mlsanityData to 14 Oct 2026

Pruning vs Quantization: Which is Better?

arXiv 2307.02973 ↗First posted 6 Jul 2023cs.LGcited by 46

Not on the map. Why?

Abstract (arXiv)

Neural network pruning and quantization techniques are almost as old as neural networks themselves. However, to date only ad-hoc comparisons between the two have been published. In this paper, we set out to answer the question on which is better: neural network quantization or pruning? By answering this question, we hope to inform design decisions made on neural network hardware going forward. We provide an extensive comparison between the two techniques for compressing deep neural networks. First, we give an analytical comparison of expected quantization and pruning error for general data distributions. Then, we provide lower bounds for the per-layer pruning and quantization error in trained networks, and compare these to empirical error after optimization. Finally, we provide an extensive experimental comparison for training 8 large-scale models on 3 tasks. Our results show that in most cases quantization outperforms pruning. Only in some scenarios with very high compression ratio, pruning might be beneficial from an accuracy standpoint.

Key facts. Cited by 46 papers in the map (data through 14 Oct 2026). Builds on 47 of its 68 references that are in the map. First posted on arXiv 6 Jul 2023.

Builds on

47 of 68 references.

References · 47

  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. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  5. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  6. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  7. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  8. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
  9. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
  10. 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,161
  11. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 2,031
  12. 2019Searching for MobileNetV3cited by 1,994
35 more
  1. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  2. 2023SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shotcited by 1,121
  3. 2019EfficientDet: Scalable and Efficient Object Detectioncited by 1,020
  4. 2018SNIP: Single-shot Network Pruning based on Connection Sensitivitycited by 812
  5. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 702
  6. 2019Learned Step Size Quantizationcited by 628
  7. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 621
  8. 2017To prune, or not to prune: exploring the efficacy of pruning for model compressioncited by 603
  9. 2017Learning Sparse Neural Networks through L0 Regularizationcited by 583
  10. 2020What is the State of Neural Network Pruning?cited by 583
  11. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 569
  12. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 479
  13. 2019The State of Sparsity in Deep Neural Networkscited by 476
  14. 2021A White Paper on Neural Network Quantizationcited by 472
  15. 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 467
  16. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 402
  17. 2015Deep Learning with Limited Numerical Precisioncited by 376
  18. 2019Data-Free Quantization Through Weight Equalization and Bias Correctioncited by 336
  19. 2022Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruningcited by 322
  20. 2020ZeroQ: A Novel Zero Shot Quantization Frameworkcited by 269
  21. 2017NISP: Pruning Networks using Neuron Importance Score Propagationcited by 228
  22. 2021Accelerating Sparse Deep Neural Networkscited by 223
  23. 2019Low-bit Quantization of Neural Networks for Efficient Inferencecited by 193
  24. 2020LSQ+: Improving low-bit quantization through learnable offsets and better initializationcited by 189
  25. 2020Characterising Bias in Compressed Modelscited by 167
  26. 2015Fixed Point Quantization of Deep Convolutional Networkscited by 143
  27. 2022Overcoming Oscillations in Quantization-Aware Trainingcited by 119
  28. 2015Accelerating Very Deep Convolutional Networks for Classification and Detectioncited by 109
  29. 2019A Signal Propagation Perspective for Pruning Neural Networks at Initializationcited by 96
  30. 2020Bayesian Bits: Unifying Quantization and Pruningcited by 94
  31. 2022FP8 Quantization: The Power of the Exponentcited by 88
  32. 2017Exploring Sparsity in Recurrent Neural Networkscited by 75
  33. 2022OPQ: Compressing Deep Neural Networks with One-shot Pruning-Quantizationcited by 19
  34. 2022Cyclical Pruning for Sparse Neural Networkscited by 6
  35. 2015A Semidefinite Programming Method for Integer Convex Quadratic Minimizationcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 18; 2025: 15; 2026: 12.

46 papers.

Sort:

By area:

