AI Paper Atlas

Research map / Architectures and efficiency / Quantization

NoisyQuant: Noisy Bias-Enhanced Post-Training Activation Quantization for Vision Transformers

Yijiang Liu, Huanrui Yang, Zhen Dong, Kurt Keutzer, Li Du, Shanghang Zhang

CVPR 2023 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 52 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add fixed uniform noisy bias before activation quantizationEasyQuant on ViT-S · ImageNet-2012, W6A6top-1 accuracy: 75.13 -> 76.86Improved
Add fixed uniform noisy bias before activation quantizationEasyQuant on DeiT-S · ImageNet-2012, W6A6top-1 accuracy: 75.27 -> 76.37Improved
Add fixed uniform noisy bias before activation quantizationPTQ4ViT on DeiT-S · ImageNet-2012, W6A6top-1 accuracy: 76.28 -> 77.43Improved
Add fixed uniform noisy bias before activation quantizationPTQ4ViT on ViT-S · ImageNet-2012, W6A6top-1 accuracy: 78.63 -> 78.65No clear effect
Add fixed uniform noisy bias before activation quantizationPTQ4ViT on Swin-B · ImageNet-2012, W6A6top-1 accuracy: 84.01 -> 84.68Improved
Add noisy bias to each layer type separately (ablation)EasyQuant on DeiT-S · ImageNet-2012, W6A6top-1 accuracy: 75.27 -> 75.38 (qkv); 75.27 -> 75.45 (proj); 75.27 -> 75.33 (fc1); 75.27 -> 76.21 (fc2)Improved
Add noisy bias to qkv, proj, fc1, and fc2 layers (ablation)EasyQuant on DeiT-S · ImageNet-2012, W6A6top-1 accuracy: 75.27 -> 76.37Improved
Add noisy bias to each layer type separately (ablation)EasyQuant on Swin-T · ImageNet-2012, W6A6top-1 accuracy: 79.51 -> 79.53 (qkv); 79.51 -> 79.56 (proj); 79.51 -> 79.52 (fc1); 79.51 -> 79.80 (fc2)Improved
Add noisy bias to qkv, proj, fc1, and fc2 layers (ablation)EasyQuant on Swin-T · ImageNet-2012, W6A6top-1 accuracy: 79.51 -> 80.01Improved
Use calibration size 32 rather than 1024 (ablation)NoisyQuant · ImageNet-2012, W6A6top-1 accuracy: 76.86 -> 76.81 (ViT-S); 76.37 -> 76.30 (DeiT-S); 80.01 -> 79.97 (Swin-T)No clear effect
Use calibration size 128 rather than 1024 (ablation)NoisyQuant · ImageNet-2012, W6A6top-1 accuracy: 76.86 -> 76.87 (ViT-S); 76.37 -> 76.47 (DeiT-S); 80.01 -> 80.13 (Swin-T)No clear effect
Store noisy bias and denoising bias as INT16 instead of FP32 (ablation)NoisyQuant · ImageNet-2012top-1 accuracyNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy