mlsanityData to 14 Oct 2026

Pack-PTQ: Advancing Post-training Quantization of Neural Networks by Pack-wise Reconstruction

arXiv 2505.00259 ↗First posted 1 May 2025cs.CV cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Post-training quantization (PTQ) has evolved as a prominent solution for compressing complex models, which advocates a small calibration dataset and avoids end-to-end retraining. However, most existing PTQ methods employ block-wise reconstruction, which neglects cross-block dependency and exhibits a notable accuracy drop in low-bit cases. To address these limitations, this paper presents a novel PTQ method, dubbed Pack-PTQ. First, we design a Hessian-guided adaptive packing mechanism to partition blocks into non-overlapping packs, which serve as the base unit for reconstruction, thereby preserving the cross-block dependency and enabling accurate quantization parameters estimation. Second, based on the pack configuration, we propose a mixed-precision quantization approach to assign varied bit-widths to packs according to their distinct sensitivities, thereby further enhancing performance. Extensive experiments on 2D image and 3D point cloud classification tasks, using various network architectures, demonstrate the superiority of our method over the state-of-the-art PTQ methods.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 36 of its 43 references that are in the map. First posted on arXiv 1 May 2025.

Builds on

36 of 43 references.

References · 36

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  4. 2017Mask R-CNNcited by 7,277
  5. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  6. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  7. 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
  8. 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,940
  9. 2020Knowledge Distillation: A Surveycited by 1,603
  10. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  11. 2020Designing Network Design Spacescited by 960
  12. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 943
24 more
  1. 2021Segmenter: Transformer for Semantic Segmentationcited by 855
  2. 2019Learned Step Size Quantizationcited by 628
  3. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 621
  4. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 479
  5. 2021A White Paper on Neural Network Quantizationcited by 472
  6. 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 467
  7. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 402
  8. 2019Data-Free Quantization Through Weight Equalization and Bias Correctioncited by 336
  9. 2019HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networkscited by 261
  10. 2023A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and Recommendationscited by 202
  11. 2022QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantizationcited by 180
  12. 2021PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantizationcited by 179
  13. 2022RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformerscited by 137
  14. 2022Overcoming Oscillations in Quantization-Aware Trainingcited by 119
  15. 2022PD-Quant: Post-Training Quantization based on Prediction Difference Metriccited by 73
  16. 2023Towards Accurate Post-Training Quantization for Vision Transformercited by 71
  17. 2021Towards Mixed-Precision Quantization of Neural Networks via Constrained Optimizationcited by 61
  18. 2022NoisyQuant: Noisy Bias-Enhanced Post-Training Activation Quantization for Vision Transformerscited by 61
  19. 2024APTQ: Attention-aware Post-Training Mixed-Precision Quantization for Large Language Modelscited by 34
  20. 2024AdaLog: Post-Training Quantization for Vision Transformers with Adaptive Logarithm Quantizercited by 29
  21. 2023I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs Quantizationcited by 19
  22. 2022Genie: Show Me the Data for Quantizationcited by 17
  23. 2023Solving Oscillation Problem in Post-Training Quantization Through a Theoretical Perspectivecited by 17
  24. 2022RAPQ: Rescuing Accuracy for Power-of-Two Low-bit Post-training Quantizationcited by 15

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026ScalePredictor: Instance-aware Scale Learning for Accurate Quantization of Vision Transformerscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024COMQ: A Backpropagation-Free Algorithm for Post-Training Quantization
  2. 2023Benchmarking the Reliability of Post-training Quantization: a Particular Focus on Worst-case Performance
  3. 2026The Devil Is in the Reconstruction Loss Scale: Rethinking Optimization in LLM Quantization
  4. 2026Understanding Quantization-Aware Training: Gradients at Quantized Weights Bias to the Low-Loss Basin
  5. 2025Beacon: Post-Training Quantization with Integrated Grid Selection
  6. 2025APHQ-ViT: Post-Training Quantization with Average Perturbation Hessian Based Reconstruction for Vision Transformers
  7. 2023Augmenting Hessians with Inter-Layer Dependencies for Mixed-Precision Post-Training Quantization
  8. 2026Scale Sensitivity in Low-Bit Post-Training Quantization: Curvature of the Quantization Error Landscape