  1. 2023On the Impact of Calibration Data in Post-training Quantization and Pruningcited by 58
  2. 2024MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantizationcited by 28
  3. 2024Can Large Language Models Understand Context?cited by 26
  4. 2024Effective Interplay between Sparsity and Quantization: From Theory to Practicecited by 19
  5. 2024Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMscited by 18
  6. 2024ESPACE: Dimensionality Reduction of Activations for Model Compressioncited by 15
  7. 2025SQAP-VLA: A Synergistic Quantization-Aware Pruning Framework for High-Performance Vision-Language-Action Modelscited by 15
  8. 2024Pruning By Explaining Revisited: Optimizing Attribution Methods to Prune CNNs and Transformerscited by 11
  9. 2024Efficient LLM Inference using Dynamic Input Pruning and Cache-Aware Maskingcited by 11
  10. 2025Surgical Scene Understanding in the Era of Foundation AI Models: A Comprehensive Reviewcited by 8
  11. 2024On the Limitations of Language Targeted Pruning: Investigating the Calibration Language Impact in Multilingual LLM Pruningcited by 6
  12. 2024MATCH: Model-Aware TVM-based Compilation for Heterogeneous Edge Devicescited by 5
34 more
  1. 2025From Tiny Machine Learning to Tiny Deep Learning: A Surveycited by 5
  2. 2024SpikeBottleNet: Spike-Driven Feature Compression Architecture for Edge-Cloud Co-Inferencecited by 3
  3. 2025RSQ: Learning from Important Tokens Leads to Better Quantized LLMscited by 3
  4. 2025ProfilingAgent: Profiling-Guided Agentic Reasoning for Adaptive Model Optimizationcited by 3
  5. 2026LoPRo: Enhancing Low-Rank Quantization via Permuted Block-Wise Rotationcited by 3
  6. 2024Leveraging language models for summarizing mental state examinations: A comprehensive evaluation and dataset releasecited by 2
  7. 2024How to Parameterize Asymmetric Quantization Ranges for Quantization-Aware Trainingcited by 2
  8. 2025Information-Bottleneck Driven Binary Neural Network for Change Detectioncited by 2
  9. 2025On the Interaction of Compressibility and Adversarial Robustnesscited by 2
  10. 2025Explaining How Quantization Disparately Skews a Modelcited by 2
  11. 2026FLRQ: Faster LLM Quantization with Flexible Low-Rank Matrix Sketchingcited by 2
  12. 2024An Empirical Study of the Influence of Adversarial Fine-Tuning on Compressed Neural Networkscited by 1
  13. 2024Predicting Probabilities of Error to Combine Quantization and Early Exiting: QuEEcited by 1
  14. 2024BitQ: Tailoring Block Floating Point Precision for Improved DNN Efficiency on Resource-Constrained Devicescited by 1
  15. 2024On the Impact of White-box Deployment Strategies for Edge AI on Latency and Model Performancecited by 1
  16. 2024Advanced Knowledge Transfer: Refined Feature Distillation for Zero-Shot Quantization in Edge Computingcited by 1
  17. 2025NoWag: A Unified Framework for Shape Preserving Compression of Large Language Modelscited by 1
  18. 2025Frustratingly Easy Task-aware Pruning for Large Language Modelscited by 1
  19. 2025Link-Aware Energy-Frugal Continual Learning for Fault Detection in IoT Networkscited by 1
  20. 2026PACE: Prune-And-Compress Ensemble Modelscited by 1
  21. 2026GPTQ-intrinsic LoRA: A Near-optimal Algorithm for Low-precision Quantization with Low-rank Adaptationcited by 1
  22. 2024Label Set Optimization via Activation Distribution Kurtosis for Zero-shot Classification with Generative Modelscited by 0
  23. 2025UPAQ: A Framework for Real-Time and Energy-Efficient 3D Object Detection in Autonomous Vehiclescited by 0
  24. 2025As easy as PIE: understanding when pruning causes language models to disagreecited by 0
  25. 2025Efficient Quantization-Aware Neural Receivers: Beyond Post-Training Quantizationcited by 0
  26. 2025DivQAT: Enhancing Robustness of Quantized Convolutional Neural Networks against Model Extraction Attackscited by 0
  27. 2026Fine-Grained Model Merging via Modular Expert Recombinationcited by 0
  28. 2026Linguistic properties and model scale in brain encoding: from small to compressed language modelscited by 0
  29. 2026Pruning Extensions and Efficiency Trade-Offs for Sustainable Time Series Classificationcited by 0
  30. 2026A Comparative Study of CNN Optimization Methods for Edge AI: Exploring the Role of Early Exitscited by 0
  31. 2026Fast Tensorization of Neural Networks via Slice-wise Feature Distillationcited by 0
  32. 2026Joint Structural Pruning and Mixed-Precision Quantization for LLM Compressioncited by 0
  33. 2026Harvesting AI Computation at the Edge via Generic Approximationcited by 0
  34. 2026Beyond Reconstruction Loss in Post-Training Quantization: Balanced Fitting for Large Vision-Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2021Pruning and Quantization for Deep Neural Network Acceleration: A Survey
  2. 2025Integrating Pruning with Quantization for Efficient Deep Neural Networks Compression
  3. 2021Training Deep Neural Networks with Joint Quantization and Pruning of Weights and Activations
  4. 2023A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and Recommendations
  5. 2025Pruning and Quantization Impact on Graph Neural Networks
  6. 2021Ps and Qs: Quantization-aware pruning for efficient low latency neural network inference
  7. 2025Theoretical Compression Bounds for Wide Multilayer Perceptrons
  8. 2025Automatic Joint Structured Pruning and Quantization for Efficient Neural Network Training and Compression