mlsanityData to 14 Oct 2026

Cat: Post-Training Quantization Error Reduction via Cluster-based Affine Transformation

arXiv 2509.26277 ↗First posted 30 Sep 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Post-Training Quantization (PTQ) reduces the memory footprint and computational overhead of deep neural networks by converting full-precision (FP) values into quantized and compressed data types. While PTQ is more cost-efficient than Quantization-Aware Training (QAT), it is highly susceptible to accuracy degradation under a low-bit quantization (LQ) regime (e.g., 2-bit). Affine transformation is a classical technique used to reduce the discrepancy between the information processed by a quantized model and that processed by its full-precision counterpart; however, we find that using plain affine transformation, which applies a uniform affine parameter set for all outputs, worsens the results in low-bit PTQ. To address this, we propose Cluster-based Affine Transformation (CAT), an error-reduction framework that employs cluster-specific parameters to align LQ outputs with FP counterparts. CAT refines LQ outputs with only a negligible number of additional parameters, without requiring fine-tuning of the model or quantization parameters. We further introduce a novel PTQ framework integrated with CAT. Experiments on ImageNet-1K show that this framework consistently outperforms prior PTQ methods across diverse architectures and LQ settings, achieving up to 53.18% Top-1 accuracy on W2A2 ResNet-18. Moreover, CAT enhances existing PTQ baselines by more than 3% when used as a plug-in. We plan to release our implementation alongside the publication of this paper.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 40 of its 55 references that are in the map. First posted on arXiv 30 Sep 2025.

Builds on

40 of 55 references.

References · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  3. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  4. 2024SAM 2: Segment Anything in Images and Videoscited by 2,863
  5. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  6. 2020Designing Network Design Spacescited by 960
  7. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 943
  8. 2019Learned Step Size Quantizationcited by 628
  9. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 621
  10. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 443
  11. 2016Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activationscited by 426
  12. 2019Q8BERT: Quantized 8Bit BERTcited by 300
28 more
  1. 2018LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networkscited by 274
  2. 2020ZeroQ: A Novel Zero Shot Quantization Frameworkcited by 269
  3. 2023LoftQ: LoRA-Fine-Tuning-Aware Quantization for Large Language Modelscited by 196
  4. 2022QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantizationcited by 180
  5. 2021PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantizationcited by 179
  6. 2021FQ-ViT: Post-Training Quantization for Fully Quantized Vision Transformercited by 151
  7. 2022RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformerscited by 137
  8. 2024EfficientQAT: Efficient Quantization-Aware Training for Large Language Modelscited by 127
  9. 2019Loss Aware Post-training Quantizationcited by 97
  10. 2020Degree-Quant: Quantization-Aware Training for Graph Neural Networkscited by 96
  11. 2024Q-DiT: Accurate Post-Training Quantization for Diffusion Transformerscited by 94
  12. 2023EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Modelscited by 78
  13. 2022PD-Quant: Post-Training Quantization based on Prediction Difference Metriccited by 73
  14. 2023Towards Accurate Post-Training Quantization for Vision Transformercited by 71
  15. 2024AffineQuant: Affine Transformation Quantization for Large Language Modelscited by 67
  16. 2019QKD: Quantization-aware Knowledge Distillationcited by 44
  17. 2024Boosting Vision-Language Models with Transductioncited by 36
  18. 2024ResQ: Mixed-Precision Quantization of Large Language Models with Low-Rank Residualscited by 36
  19. 2023CBQ: Cross-Block Quantization for Large Language Modelscited by 34
  20. 2024Towards Accurate Post-Training Quantization of Vision Transformers via Error Reductioncited by 23
  21. 2025Retrieval-Augmented Perception: High-Resolution Image Perception Meets Visual RAGcited by 23
  22. 2024Effective Interplay between Sparsity and Quantization: From Theory to Practicecited by 19
  23. 2025APHQ-ViT: Post-Training Quantization with Average Perturbation Hessian Based Reconstruction for Vision Transformerscited by 19
  24. 2024PassionSR: Post-Training Quantization with Adaptive Scale in One-Step Diffusion based Image Super-Resolutioncited by 9
  25. 2025QuartDepth: Post-Training Quantization for Real-Time Depth Estimation on the Edgecited by 8
  26. 2023Q-HyViT: Post-Training Quantization of Hybrid Vision Transformers with Bridge Block Reconstruction for IoT Systemscited by 6
  27. 2025RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Modelscited by 5
  28. 2025On Quantizing Neural Representation for Variable-Rate Video Codingcited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Sensitivity-Aware Post-Training Quantization for Deep Neural Networks
  2. 2023Benchmarking the Reliability of Post-training Quantization: a Particular Focus on Worst-case Performance
  3. 2024COMQ: A Backpropagation-Free Algorithm for Post-Training Quantization
  4. 2021Post-Training Sparsity-Aware Quantization
  5. 2026Efficient Tuning Before Low-Bit Post-Training Quantization for Stochastic Gradient Descent-optimized Models
  6. 2021A White Paper on Neural Network Quantization
  7. 2025Provable Post-Training Quantization: Theoretical Analysis of OPTQ and Qronos
  8. 2022QFT: Post-training quantization via fast joint finetuning of all degrees of freedom