Search Results for author:
Found 7 papers, 5 papers with code
Enhanced Distribution Alignment for Post-Training Quantization of Diffusion Models
Specifically, at the calibration sample level, we select calibration samples based on the density and diversity in the latent space, thus facilitating the alignment of their distribution with the overall samples; and at the reconstruction output level, we propose Fine-grained Block Reconstruction, which can align the outputs of the quantized model and the full-precision model at different network granularity.
BinaryViT: Towards Efficient and Accurate Binary Vision Transformers
In this paper, we first argue empirically that the severe performance degradation is mainly caused by the weight oscillation in the binarization training and the information distortion in the activation of ViTs.
Patch-wise Mixed-Precision Quantization of Vision Transformer
As emerging hardware begins to support mixed bit-width arithmetic computation, mixed-precision quantization is widely used to reduce the complexity of neural networks.
RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformers
Post-training quantization (PTQ), which only requires a tiny dataset for calibration without end-to-end retraining, is a light and practical model compression technique.
PSAQ-ViT V2: Towards Accurate and General Data-Free Quantization for Vision Transformers
In this paper, we propose PSAQ-ViT V2, a more accurate and general data-free quantization framework for ViTs, built on top of PSAQ-ViT.
HFT: Lifting Perspective Representations via Hybrid Feature Transformation
In order to reap the benefits and avoid the drawbacks of CBFT and CFFT, we propose a novel framework with a Hybrid Feature Transformation module (HFT).
Patch Similarity Aware Data-Free Quantization for Vision Transformers
The above insights guide us to design a relative value metric to optimize the Gaussian noise to approximate the real images, which are then utilized to calibrate the quantization parameters.
