mlsanityData to 14 Oct 2026

I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs Quantization

arXiv 2311.10126 ↗First posted 16 Nov 2023cs.CVcited by 19

Not on the map. Why?

Abstract (arXiv)

Albeit the scalable performance of vision transformers (ViTs), the dense computational costs (training & inference) undermine their position in industrial applications. Post-training quantization (PTQ), tuning ViTs with a tiny dataset and running in a low-bit format, well addresses the cost issue but unluckily bears more performance drops in lower-bit cases. In this paper, we introduce I&S-ViT, a novel method that regulates the PTQ of ViTs in an inclusive and stable fashion. I&S-ViT first identifies two issues in the PTQ of ViTs: (1) Quantization inefficiency in the prevalent log2 quantizer for post-Softmax activations; (2) Rugged and magnified loss landscape in coarse-grained quantization granularity for post-LayerNorm activations. Then, I&S-ViT addresses these issues by introducing: (1) A novel shift-uniform-log2 quantizer (SULQ) that incorporates a shift mechanism followed by uniform quantization to achieve both an inclusive domain representation and accurate distribution approximation; (2) A three-stage smooth optimization strategy (SOS) that amalgamates the strengths of channel-wise and layer-wise quantization to enable stable learning. Comprehensive evaluations across diverse vision tasks validate I&S-ViT' superiority over existing PTQ of ViTs methods, particularly in low-bit scenarios. For instance, I&S-ViT elevates the performance of 3-bit ViT-B by an impressive 50.68%.

Key facts. Cited by 19 papers in the map (data through 14 Oct 2026). Builds on 52 of its 61 references that are in the map. First posted on arXiv 16 Nov 2023.

Builds on

52 of 61 references.

References · 52

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  5. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  6. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  7. 2020End-to-End Object Detection with Transformerscited by 7,414
  8. 2017Mask R-CNNcited by 7,277
  9. 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
  10. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  11. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,887
  12. 2021ViViT: A Video Vision Transformercited by 1,620
38 more
  1. 2020Sharpness-Aware Minimization for Efficiently Improving Generalizationcited by 1,417
  2. 2020Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformerscited by 1,369
  3. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,276
  4. 2017Visualizing the Loss Landscape of Neural Netscited by 1,130
  5. 2022Token Merging: Your ViT But Fastercited by 1,021
  6. 2021MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformercited by 642
  7. 2019Learned Step Size Quantizationcited by 628
  8. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 621
  9. 2018Quantizing deep convolutional networks for efficient inference: A whitepapercited by 521
  10. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 479
  11. 2021Post-Training Quantization for Vision Transformercited by 278
  12. 2019Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networkscited by 238
  13. 2022QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantizationcited by 180
  14. 2021PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantizationcited by 179
  15. 2020BinaryBERT: Pushing the Limit of BERT Quantizationcited by 178
  16. 2018Learning to Quantize Deep Networks by Optimizing Quantization Intervals with Task Losscited by 160
  17. 2021FQ-ViT: Post-Training Quantization for Fully Quantized Vision Transformercited by 151
  18. 2019Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networkscited by 148
  19. 2022RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformerscited by 137
  20. 2022Q-ViT: Accurate and Fully Quantized Low-bit Vision Transformercited by 108
  21. 2022I-ViT: Integer-only Quantization for Efficient Vision Transformer Inferencecited by 95
  22. 2022MiniViT: Compressing Vision Transformers with Weight Multiplexingcited by 88
  23. 2023DiffRate : Differentiable Compression Rate for Efficient Vision Transformerscited by 77
  24. 2022PD-Quant: Post-Training Quantization based on Prediction Difference Metriccited by 73
  25. 2023Towards Accurate Post-Training Quantization for Vision Transformercited by 71
  26. 2022NoisyQuant: Noisy Bias-Enhanced Post-Training Activation Quantization for Vision Transformerscited by 61
  27. 2020EasyQuant: Post-training Quantization via Scale Optimizationcited by 57
  28. 2023Oscillation-free Quantization for Low-bit Vision Transformerscited by 57
  29. 2022Patch Similarity Aware Data-Free Quantization for Vision Transformerscited by 56
  30. 2021ReCU: Reviving the Dead Weights in Binary Neural Networkscited by 52
  31. 2021IntraQ: Learning Synthetic Images with Intra-Class Heterogeneity for Zero-Shot Network Quantizationcited by 43
  32. 2022Q-ViT: Fully Differentiable Quantization for Vision Transformercited by 37
  33. 2022SDQ: Stochastic Differentiable Quantization with Mixed Precisioncited by 28
  34. 2022Dynamic Dual Trainable Bounds for Ultra-low Precision Super-Resolution Networkscited by 21
  35. 2023PowerQuant: Automorphism Search for Non-Uniform Quantizationcited by 20
  36. 2023Jumping through Local Minima: Quantization in the Loss Landscape of Vision Transformerscited by 18
  37. 2021Fine-grained Data Distribution Alignment for Post-Training Quantizationcited by 9
  38. 2023MBQuant: A Novel Multi-Branch Topology Method for Arbitrary Bit-width Network Quantizationcited by 6

