paper-with-me

Papers

Calibration and Transformation-Free Weight-Only LLMs Quantization via Dynamic Grouping

2025-09-03 · Xinzhe Zheng, Zhen-Qun Yang, Zishan Liu, Haoran Xie, S. Joe Qin, Arlene Chen, Fangzhen Lin arxiv

Large Language Models (LLMs) deliver strong performance but are difficult to deploy under tight memory and compute constraints. Low-bit post-training quantization (PTQ) is a promising direction; however, it typically relies on calibration data, auxiliary transformations, and GPU tools. To address these limitations, we propose MSB (Multi Scale Binary), a calibration-free and transformation-free PTQ method that generalizes binary quantization to multi-bit settings. MSB optimizes a dynamic grouping criterion that minimizes within group variance, yielding group-wise multiscale levels that can be applied consistently across granularities from per tensor to block-wise configurations with 64 elements groups per row, without calibration or intermediate transforms. We implement the optimization in a CPU based solver for the quantization step and evaluate using standard bfloat16 execution without low-bit packing. On Llama 3.2 3B, MSB achieves 8.43 perplexity on WikiText-2 under 4-bit weight only block-wise quantization, compared to 7.81 in full precision and 12.23 with GPTQ its default setup. Overall, MSB provides a new optimization perspective for low-bit PTQ while simplifying the pipeline by removing calibration and transformations.

📄 PDF Abstract BibTeX arXiv:2509.03054

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

ACE: Adaptive Calibration-Free Expert Skipping for MoE-based LLMs

2026-09-04 · Zukang Xu, Zhixiong Zhao, Xing Hu, Jiangyong Yu 외 arxiv

Mixture-of-Experts (MoE) architectures provide an efficient paradigm for scaling large language models (LLMs), yet fixed top-k routing activates the same number of expert slots for every token, causing substantial redund…

CafeQ: Calibration-free Quantization via Learned Transformations and Adaptive Rounding

2025-11-24 · Ziteng Sun, Adrian Benton, Samuel Kushnir, Asher Trockman 외 arxiv

Post-training quantization is an effective method for reducing the serving cost of large language models, where the standard approach is to use a round-to-nearest quantization level scheme. However, this often introduces…

FreeAct: Freeing Activations for LLM Quantization

2026-03-02 · Xiaohao Liu, Xiaobo Xia, Manyi Zhang, Ji-Fu Li 외 arxiv

Quantization is pivotal for mitigating the significant memory and computational overhead of Large Language Models (LLMs). While emerging transformation-based methods have successfully enhanced quantization by projecting …

FLAT-LLM: Fine-grained Low-rank Activation Space Transformation for Large Language Model Compression

2025-05-29 · Jiayi Tian, Ryan Solgi, Jinming Lu, Yifan Yang 외

Large Language Models (LLMs) have enabled remarkable progress in natural language processing, yet their high computational and memory demands pose challenges for deployment in resource-constrained environments. Although …

Language ModelingLanguage ModellingLarge Language ModelModel Compression

Boundary-weighted logit consistency improves calibration of segmentation networks

2023-07-16 · Neerav Karani, Neel Dey, Polina Golland

Neural network prediction probabilities and accuracy are often only weakly-correlated. Inherent label ambiguity in training data for image segmentation aggravates such miscalibration. We show that logit consistency acros…

Image SegmentationMRI segmentationSegmentationSemantic Segmentation