mlsanityData to 14 Oct 2026

Loss Aware Post-training Quantization

arXiv 1911.07190 ↗First posted 17 Nov 2019cs.LG cs.CVcited by 97

Not on the map. Why?

Abstract (arXiv)

Neural network quantization enables the deployment of large models on resource-constrained devices. Current post-training quantization methods fall short in terms of accuracy for INT4 (or lower) but provide reasonable accuracy for INT8 (or above). In this work, we study the effect of quantization on the structure of the loss landscape. Additionally, we show that the structure is flat and separable for mild quantization, enabling straightforward post-training quantization methods to achieve good results. We show that with more aggressive quantization, the loss landscape becomes highly non-separable with steep curvature, making the selection of quantization parameters more challenging. Armed with this understanding, we design a method that quantizes the layer parameters jointly, enabling significant accuracy improvement over current post-training quantization methods. Reference implementation is available at https://github.com/ynahshan/nn-quantization-pytorch/tree/master/lapq

Key facts. Cited by 97 papers in the map (data through 14 Oct 2026). First posted on arXiv 17 Nov 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 13; 2022: 24; 2023: 18; 2024: 22; 2025: 11; 2026: 9.

97 papers.

Sort:

By area:

  1. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  2. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 479
  3. 2022Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruningcited by 322
  4. 2022Post-training Quantization on Diffusion Modelscited by 267
  5. 2022Outlier Suppression: Pushing the Limit of Low-bit Transformer Language Modelscited by 181
  6. 2022QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantizationcited by 180
  7. 2022Overcoming Oscillations in Quantization-Aware Trainingcited by 119
  8. 2023OWQ: Outlier-Aware Weight Quantization for Efficient Fine-Tuning and Inference of Large Language Modelscited by 114
  9. 2021Accelerated Sparse Neural Training: A Provable and Efficient Method to Find N:M Transposable Maskscited by 103
  10. 2022PD-Quant: Post-Training Quantization based on Prediction Difference Metriccited by 73
  11. 2024QuEST: Low-bit Diffusion Model Quantization via Efficient Selective Finetuningcited by 49
  12. 2023FlexRound: Learnable Rounding based on Element-wise Division for Post-Training Quantizationcited by 48
38 more
  1. 2023Towards Accurate Post-training Quantization for Diffusion Modelscited by 47
  2. 2024Q-VLM: Post-training Quantization for Large Vision-Language Modelscited by 47
  3. 2021Towards Efficient Post-training Quantization of Pre-trained Language Modelscited by 39
  4. 2021Auto-Split: A General Framework of Collaborative Edge-Cloud AIcited by 30
  5. 2021Accurate Neural Training with 4-bit Matrix Multiplications at Standard Formatscited by 26
  6. 2024Instance-Aware Group Quantization for Vision Transformerscited by 25
  7. 2022It's All In the Teacher: Zero-Shot Quantization Brought Closer to the Teachercited by 20
  8. 2023Jumping through Local Minima: Quantization in the Loss Landscape of Vision Transformerscited by 18
  9. 2023Solving Oscillation Problem in Post-Training Quantization Through a Theoretical Perspectivecited by 17
  10. 2024On Efficient Variants of Segment Anything Model: A Surveycited by 16
  11. 2022The Effect of Model Compression on Fairness in Facial Expression Recognitioncited by 13
  12. 2022Quantization Robust Federated Learning for Efficient Inference on Heterogeneous Devicescited by 12
  13. 2023Poster: Self-Supervised Quantization-Aware Knowledge Distillationcited by 11
  14. 2025QuantV2X: A Fully Quantized Multi-Agent System for Cooperative Perceptioncited by 11
  15. 2021HPTQ: Hardware-Friendly Post Training Quantizationcited by 10
  16. 2022Rate-Distortion Optimized Post-Training Quantization for Learned Image Compressioncited by 10
  17. 2023EQ-Net: Elastic Quantization Neural Networkscited by 10
  18. 2024IntLoRA: Integral Low-rank Adaptation of Quantized Diffusion Modelscited by 10
  19. 2021Confounding Tradeoffs for Neural Network Quantizationcited by 9
  20. 2021Fine-grained Data Distribution Alignment for Post-Training Quantizationcited by 9
  21. 2022Minimum Variance Unbiased N:M Sparsity for the Neural Gradientscited by 9
  22. 2023GenQ: Quantization in Low Data Regimes with Generative Synthetic Datacited by 8
  23. 2025Put Teacher in Student's Shoes: Cross-Distillation for Ultra-compact Model Compression Frameworkcited by 8
  24. 2022SQuAT: Sharpness- and Quantization-Aware Training for BERTcited by 7
  25. 2024QEFT: Quantization for Efficient Fine-Tuning of LLMscited by 7
  26. 2023EPTQ: Enhanced Post-Training Quantization via Hessian-guided Network-wise Optimizationcited by 6
  27. 2024Improving Quantization-aware Training of Low-Precision Network via Block Replacement on Full-Precision Counterpartcited by 6
  28. 2025FP4DiT: Towards Effective Floating Point Quantization for Diffusion Transformerscited by 6
  29. 2024MetaAug: Meta-Data Augmentation for Post-Training Quantizationcited by 5
  30. 2024Panacea: Novel DNN Accelerator using Accuracy-Preserving Asymmetric Quantization and Energy-Saving Bit-Slice Sparsitycited by 5
  31. 2025Task Vector Quantization for Memory-Efficient Model Mergingcited by 5
  32. 2025Radio: Rate-Distortion Optimization for Large Language Model Compressioncited by 5
  33. 2025Sharpness-Aware Data Generation for Zero-shot Quantizationcited by 5
  34. 2022Sub-8-bit quantization for on-device speech recognition: a regularization-free approachcited by 4
  35. 2022Vertical Layering of Quantized Neural Networks for Heterogeneous Inferencecited by 4
  36. 2023MixQuant: Mixed Precision Quantization with a Bit-width Optimization Searchcited by 4
  37. 2024decoupleQ: Towards 2-bit Post-Training Uniform Quantization via decoupling Parameters into Integer and Floating Pointscited by 4
  38. 2024PTQ4RIS: Post-Training Quantization for Referring Image Segmentationcited by 4

50 most cited of 97 All 97 (JSON)