QNCD: Quantization Noise Correction for Diffusion Models
Diffusion models have revolutionized image synthesis, setting new benchmarks in quality and creativity. However, their widespread adoption is hindered by the intensive computation required during the iterative denoising process. Post-training quantization (PTQ) presents a solution to accelerate sampling, aibeit at the expense of sample quality, extremely in low-bit settings. Addressing this, our study introduces a unified Quantization Noise Correction Scheme (QNCD), aimed at minishing quantization noise throughout the sampling process. We identify two primary quantization challenges: intra and inter quantization noise. Intra quantization noise, mainly exacerbated by embeddings in the resblock module, extends activation quantization ranges, increasing disturbances in each single denosing step. Besides, inter quantization noise stems from cumulative quantization deviations across the entire denoising process, altering data distributions step-by-step. QNCD combats these through embedding-derived feature smoothing for eliminating intra quantization noise and an effective runtime noise estimatiation module for dynamicly filtering inter quantization noise. Extensive experiments demonstrate that our method outperforms previous quantization methods for diffusion models, achieving lossless results in W4A8 and W8A8 quantization settings on ImageNet (LDM-4). Code is available at: https://github.com/huanpengchu/QNCD
Code (1)
Tasks
DenoisingImage GenerationQuantizationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Q-Drift: Quantization-Aware Drift Correction for Diffusion Model Sampling
Post-training quantization (PTQ) is a practical path to deploy large diffusion models, but quantization noise can accumulate over the denoising trajectory and degrade generation quality. We propose Q-Drift, a principled …
Timestep-Aware Correction for Quantized Diffusion Models
Diffusion models have marked a significant breakthrough in the synthesis of semantically coherent images. However, their extensive noise estimation networks and the iterative generation process limit their wider applicat…
AttributeNoise EstimationQuantizationSoftmax Bias Correction for Quantized Generative Models
Post-training quantization (PTQ) is the go-to compression technique for large generative models, such as stable diffusion or large language models. PTQ methods commonly keep the softmax activation in higher precision as …
Language ModelingLanguage ModellingQuantizationQuantized Keys Steal Attention: Bias Correction for KV-Cache Compression in Video Diffusion
Chunk-wise autoregressive video diffusion models rely on a KV cache of previously generated chunks to avoid redundant computation, but this cache quickly becomes a memory bottleneck as videos grow longer. Methods that qu…
Bridging the Gap between Gaussian Diffusion Models and Universal Quantization for Image Compression
Generative neural image compression supports data representation at extremely low bitrate, synthesizing details at the client and consistently producing highly realistic images. By leveraging the similarities between qua…
Image CompressionQuantization