把LLM剪枝难题化为背包问题，SNIPER统一深度+宽度剪枝，精准控预算！
arXiv:2608.12953v1 Announce Type: cross Abstract: Structured pruning is a promising approach for compressing large language models (LLMs), yet existin…
深入剖析免训练低秩压缩中校准与截断误差的传播机制，为LLM高效压缩提供理论支撑。
arXiv:2608.08506v1 Announce Type: new Abstract: Training-free low-rank compression frameworks have been gaining prominence for LLM compression given t…
大模型剪枝新招：高层注意力剪枝加缩放，推理提速不损精度，论文原文速看。
arXiv:2507.01900v3 Announce Type: replace-cross Abstract: Pruning is a highly effective approach for compressing large language models (LLMs), signifi…
新型量化方法ARCQuant通过增强残差通道，显著提升NVFP4在大语言模型上的精度表现，为低比特量化开辟新路径。
arXiv:2601.07475v2 Announce Type: replace-cross Abstract: The emergence of fine-grained numerical formats like NVFP4 presents new opportunities for ef…
校准引导的LLM压缩竟有输出空间分配开销？这篇实证研究用数据揭示量化压缩的隐性成本，值得技术玩家细读。
arXiv:2606.27785v1 Announce Type: cross Abstract: Training-free compression methods for large language models (LLMs) often use calibration data to gui…
被ICML 2026接收为oral，提出一种兼顾成本与精度的三元量化新方法，为大模型高效部署提供突破。
arXiv:2606.26650v1 Announce Type: cross Abstract: In this paper, we present CAT-Q, Cost-efficient and Accurate Ternary Quantization, for compressing a…
提出统一秩分配方法，突破低秩分解压缩LLM的瓶颈，兼顾效率与性能。
arXiv:2606.21847v1 Announce Type: cross Abstract: Low-rank decomposition serves as a promising compression paradigm for large language models, however…
新型SVD压缩框架，利用反向信号影响度量引导低秩近似，在保持大模型功能的同时实现高效压缩。
arXiv:2606.19993v1 Announce Type: new Abstract: We present Activation- and Influence-Aware Ranks (AIR), an SVD-based LLM compression framework that gu…
LLM压缩新突破：将块删除问题转化为约束二元优化，物理模型Ising玻璃降低下游性能损失
arXiv:2602.00161v2 Announce Type: replace Abstract: In this paper, we formulate the compression of large language models (LLMs) by optimally deleting …
仅用10位比特完成Haiku到Opus量级压缩？这篇论文提出基于“二十问”的交互式有损压缩协议，小模型通过提问逐比特获取大模型知识，实现极高压缩率
arXiv:2604.02343v2 Announce Type: replace-cross Abstract: We study the compression of LLM-generated text across lossless and lossy regimes, characteri…
突破性LLM压缩至低于1比特，结合草图算法与硬件友好算子，实现极致模型瘦身。
arXiv:2506.17255v2 Announce Type: replace-cross Abstract: Large language models (LLMs) require larger GPU memory size these days, necessitating effici…
新方法Swift-SVD实现理论最优性与实际效率兼得，专为低秩大模型压缩而生，ICML 2026收录。
arXiv:2604.01609v2 Announce Type: replace Abstract: The deployment of Large Language Models is constrained by the memory and bandwidth demands of stat…
联合结构化剪枝与混合精度量化，为LLM压缩提供高效新方案，兼顾精度与速度。
arXiv:2606.07819v1 Announce Type: cross Abstract: Recently, the efficiency of Large Language Models (LLMs) deployment has become a critical concern in…
用SVD低秩分解加学习缩放矩阵，实现高效LLM压缩，精准降维保性能。
arXiv:2606.07098v1 Announce Type: cross Abstract: We present SigmaScale, a method for learning auxiliary scaling matrices $S$ to aid truncated Singula…
联合优化架构与量化，突破LLM压缩效率瓶颈
arXiv:2606.04063v1 Announce Type: cross Abstract: Deploying large language models (LLMs) is challenging due to their significant memory and computatio…
重新审视张量分解在LLM压缩中的核心作用，为后训练压缩提供新思路。
arXiv:2606.03465v1 Announce Type: cross Abstract: Post-training compression is essential for deploying large language models (LLMs) under tight resour…
压缩LLM时，准确率不是唯一指标——新基准用保形概率评估不确定性保留。
arXiv:2606.01850v1 Announce Type: new Abstract: Model compression techniques such as quantization and pruning are widely used to reduce the deployment…
新方法ProjQ让大模型压缩更聪明：后训练量化与低秩适配协同去噪，缓解顺序部署的精度损失。
arXiv:2606.00494v1 Announce Type: new Abstract: Post-Training Quantization (PTQ) and Low-Rank Adaptation (LoRA) constitute the standard pipeline for e…
重新定义替换式LLM压缩的粒度，从层到子模块，为模型轻量化提供新视角。
arXiv:2606.02559v1 Announce Type: cross Abstract: Post-training compression of Large Language Models (LLMs) removes entire architectural components, e…
跨层子空间耦合统一框架，揭示大模型SVD压缩方法的新极限与潜力
arXiv:2605.30836v1 Announce Type: new Abstract: Recent SVD based compression methods for large language models like SVD LLM and Basis Sharing can be u…