1
CSS and Momentum
CSS在2021年迎来重大突破，容器查询和:has伪类即将落地，将重塑前端开发方式。
I reckon 2021 was the best year for CSS since…2015? I haven’t felt this level of excitement and momentum in years. In the last twelve months container…
2
Packing My Library
阿尔贝托·曼古埃尔对创造与失去的深刻洞察，引你在阅读中思考存在的意义
I was reading Alberto Manguel’s Packing My Library on the train back to SF the other day and the moment I read this passage I hurriedly underlined it …
3
Back to the Future with RSS
怀念RSS的开放与自由：无广告、无算法、完全掌控，回到未来重拾信息聚合初心
Nicky Case writes about why RSS was so damn good : Imagine an open version of Twitter or Facebook News Feed, with no psy-op ads, owned by no oligopoly…
4
Norse Mythology
尼尔·盖曼重述北欧神话，两天读完的短篇集，视角独特引人入胜
I ravaged this collection of short stories by Neil Gaiman called Norse Mythology ; it reminds me of Karin Tidbeck’s Jagannath and brought me back to r…
5
Pairing is the key to evangelizing your design system
分享推广设计系统的秘诀：通过配对协作打破对立，让设计师与工程师一起构建更高效的产品流程。
A while ago I wrote a piece for the folks at Figma all about one of the best ways to evangelize your design system: pairing with designers and enginee…
6
With New Browser Tech, Apple Preserves Privacy and Google Preserves Trackers
苹果与谷歌浏览器技术背后的隐私博弈：一个守护用户，一个纵容追踪
Writing for the Electronic Frontier Foundation, Alan Toner describes the current state of advertising on the web and criticizes Google for not just al…
7
[Outliers] Phil Knight: The Obsession That Built Nike
Nike创始人Phil Knight在四面楚歌时如何靠疯狂信念逆袭，这段创业史比小说更刺激。
Phil Knight is the founder of Nike, the brand that reshaped sports and became one of the most powerful companies in the world. What would you do if yo…
8
Templates in ElevenCreative
一键完成创意到成品，ElevenCreative模板让设计更高效。
From idea to finished asset in one click Discussion | Link
9
The US is betting on AI to catch insider trading in prediction markets
美国CFTC利用AI监控Polymarket等预测市场，严查可疑内幕交易，监管新动向值得关注
The Commodity Futures Trading Commission wants us to know it's taking this very seriously.
10
Users turn to jailbreaking their older Kindles as Amazon ends support
亚马逊终止旧款Kindle技术支持，用户纷纷越狱以解锁自定义字体、屏保和第三方应用延长设备寿命
It may be possible to jailbreak an older, end-of-support Kindle and continue adding books to it. But doing so carries risks.
11
PolitNuggets: Benchmarking Agentic Discovery of Long-Tail Political Facts
PolitNuggets基准测试评估大模型在多语言长尾政治事实发现中的能力，为智能体信息合成提供新标尺。
arXiv:2605.14002v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) embedded in agentic frameworks have transformed information retrieval fr…
12
Parallelizing Counterfactual Regret Minimization
将并行计算引入反事实遗憾最小化算法，大幅提升游戏求解效率，AI博弈新突破
arXiv:2605.14277v1 Announce Type: new Abstract: Parallelization has played an instrumental role in the field of artificial intelligence (AI), drastica…
13
Metis AI: The Overlooked Middle Zone Between AI-Native and World-Movers
提出“Metis AI”新概念，揭示数字任务中AI失效的隐秘界限，挑战传统物理/数字二分法
arXiv:2605.14407v1 Announce Type: new Abstract: The dominant discourse on AI limitations frames the boundary of AI capability as a divide between digi…
14
PyCSP3-Scheduling: A Scheduling Extension for PyCSP3
PyCSP3新增调度扩展，原生支持区间变量和资源函数，简化约束建模与求解分离。
arXiv:2605.14559v1 Announce Type: new Abstract: PyCSP$^3$ provides a productive way to build constraint models for solving combinatorial constrained p…
15
SliceGraph: Mapping Process Isomers in Multi-Run Chain-of-Thought Reasoning
用互kNN构建过程图，追踪多链推理中路径的共享与分裂，揭示思维过程几何。
arXiv:2605.14619v1 Announce Type: new Abstract: Multi-run chain-of-thought reasoning is usually collapsed to final-answer aggregates, which discard ho…
16
Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning
揭秘大模型“未来知识”作弊：时间批判让AI学会说“不知道”
arXiv:2605.14636v1 Announce Type: new Abstract: Large language models (LLMs) often fail to reason under temporal cutoffs: when prompted to answer from…
17
Monitoring Data-aware Temporal Properties (Extended Version)
提出一种面向数据感知时序属性的监测方法，适用于未知动态AI系统的实时评估。
arXiv:2605.14666v1 Announce Type: new Abstract: Dynamic systems in AI are often complex and heterogeneous, so that an internal specification is not ac…
18
On Strong Equivalence Notions in Logic Programming and Abstract Argumentation
逻辑编程与抽象论证中的强等价概念，为知识库替换提供理论保障。
arXiv:2605.14721v1 Announce Type: new Abstract: Strong equivalence between knowledge bases ensures the possibility of replacing one with the other wit…
19
OpenDeepThink: Parallel Reasoning via Bradley--Terry Aggregation
提出OpenDeepThink方法，用Bradley-Terry模型实现并行推理，无需真实验证器即可筛选最佳候选，为LLM推理扩展新路径。
arXiv:2605.15177v1 Announce Type: new Abstract: Test-time compute scaling is a primary axis for improving LLM reasoning. Existing methods primarily sc…
20
GAMBIT: A Three-Mode Benchmark for Adversarial Robustness in Multi-Agent LLM Collectives
首个专注多智能体LLM集体对抗鲁棒性的三模式基准，揭示单一欺骗智能体如何突破现有防御。
arXiv:2605.09027v2 Announce Type: cross Abstract: In multi-agent systems (MAS), a single deceptive agent can nullify all gains of an agentic AI collec…