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 6; 2025: 6; 2026: 7.

19 papers.

Sort:

By area:

  1. 2026Flow caching for autoregressive video generationcited by 25
  2. 2024DopQ-ViT: Towards Distribution-Friendly and Outlier-Aware Post-Training Quantization for Vision Transformerscited by 19
  3. 2025APHQ-ViT: Post-Training Quantization with Average Perturbation Hessian Based Reconstruction for Vision Transformerscited by 19
  4. 2025FIMA-Q: Post-Training Quantization for Vision Transformers by Fisher Information Matrix Approximationcited by 19
  5. 2024Model Quantization and Hardware Acceleration for Vision Transformers: A Comprehensive Surveycited by 13
  6. 2025LRQ-DiT: Log-Rotation Post-Training Quantization of Diffusion Transformers for Image and Video Generationcited by 12
  7. 2025Low-bit Model Quantization for Deep Neural Networks: A Surveycited by 11
  8. 2024ADFQ-ViT: Activation-Distribution-Friendly Post-Training Quantization for Vision Transformerscited by 7
  9. 2026Motion-Aware Caching for Efficient Autoregressive Video Generationcited by 6
  10. 2024Mixed Non-linear Quantization for Vision Transformerscited by 3
  11. 2024Semantic Alignment and Reinforcement for Data-Free Quantization of Vision Transformerscited by 3
  12. 2025AHCQ-SAM: Toward Accurate and Hardware-Compatible Post-Training Segment Anything Model Quantizationcited by 3
7 more
  1. 2025Pack-PTQ: Advancing Post-training Quantization of Neural Networks by Pack-wise Reconstructioncited by 1
  2. 2024Progressive Fine-to-Coarse Reconstruction for Accurate Low-Bit Post-Training Quantization in Vision Transformerscited by 0
  3. 2026ALGOGEN: Tool-Generated Verifiable Traces for Reliable Algorithm Visualizationcited by 0
  4. 2026MARR: Module-Adaptive Residual Reconstruction for Low-Bit Post-Training Quantizationcited by 0
  5. 2026ScalePredictor: Instance-aware Scale Learning for Accurate Quantization of Vision Transformerscited by 0
  6. 2026GLF-Q: Global-Local Feature-based Quantization for Vision Transformerscited by 0
  7. 2026STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025AIQViT: Architecture-Informed Post-Training Quantization for Vision Transformers
  2. 2024AdaLog: Post-Training Quantization for Vision Transformers with Adaptive Logarithm Quantizer
  3. 2023TSPTQ-ViT: Two-scaled post-training quantization for vision transformer
  4. 2024MPTQ-ViT: Mixed-Precision Post-Training Quantization for Vision Transformer
  5. 2025IPTQ-ViT: Post-Training Quantization of Non-linear Functions for Integer-only Vision Transformers
  6. 2024P2-ViT: Power-of-Two Post-Training Quantization and Acceleration for Fully Quantized Vision Transformer
  7. 2024Trio-ViT: Post-Training Quantization and Acceleration for Softmax-Free Efficient Vision Transformer
  8. 2024Towards Accurate Post-Training Quantization of Vision Transformers via Error Reduction