mlsanityData to 14 Oct 2026

IPTQ-ViT: Post-Training Quantization of Non-linear Functions for Integer-only Vision Transformers

arXiv 2511.15369 ↗First posted 19 Nov 2025cs.CV cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Previous Quantization-Aware Training (QAT) methods for vision transformers rely on expensive retraining to recover accuracy loss in non-linear layer quantization, limiting their use in resource-constrained environments. In contrast, existing Post-Training Quantization (PTQ) methods either partially quantize non-linear functions or adjust activation distributions to maintain accuracy but fail to achieve fully integer-only inference. In this paper, we introduce IPTQ-ViT, a novel PTQ framework for fully integer-only vision transformers without retraining. We present approximation functions: a polynomial-based GELU optimized for vision data and a bit-shifting-based Softmax designed to improve approximation accuracy in PTQ. In addition, we propose a unified metric integrating quantization sensitivity, perturbation, and computational cost to select the optimal approximation function per activation layer. IPTQ-ViT outperforms previous PTQ methods, achieving up to 6.44%p (avg. 1.78%p) top-1 accuracy improvement for image classification, 1.0 mAP for object detection. IPTQ-ViT outperforms partial floating-point PTQ methods under W8A8 and W4A8, and achieves accuracy and latency comparable to integer-only QAT methods. We plan to release our code https://github.com/gihwan-kim/IPTQ-ViT.git.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 31 of its 39 references that are in the map. First posted on arXiv 19 Nov 2025.

Builds on

31 of 39 references.

References · 31

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  4. 2020End-to-End Object Detection with Transformerscited by 7,414
  5. 2017Mask R-CNNcited by 7,277
  6. 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
  7. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  8. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  9. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,385
  10. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,276
  11. 2021Segmenter: Transformer for Semantic Segmentationcited by 855
  12. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 815
19 more
  1. 2021CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classificationcited by 806
  2. 2018Quantizing deep convolutional networks for efficient inference: A whitepapercited by 521
  3. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 402
  4. 2019HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networkscited by 261
  5. 2021I-BERT: Integer-only BERT Quantizationcited by 255
  6. 2020Binary Neural Networks: A Surveycited by 190
  7. 2020Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluationcited by 188
  8. 2021PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantizationcited by 179
  9. 2021FQ-ViT: Post-Training Quantization for Fully Quantized Vision Transformercited by 151
  10. 2022RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformerscited by 137
  11. 2022I-ViT: Integer-only Quantization for Efficient Vision Transformer Inferencecited by 95
  12. 2022NoisyQuant: Noisy Bias-Enhanced Post-Training Activation Quantization for Vision Transformerscited by 61
  13. 2022Patch Similarity Aware Data-Free Quantization for Vision Transformerscited by 56
  14. 2022PSAQ-ViT V2: Towards Accurate and General Data-Free Quantization for Vision Transformerscited by 42
  15. 2023OcTr: Octree-based Transformer for 3D Object Detectioncited by 32
  16. 2024ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterizationcited by 30
  17. 2024I-LLM: Efficient Integer-Only Inference for Fully-Quantized Low-Bit Large Language Modelscited by 23
  18. 2024CLAMP-ViT: Contrastive Data-Free Learning for Adaptive Post-Training Quantization of ViTscited by 17
  19. 2024Mixed Non-linear Quantization for Vision Transformerscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Enabling Fully Integer-Only Inference for Lightweight Detection Transformerscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025AIQViT: Architecture-Informed Post-Training Quantization for Vision Transformers
  2. 2023I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs Quantization
  3. 2024DopQ-ViT: Towards Distribution-Friendly and Outlier-Aware Post-Training Quantization for Vision Transformers
  4. 2024MPTQ-ViT: Mixed-Precision Post-Training Quantization for Vision Transformer
  5. 2023TSPTQ-ViT: Two-scaled post-training quantization for vision transformer
  6. 2024ADFQ-ViT: Activation-Distribution-Friendly Post-Training Quantization for Vision Transformers
  7. 2025GPLQ: A General, Practical, and Lightning QAT Method for Vision Transformers
  8. 2025APHQ-ViT: Post-Training Quantization with Average Perturbation Hessian Based Reconstruction for Vision Transformers