#And

安全攻防 None

Dirty Frag Vulnerability Made Public Early: Root Privilege On All Distributions

One week after the Copy Fail vulnerability, a new Linux local privilege escalation bug has been made public。

the vulnerability and

2026-05-08 Michael Larabel

智能AI None

AMD 的本地开源 AI 现在可以轻松与您的 Gmail 交互

AMD software engineers continue rapidly advancing their open-source software efforts around local AI/LLM use on consumer

AMD local software

2026-05-08 Michael Larabel

安全攻防 None

超越神经网络的数据驱动变分基础学习：自适应基础发现的非神经框架

arXiv:2605.05221v1 Announce Type: new Abstract: Classical representation systems such as Fourier series, wavelets, and f

and basis the

2026-05-08 Andrew Kiruluta

安全攻防 None

MidSteer：用于转向生成模型的最佳仿射框架

arXiv:2605.05220v1 Announce Type: new Abstract: Steering intermediate representations has emerged as a powerful strategy

and for concept

2026-05-08 Tatiana Gaintseva, Andrew Stepanov, Ziquan Liu, Martin Benning, Gregory Slabaugh, Jiankang Deng, Ismail Elezi

智能AI None

用于混合和循环 LLM 服务的稀疏前缀缓存

05219v1 Announce Type: new Abstract: Prefix caching is a key latency optimization for autoregressive LLM serving, yet ex

the and recurrent

2026-05-08 Mikhail Shirokikh, Sergey Nikolenko

安全攻防 None

用于混沌预测的视界约束罗生门集

arXiv:2605.05218v1 Announce Type: new Abstract: Predictive multiplicity and chaotic dynamics represent two fundamental c

that the and

2026-05-08 Gauri Kale, Rahul Vishwakarma, Holly Diamond, Ava Hedayatipour, Amin Rezaei

智能AI None

具有可学习损失平衡和迁移学习功能的物理神经网络

arXiv:2605.05217v1 Announce Type: new Abstract: We propose a self-supervised physics-informed neural network (PINN) fram

and physics data

2026-05-08 Reza Pirayeshshirazinezhad

智能AI None

SAT：协调员的顺序代理调整免费即插即用多 LLM 培训并保证单调改进

05216v1 Announce Type: new Abstract: Large language models (LLMs) with a large number of parameters achieve strong perfo

the and with

2026-05-08 Yi Xie, Yangyang Xu, Yi Fan, Bo Liu

智能AI None

使用人口分层模型进行全国范围内基于 EHR 的慢性鼻窦炎预测

05213v1 Announce Type: new Abstract: Chronic rhinosinusitis (CRS) is a common heterogeneous inflammatory disorder that c

and CRS that

2026-05-08 Sicong Chang, Yidan Shen, Justina Varghese, Akshay R Prabhakar, Sebastian Guadarrama-Sistos-Vazquez, Jiefu Chen, Masayoshi Takashima, Omar G. Ahmed, Renjie Hu, Xin Fu

智能AI None

平坦极小值是一个幻觉吗？

05209v1 Announce Type: new Abstract: Neural networks that land in flat regions of the loss landscape tend to generalise

the and weakness

2026-05-08 Michael Timothy Bennett

智能AI None

代理岛：多代理游戏的饱和和抗污染基准

04312v1 Announce Type: new Abstract: Static capabilities benchmarks suffer from saturation and contamination, making it

and model models

2026-05-08 Connacher Murphy

智能AI None

推测生成的并行前缀验证

arXiv:2605.04263v1 Announce Type: new Abstract: We introduce PARSE (PArallel pRefix Speculative Engine), a speculative g

verification and model

2026-05-08 Yuncheng Yao, Yuxuan Xia, Shengjie Wang, Danyang Zhuo

智能AI None

Pro$^2$Assist：通过多模式自我中心感知为长期程序任务提供持续的步骤感知主动协助

04227v1 Announce Type: new Abstract: Procedural tasks with multiple ordered steps are ubiquitous in daily life。

Pro Assist and

2026-05-08 Lilin Xu, Bufang Yang, Siyang Jiang, Kaiwei Liu, Kaiyuan Hou, Yuang Fan, Hongkai Chen, Zhenyu Yan, Xiaofan Jiang

开发者生态 None

ANDRE：基于注意力的神经符号可微规则提取器

arXiv:2605.04193v1 Announce Type: new Abstract: Inductive Logic Programming (ILP) aims to learn interpretable first-orde

and rule ILP

2026-05-08 Iman Sharifi, Peng Wei, Saber Fallah

智能AI None

通过时间扩展交互图对手术团队动态进行可操作的实时建模

04169v1 Announce Type: new Abstract: Surgical team performance arises from complex interactions between technical execut

and team time

2026-05-08 Vincenzo Marco De Luca, Antonio Longa, Giovanna Varni, Andrea Passerini

数码硬件 None

正则化中心强调时间差异学习

04100v1 Announce Type: new Abstract: Off-policy temporal-difference (TD) learning with function approximation faces a st

the and policy

2026-05-08 Xingguo Chen, Chaohui Wu, Jinguo Ye, Chao Li, Shangdong Yang, Guang Yang, Tianyu Liang, Wenhao Wang

智能AI None

LCM：无损上下文管理

04050v1 Announce Type: new Abstract: We introduce Lossless Context Management (LCM), a deterministic architecture for LL

context LCM and

2026-05-08 Clint Ehrlich, Theodore Blackman

智能AI None

Uber 使用 OpenAI 帮助人们更智能地赚钱并更快地预订

Uber uses OpenAI to power AI assistants and voice features that help drivers earn smarter and riders book faster across

and Uber uses

2026-05-08 OpenAI博客

智能AI None

Singular Bank 通过 ChatGPT 和 Codex 帮助银行家快速行动

Singular Bank built Singularity, an internal assistant using ChatGPT and Codex to help bankers save 60–90 minutes daily

and Singular Bank

2026-05-08 OpenAI博客

智能AI None

Simplex 使用 Codex 重新思考软件开发

Simplex boosts software development with ChatGPT Enterprise and Codex, reducing design, build, and testing time while sc

and Simplex boosts

2026-05-08 OpenAI博客