1
启境汽车 CEO 刘嘉铭：启境坚定不移携手华为乾崑的大原则，不会变
IT之家 8 月 13 日消息，最近，汽车行业和华为的合作成为了热门话题。启境汽车 CEO 刘嘉铭今日发文，代表启境表达了观点。 IT之家附刘嘉铭微博原文如下： 先说结论， 启境坚定不移携手华为乾崑的大原则，不会变 。 首先，启境品牌是广汽集团携手华为乾崑联合打造的高端智能电动车品牌。在行业“电动化…
2
Extractable Memorization From First Principles
从第一原理出发，揭示LLM可提取记忆研究中两种相反的效度陷阱，平衡过度提取与不可靠证据的争议。
arXiv:2607.12649v1 Announce Type: new Abstract: Recent work on extractable memorization in LLMs suffers from two contrasting validity problems. Some s…
3
国务院：原则同意《中医药振兴发展“十五五”规划》
36氪获悉，国务院发布关于《中医药振兴发展“十五五”规划》的批复，原则同意《中医药振兴发展“十五五”规划》。《规划》实施要坚持以习近平新时代中国特色社会主义思想为指导，深入贯彻党的二十大和二十届历次全会精神，认真落实四中全会部署，全面贯彻习近平总书记关于健康中国建设和中医药工作的重要论述，坚定不移走…
4
Software Development Changed. Good Engineering Didn’t.
变化比想象中快，不变比想象中重要——判断好工程师的标准从未改变。
Over the last 10 years, software development has changed dramatically. Products are built faster, teams use more cloud services, data has become centr…
5
梁汝波发布全员邮件，字节跳动时隔 4 年更新领导力原则
字节跳动时隔4年更新领导力原则，新增“做有高度的事”与“敢于设定高目标”，并强化危机感与一线视角，看点十足。
6 月 29 日晚间，字节跳动 CEO 梁汝波发布全员邮件。梁汝波在邮件中表示，基于行业正在发生的变革和组织自身发展需求，我们最近重新梳理和更新了我们的文化内容，希望在面对新的机遇和挑战时，明确我们要以什么样的方式出发。 梁汝波在邮件中称，“激发创造，丰富生活”是字节跳动不变的使命。过去谈使命，更多…
6
美国政府以《网联汽车原则》为由，禁止新款极星电动汽车在美销售
IT之家 6 月 26 日消息，据路透社报道，极星（Polestar）昨日表示，美国政府要求公司停止在美国市场销售新款电动汽车。受此消息影响，极星股价一度下跌 6.3%。 据报道，美国商务部以《网联汽车原则》为由，拒绝了极星的销售许可请求。根据该规则，拥有中国相关联网技术（IT之家注：涵盖蓝牙、Wi…
7
AI Slop Cannons
不想让AI输出变成垃圾？五条原则教你精准使用AI，真正创造价值。
Article URL: https://www.augmentedswe.com/p/ai-slop-cannons-and-their-consequences Comments URL: https://news.ycombinator.com/item?id=48602557 Points:…
8
Rest & Restfulness API Design Principles
通俗拆解REST API核心设计原则，从资源表示到实践细节，适合开发者快速掌握规范
Re presentational S tate T ransfer is an architectural style used for designing the network applications. Instead of relying on complex protocols REST…
9
Quanto guideline um agente de código precisa?
从Anthropic的规划与生成代理配置出发，探讨代码代理究竟需要多少指导方针。
Disclaimer Este texto foi inicialmente concebido pela IA Generativa em função da transcrição de um vídeo do Dev Eficiente. Se preferir acompanhar por …
10
Helm Charts for a Simple SPA: Over-Engineering or Smart DevOps?
深度探讨Helm Charts对简单SPA应用是否是过度工程，用实际YAML对比揭示DevOps决策取舍。
Ah, the Single Page Application (SPA). You’ve built it in React, Vue, or Svelte. You’ve run npm run build, and out pops a dist/ folder containing a fe…
11
Lean Inference: Lean Manufacturing Principles Applied to AI
将制造业精益原则引入AI推理，降本增效的新思路。
Article URL: https://neurometric.substack.com/p/lean-inference-workflows-applying Comments URL: https://news.ycombinator.com/item?id=48387109 Points: …
12
How To Make Your Design System AI-Ready
让设计系统无缝对接AI，三大层次与审计工具实战指南，设计团队必读。
Practical guide on how to reduce drifts, minimize mistakes, maintain context, and improve the quality of AI-generated prototypes. Brought to you by De…
13
Safety Must Precede the Deployment of Open-Ended AI
ICML'26 论文强调开放型 AI 部署前必须优先保障安全，为行业敲响警钟。
arXiv:2502.04512v4 Announce Type: replace Abstract: AI advancements have been significantly driven by a combination of foundation models and curiosity…
14
Towards Efficient LLMs Annealing with Principled Sample Selection
提出原则性样本选择方法提升LLM退火效率，向高质量模型训练迈进。
arXiv:2605.31175v1 Announce Type: new Abstract: The annealing phase is a pivotal convergence stage in LLM pre-training that ultimately determines fina…
15
国务院：国家支持投资者按照市场化原则开展对外投资活动，积极参与国际合作竞争
36氪获悉，《国务院关于对外投资的规定》现予公布。国家支持投资者按照市场化原则开展对外投资活动，积极参与国际合作竞争。投资者依法享有对外投资自主权，自主决策、自担风险、自负盈亏。投资者开展对外投资及其相关活动，应当遵守法律法规和国际惯例，尊重当地习俗和文化传统，遵守商业道德，诚实守信、公平竞争，履行…
16
Knowing When to Quit: A Principled Framework for Dynamic Abstention in LLM Reasoning
让LLM学会"知难而退"：一个原则性框架，帮助模型动态判断何时应该放弃推理，提升可靠性与效率。
arXiv:2604.18419v4 Announce Type: replace Abstract: LLMs utilizing chain-of-thought reasoning often waste substantial compute by producing long, incor…
17
We Think, Therefore We Align LLMs to Helpful, Harmless and Honest Before They Go Wrong
LLM对齐新思路：在模型出错前就教会它思考“有用、无害、诚实”
arXiv:2509.22510v3 Announce Type: replace Abstract: Alignment of Large Language Models (LLMs) is the ability to satisfy desired objectives during gene…
18
From Descriptive to Prescriptive: Uncover the Social Value Alignment of LLM-based Agents
提出GraphRAG框架将原则转为价值指令，解决LLM智能体在自认知、困境决策及情绪上的对齐缺陷。
arXiv:2605.14034v1 Announce Type: new Abstract: Wide applications of LLM-based agents require strong alignment with human social values. However, curr…
19
Don’t draw the UI, draw the priority
别再急于画界面，先列出内容优先级清单，这才是高效设计的起点。
Someone gave me some great feedback in the moments when I’m struggling with a design: “Don’t draw the UI, draw the priority instead.” What they meant …
20
Enhanced and Efficient Reasoning in Large Learning Models
图灵奖得主Valiant提出高效原则性推理方法，破解大模型可信度难题
arXiv:2605.14036v1 Announce Type: new Abstract: In current Large Language Models we can trust the production of smoothly flowing prose on the basis of…