# Linear-Time Non-Local Codec Placement for Machine Learning Model Compression

arXiv: https://arxiv.org/abs/2610.14703 · HTML: https://mlsanity.com/paper/2610.14703/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 19 references that are in the map.
- First posted on arXiv 10 Oct 2026.

## Builds on (most cited first, 13 of 13)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://mlsanity.com/paper/1807.11164/) (2018)
- [QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs](https://mlsanity.com/paper/2404.00456/) (2024)
- [Learned Step Size Quantization](https://mlsanity.com/paper/1902.08153/) (2019)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weights](https://mlsanity.com/paper/1702.03044/) (2017)
- [Microscaling Data Formats for Deep Learning](https://mlsanity.com/paper/2310.10537/) (2023)
- [TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate](https://mlsanity.com/paper/2504.19874/) (2025)
- [TorchAO: PyTorch-Native Training-to-Serving Model Optimization](https://mlsanity.com/paper/2507.16099/) (2025)
- [ExecuTorch -- A Unified PyTorch Solution to Run AI Models On-Device](https://mlsanity.com/paper/2605.08195/) (2026)

## Abstract (arXiv)

On-device inference imposes constraints on model storage and operand traffic and thus benefits from compressed operand representations, which mandate the use of encoding and decoding operators. However, na"ive placement of codec operators can cancel the operand traffic gains sought from compression. In this work, I present an algorithm to place codec operators into a source computational graph to make it compatible with target hardware. The edges of the graph are colored so that each induced subgraph is weakly connected and has a maximal node that both postdominates all subgraph inputs and dominates all subgraph outputs; this unique node - dubbed the textitabsolute dominator - is the natural location to observe the traffic flowing through the subgraph and learn parametric codec operators. The algorithm's time complexity is linear in the source graph's size. With respect to na"ive local codec placement, the algorithm reduces modeled operand traffic by 24%-44% across four models, including 24% on SigLIP and 34% on SmolLM2.