#And

安全攻防 None

Ivanti、Fortinet、SAP、VMware、n8n 补丁 RCE、SQL 注入、权限提升缺陷

Ivanti, Fortinet, n8n, SAP, and VMware have released security fixes for various vulnerabilities that could be exploited

Ivanti and that

2026-05-18 info@thehackernews.com (The Hacker News)

安全攻防 None

开发人员工作站现已成为软件供应链的一部分

Supply chain attackers are not only trying to slip malicious code into trusted software。They are trying to steal the acc

and are trying

2026-05-18 info@thehackernews.com (The Hacker News)

开发者生态 None

可证明安全操作系统 (PSOS) 的基础 (1979) [pdf]

【HN用户评论摘要】 I understand why in 1979 and perhaps until mid 1990s capability OS architecture might have been irrelevant an

and understand why

2026-05-18 rurban

智能AI None

韩国 LetinAR 正在构建人工智能眼镜背后的光学器件

A lens the size of a thumbnail — and the South Korean startup that makes it could become the optical backbone of the AI

the lens size

2026-05-18 Kate Park

开源推荐 None

GitHub 热门项目：LLM101n

GitHub项目：LLM101n 仓库地址：https://github。

Chapter LLM build

2026-05-18 GitHub Trending

开源推荐 None

GitHub 热门项目：OpenManus

GitHub项目：OpenManus 仓库地址：https://github。com/FoundationAgents/OpenManus Stars：56294 | 作者：FoundationAgents 项目描述：No fortress

https OpenManus com

2026-05-18 GitHub Trending

科技头条 None

一对丹麦夫妇特立独行的非洲研究在小罗伯特·肯尼迪的疫苗政策中发挥了重要作用

The work of Peter Aaby and Christine Stabell Benn has long been controversial。Until Robert F。became US health policy chi

The work Peter

2026-05-18 Arthur Allen, KFF Health News

开发者生态 None

是时候放弃意识辩论带来的二元论了

There Is No ‘Hard Problem Of Consciousness’ Consciousness is not separate from the physical world — our “soul” is of the

the and our

2026-05-18 ahalbert4

智能AI None

解读细胞，设计治疗方法：用于功能导向药物设计的微扰条件分子扩散

arXiv:2605.15243v1 Announce Type: new Abstract: When reliable target structures are unavailable at scale or phenotypes a

textbf the and

2026-05-18 Ziyu Xu, Zijian Zhang, Liang Wang, Zhiyuan Liu, Qiang Liu, Shu Wu, Liang Wang

智能AI None

通过合规自我蒸馏降低法学硕士安全调整中的安全税

15239v1 Announce Type: new Abstract: Safety alignment often improves robustness to harmful queries at the cost of reason

safety and the

2026-05-18 Yu Fu, Longxuan Yu, Haz Sameen Shahgir, Zhipeng Wei, Hui Liu, N. Benjamin Erichson, Yue Dong

智能AI None

MuteBench：不完全多模态融合的模态不可用容差评估

15235v1 Announce Type: new Abstract: Multimodal physiological data powers clinical AI systems from intensive care units

modality missing the

2026-05-18 Wugeng Zheng, Ziwen Kan, Tianlong Chen, Chen Chen, Song Wang

智能AI None

Mask-Morph Graph U-Net：一种基于网格的通用代理，用于大几何变化下的耐撞性场预测

15231v1 Announce Type: new Abstract: Nonlinear finite element crash simulations are accurate but computationally expensi

graph and edge

2026-05-18 Haoran Li, Tobias Lehrer, Yingxue Zhao, Haosu Zhou, Philipp Stocker, Tobias Pfaff, Nan Li

智能AI None

量化破坏一致性：压缩法学硕士在模型和精度水平上出现偏差

arXiv:2605.15208v1 Announce Type: new Abstract: Large Language Models are routinely compressed via post-training quantiz

bit and new

2026-05-18 Plawan Kumar Rath, Rahul Maliakkal

智能AI None

TeamTR：多代理 LLM 协调的信任区域微调

15207v1 Announce Type: new Abstract: Multi-agent LLM systems have shown promise for complex reasoning, yet recent evalua

and agent that

2026-05-18 Yi Xie, Siao Liu, Falong Fan, Yuanqi Yao, Yue Zhao, Bo Liu

智能AI None

AgentStop：提前终止本地 AI 代理以节省消费设备的能源

arXiv:2605.15206v1 Announce Type: new Abstract: Autonomous agents powered by large language models (LLMs) are increasing

and agents based

2026-05-18 Dzung Pham, Kleomenis Katevas, Ali Shahin Shamsabadi, Hamed Haddadi

智能AI None

ICRL：通过强化学习学习将自我批评内化

15224v1 Announce Type: new Abstract: Large language model-based agents make mistakes, yet critique can often guide the s

the critique solver

2026-05-18 Jianbo Lin, Xiaomin Yu, Yi Xin, Yifu Guo, Zhuosong Jiang, Zhongqi Yue, Weishi Wang, Heqing Zou, Chengwei Qin, Hui Xiong

智能AI None

NOVA：通过人工智能发现知识的基本限制

15219v1 Announce Type: new Abstract: Can AI systems discover genuinely new knowledge through iterative self improvement,

the knowledge and

2026-05-18 Salman Avestimehr, Ken Duffy, Muriel M\'edard

智能AI None

CAX-Agent：用于可靠 APDL 自动化的轻量级代理线束

arXiv:2605.15218v1 Announce Type: new Abstract: Large language models deployed for MAPDL finite-element simulation face

and recovery the

2026-05-18 Chenying Lin, Yichen Hai, Yi He, Ran Wang, Haiyan Qiang, Liang Yu

智能AI None

公平的产出，有偏见的内部：法学硕士高风险决策中潜在偏见的因果效力和不对称性

15217v1 Announce Type: new Abstract: Instruction-tuned language models exhibit behavioural fairness in high-stakes decis

and models decisions

2026-05-18 Jagdish Tripathy, Marcus Buckmann

智能AI None

SkillSmith：将代理技能编译为边界引导运行时接口

15215v1 Announce Type: new Abstract: Recently, skills have been widely adopted in large language model (LLM)-based agent

and skills runtime

2026-05-18 Duling Xu, Zheng Chen, Zaifeng Pan, Jiawei Guan, Dong Dong, Jialin Li, Bangzheng Pu