paper-with-me

홈 › Papers

Sherry: Hardware-Efficient 1.25-Bit Ternary Quantization via Fine-grained Sparsification

2026-01-12 · Hong Huang, Decheng Wu, Qiangqiang Hu, Guanghua Yu, Jinhai Yang, Jianchen Zhu, Xue Liu, Dapeng Wu arxiv

The deployment of Large Language Models (LLMs) on resource-constrained edge devices is increasingly hindered by prohibitive memory and computational requirements. While ternary quantization offers a compelling solution by reducing weights to {-1, 0, +1}, current implementations suffer from a fundamental misalignment with commodity hardware. Most existing methods must choose between 2-bit aligned packing, which incurs significant bit wastage, or 1.67-bit irregular packing, which degrades inference speed. To resolve this tension, we propose Sherry, a hardware-efficient ternary quantization framework. Sherry introduces a 3:4 fine-grained sparsity that achieves a regularized 1.25-bit width by packing blocks of four weights into five bits, restoring power-of-two alignment. Furthermore, we identify weight trapping issue in sparse ternary training, which leads to representational collapse. To address this, Sherry introduces Arenas, an annealing residual synapse mechanism that maintains representational diversity during training. Empirical evaluations on LLaMA-3.2 across five benchmarks demonstrate that Sherry matches state-of-the-art ternary performance while significantly reducing model size. Notably, on an Intel i7-14700HX CPU, our 1B model achieves zero accuracy loss compared to SOTA baselines while providing 25% bit savings and 10% speed up. The code is available at https://github.com/Tencent/AngelSlim .

📄 PDF Abstract BibTeX arXiv:2601.07892

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Ternary MobileNets via Per-Layer Hybrid Filter Banks

2019-11-04 · Dibakar Gope, Jesse Beu, Urmish Thakker, Matthew Mattina

MobileNets family of computer vision neural networks have fueled tremendous progress in the design and organization of resource-efficient architectures in recent years. New applications with stringent real-time requireme…

Quantization

ParetoQ: Scaling Laws in Extremely Low-bit LLM Quantization

2025-02-04 · Zechun Liu, Changsheng Zhao, Hanxian Huang, Sijia Chen 외

The optimal bit-width for achieving the best trade-off between quantized model size and accuracy has been a subject of ongoing debate. While some advocate for 4-bit quantization, others propose that 1.58-bit offers super…

Quantization

SYQ: Learning Symmetric Quantization For Efficient Deep Neural Networks

2018-07-01 · CVPR 2018 6 · Julian Faraone, Nicholas Fraser, Michaela Blott, Philip H. W. Leong

Inference for state-of-the-art deep neural networks is computationally expensive, making them difficult to deploy on constrained hardware environments. An efficient way to reduce this complexity is to quantize the weight…

Quantization

The Binary and Ternary Quantization Can Improve Feature Discrimination

2025-04-18 · Weizhi Lu, Mingrui Chen, Weiyu Li

In machine learning, quantization is widely used to simplify data representation and facilitate algorithm deployment on hardware. Given the fundamental role of classification in machine learning, it is crucial to investi…

ClassificationQuantization

Compressing Low Precision Deep Neural Networks Using Sparsity-Induced Regularization in Ternary Networks

2017-09-19 · Julian Faraone, Nicholas Fraser, Giulio Gambardella, Michaela Blott 외

A low precision deep neural network training technique for producing sparse, ternary neural networks is presented. The technique incorporates hard- ware implementation costs during training to achieve significant model c…

L2 RegularizationModel CompressionQuantization