Data to 7 Oct 2026

Improving Neural Network Quantization without Retraining using Outlier Channel Splitting

arXiv 1901.09504 ↗First posted 28 Jan 2019cs.LG stat.MLcited by 150

Not on the map. Why?

Abstract (arXiv)

Quantization can improve the execution latency and energy efficiency of neural networks on both commodity GPUs and specialized accelerators. The majority of existing literature focuses on training quantized DNNs, while this work examines the less-studied topic of quantizing a floating-point model without (re)training. DNN weights and activations follow a bell-shaped distribution post-training, while practical hardware uses a linear quantization grid. This leads to challenges in dealing with outliers in the distribution. Prior work has addressed this by clipping the outliers or using specialized hardware. In this work, we propose outlier channel splitting (OCS), which duplicates channels containing outliers, then halves the channel values. The network remains functionally identical, but affected outliers are moved toward the center of the distribution. OCS requires no additional training and works on commodity hardware. Experimental evaluation on ImageNet classification and language modeling shows that OCS can outperform state-of-the-art clipping techniques with only minor overhead.

Key facts. Cited by 150 papers in the map (data through 7 Oct 2026). First posted on arXiv 28 Jan 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 42; 2022: 27; 2023: 35; 2024: 20; 2025: 18; 2026: 8.

150 papers.

Sort:

By area:

2022SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Modelscited by 1,338
2023H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Modelscited by 805
2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
2023SqueezeLLM: Dense-and-Sparse Quantizationcited by 285
2021Post-Training Quantization for Vision Transformercited by 274
2023Q-Diffusion: Quantizing Diffusion Modelscited by 266
2023Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Timecited by 247
2022A Fast Post-Training Pruning Framework for Transformerscited by 161
2023Quantizable Transformers: Removing Outliers by Helping Attention Heads Do Nothingcited by 147
2023RPTQ: Reorder-based Post-training Quantization for Large Language Modelscited by 112
2024Sequoia: Scalable, Robust, and Hardware-aware Speculative Decodingcited by 103
2022LUT-GEMM: Quantized Matrix Multiplication based on LUTs for Efficient Inference in Large-Scale Generative Language Modelscited by 98

38 more

2022PD-Quant: Post-Training Quantization based on Prediction Difference Metriccited by 73
2021Diversifying Sample Generation for Accurate Data-Free Quantizationcited by 63
2023QLLM: Accurate and Efficient Low-Bitwidth Quantization for Large Language Modelscited by 63
2023Outlier Suppression+: Accurate quantization of large language models by equivalent and optimal shifting and scalingcited by 62
2021VS-Quant: Per-vector Scaled Quantization for Accurate Low-Precision Neural Network Inferencecited by 59
2023Training Transformers with 4-bit Integerscited by 58
2025GPTAQ: Efficient Finetuning-Free Quantization for Asymmetric Calibrationcited by 54
2022SQuant: On-the-Fly Data-Free Quantization via Diagonal Hessian Approximationcited by 51
2023ZeRO++: Extremely Efficient Collective Communication for Giant Model Trainingcited by 50
2024Q-VLM: Post-training Quantization for Large Vision-Language Modelscited by 47
2023Towards Accurate Post-training Quantization for Diffusion Modelscited by 46
2023FlexRound: Learnable Rounding based on Element-wise Division for Post-Training Quantizationcited by 46
2021Zero-shot Adversarial Quantizationcited by 44
2024ProSparse: Introducing and Enhancing Intrinsic Activation Sparsity within Large Language Modelscited by 43
2024Low-Rank Quantization-Aware Training for LLMscited by 42
2021Qimera: Data-free Quantization with Synthetic Boundary Supporting Samplescited by 39
2021MQBench: Towards Reproducible and Deployable Model Quantization Benchmarkcited by 38
2021Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AIcited by 38
2021Towards Efficient Post-training Quantization of Pre-trained Language Modelscited by 37
2022Q-ViT: Fully Differentiable Quantization for Vision Transformercited by 36
2021Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safetycited by 35
2023CBQ: Cross-Block Quantization for Large Language Modelscited by 34
2021Applications and Techniques for Fast Machine Learning in Sciencecited by 32
2022Post-training Quantization for Neural Networks with Provable Guaranteescited by 32
2021Ps and Qs: Quantization-aware pruning for efficient low latency neural network inferencecited by 31
2021Auto-Split: A General Framework of Collaborative Edge-Cloud AIcited by 29
2023Efficient Post-training Quantization with FP8 Formatscited by 29
2021Pareto-Optimal Quantized ResNet Is Mostly 4-bitcited by 26
2022AlphaTuning: Quantization-Aware Parameter-Efficient Adaptation of Large-Scale Pre-Trained Language Modelscited by 22
2022It's All In the Teacher: Zero-Shot Quantization Brought Closer to the Teachercited by 20
2023PowerQuant: Automorphism Search for Non-Uniform Quantizationcited by 20
2021RED : Looking for Redundancies for Data-Free Structured Compression of Deep Neural Networkscited by 19
2021RED++ : Data-Free Pruning of Deep Neural Networks via Input Splitting and Output Mergingcited by 19
2024Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMscited by 18
2021Lightweight compression of neural network feature tensors for collaborative intelligencecited by 17
2021Qu-ANTI-zation: Exploiting Quantization Artifacts for Achieving Adversarial Outcomescited by 16
2025FPTQuant: Function-Preserving Transforms for LLM Quantizationcited by 16
2021Generative Zero-shot Network Quantizationcited by 15

50 most cited of 150 All 150 (JSON)