo Generation