SlimPer: Make Personalization Model Slim and Smart
SlimPer提出让个性化模型精简且智能的新方法,兼顾性能与效率,值得关注。
arXiv:2607.12281v1 Announce Type: cross Abstract: Transformer-style architectures are increasingly adopted for industrial recommendation systems, yet …
SlimPer提出让个性化模型精简且智能的新方法,兼顾性能与效率,值得关注。
arXiv:2607.12281v1 Announce Type: cross Abstract: Transformer-style architectures are increasingly adopted for industrial recommendation systems, yet …
开源AI网关OmniRoute,通过堆叠令牌压缩技术节省高达95%的token成本,支持250+提供商(含90+免费)。
Optimizing LLM Costs: Meet OmniRoute AI coding agents are boosting developer productivity, but they consume a massive amount of tokens. For developers…
新方法STEC通过证据压缩提升开放式多跳问答的深度搜索效率与准确性
arXiv:2607.10795v1 Announce Type: new Abstract: In open-domain multi-hop question answering (QA), LLM-based search agents offer a promising approach t…
TCL 2026 年新品 —— 真省电 Pro 二代采用“ 双排铜管冷凝器 + 直流变频 + 电子膨胀阀 ”,1.5 匹 APF 5.60(竞品日常 5.27 左右)。 1.5 匹挂机上市价 2499 元,适合卧室 / 小客厅,除了上述配置外,还升级为 美芝压缩机 。 领 100 元券: 点此领券 …
LLM代理爬虫带来新威胁,这篇论文提出压缩感知的内容保护方法,对抗恶意爬取。
arXiv:2607.08180v1 Announce Type: cross Abstract: The rise of LLM-based agents with reasoning, summarization, and memory capabilities has created a ne…
真实数据揭示LLM编码代理90%时间浪费在重读上下文,手把手教你控制token成本
I have been using LLMs & Coding Agent since early 2024. A large problem with Coding Agents & LLMs in general is context compression. To give you some …
IT之家 7 月 11 日消息,AMD 在 2026 年度 I3D 研讨会上公布了一项名为“PEPS:位置编码投影采样”的最新研究成果。 这项技术聚焦于近年来备受关注的神经纹理压缩领域,在保持画质基本不变的前提下,成功将模型参数量削减了 25%。 神经纹理压缩的核心在于训练一种叫做“隐式神经表示”的…
最新论文提出LUMI,基于LLM的无损图像压缩,无需特定tokenizer,开辟AI压缩新路径。
arXiv:2607.08221v1 Announce Type: new Abstract: Large language model (LLM)-based lossless image compression methods typically represent pixel data thr…
用Kronecker分解海森矩阵,高效提升LLM量化精度,突破后训练量化瓶颈。
arXiv:2607.07964v1 Announce Type: new Abstract: Post-training quantization (PTQ) is a widely adopted technique for compressing large language models (…
IT之家 7 月 10 日消息,科技媒体 The Information 昨日(7 月 9 日)发布博文,报道称苹果公司正接洽 PrismML 初创公司, 评估在 iPhone 上直接运行更大规模 AI 模型的可行性。 IT之家查询公开资料,PrismML 是加州理工衍生出来的 AI 初创公司,核心…
突破传统一刀切剪枝,PALS按每层百分位自适应稀疏度,精准提升大模型压缩效率。
arXiv:2607.07557v1 Announce Type: cross Abstract: One-shot pruning methods like Wanda and SparseGPT apply the same sparsity ratio to every layer of a …
提出GIFT方法,利用梯度几何信息实现低精度通信,在不牺牲模型精度的前提下显著降低LLM预训练通信开销。
arXiv:2607.07494v1 Announce Type: cross Abstract: Gradient communication is a primary scaling bottleneck in large language model (LLM) pretraining. Co…
70多位作者联合发布Nemotron混合MoE大模型压缩方案,大幅降低推理成本且保持性能。
arXiv:2607.04371v1 Announce Type: new Abstract: We present Nemotron-Labs-3-Puzzle-75B-A9B, a compressed variant of Nemotron-3-Super optimized for inte…
新方法Fair-GPTQ实现大模型量化时兼顾效率与公平性,减少偏差
arXiv:2509.15206v3 Announce Type: replace Abstract: The high memory demands of generative language models have drawn attention to quantization, which …
针对多模态大语言模型压缩难题,提出Token级响应-视觉注意力引导,提升蒸馏效果
arXiv:2607.02593v1 Announce Type: cross Abstract: While knowledge distillation (KD) is widely adopted for training lightweight models by leveraging su…
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…
将知识蒸馏引入混合xLSTM架构,探索高效模型压缩新方向
arXiv:2603.15590v2 Announce Type: replace Abstract: There have been numerous attempts to distill quadratic attention-based large language models (LLMs…
将信任区域优化引入策略蒸馏,解决模型压缩中的策略漂移问题。
arXiv:2607.04751v1 Announce Type: cross Abstract: Big goals are hard to achieve all at once; breaking them into small steps is wiser. We present Trust…
Edgee推出Compressor V2,三层次压缩策略将LLM Agent成本砍半,性能与效率兼得。
Article URL: https://www.edgee.ai/blog/posts/introducing-compressor-v2-three-compression-layers-measured-end-to-end-for-a-50-cost-reduction Comments U…
全球首款神经动力学芯片在线模拟平台,将单步运算时延压缩至2.12毫秒,脑皮层重建等任务提速最高达478倍,突破实时计算瓶颈。
IT之家 7 月 4 日消息,北京大学集成电路学院官方公众号昨日(7 月 3 日)发布博文,宣布联合发布全球 首款基于可控存内计算的忆阻器神经动力学芯片, 首次将神经动力学系统的单步运算时延压缩至 2.12 毫秒。 IT之家援引博文介绍,该项目由该校杨玉超教授带领,联合中国科学院上海微系统与信息技术…