1
神舟二十三号搭载 9 项太空实验，含水稻培育、钙钛矿电池等
IT之家 5 月 25 日消息，5 月 24 日 23 时 08 分，搭载神舟二十三号载人飞船的长征二号 F 遥二十三运载火箭在 酒泉卫星发射中心 点火发射。约 10 分钟后，神舟二十三号载人飞船进入预定轨道，发射取得圆满成功。 据中国科学院空间应用工程与技术中心透露，该中心是载人航天工程空间应用系…
2
Whatever the mirror test tells us, beluga whales pass it
白鲸通过镜子测试，但这项自我认知实验本身正受到科学界越来越大的质疑
The white whales join the short, contested list of animals that see themselves.
3
荣耀影像实验室官宣对标阿莱 ARRI 标准，正式升级为电影工业影像实验室
IT之家 5 月 24 日消息，荣耀终端股份有限公司产品线总裁方飞宣布：荣耀影像再次迈出重要一步，荣耀影像实验室对标阿莱 ARRI 技术实验室标准， 正式升级为电影工业影像实验室 。 据方飞介绍，阿莱向荣耀新交付的 ALEXA 系列电影机设备，将用于未来新品的研发使用。她表示，用户后续可以在 Rob…
4
中国空间站将首次开展钙钛矿电池动态服役实验，为月球基地储备关键技术
IT之家 5 月 24 日消息，据央视网昨日报道，神舟二十三号载人飞船将携带 9 项科学实验上行中国空间站，样品及装置总重量约 54 公斤，主要包括肝脏细胞、水稻和拟南芥种子、纳米酶、放线菌、钙钛矿太阳能电池等实验材料。 ▲ 钙钛矿基叠层项目有源样品单元 IT之家从原报道获悉，中国空间站将首次开展钙…
5
I let an AI agent loose on my network – it owned my supply chain in 12 minutes
AI代理在12分钟内攻破裸机供应链容器，展示了自主渗透测试的惊人效率与安全隐患
Article URL: https://dennysentinel.com/blog/deepseek-owned-supply-chain-12-minutes/ Comments URL: https://news.ycombinator.com/item?id=48247425 Points…
6
How to sell RL envs and data to AI labs [video]
教你把强化学习环境和数据卖给AI实验室，从0到1的商业实战经验。
Article URL: https://www.youtube.com/watch?v=gn8prz51Iqs Comments URL: https://news.ycombinator.com/item?id=48211006 Points: 1 # Comments: 0
7
How Much Training Do You Really Need to Do Hyrox?
作者亲自实验Hyrox比赛前最低训练量，用双人赛证明“偷懒”也能完赛的实用指南。
I've never felt stronger, or more scared.
8
Show HN: Asciidia – LLM-Powered Game
用LLM在游戏中随心召唤任何物体，创造龙、法术、武器，AI实时生成功能与属性，脑洞大开！
This is an experiment/demo that I wanted to share and see what people think. The main premise is that you can conjure anything and the game will do it…
9
Extreme Self-Preference in Language Models
研究发现大语言模型存在显著自我偏好，类似生物本能，挑战了AI中性假设。
arXiv:2509.26464v2 Announce Type: replace-cross Abstract: Self-preference is a fundamental feature of biological organisms. Since large language model…
10
AutoLLMResearch: Training Research Agents for Automating LLM Experiment Configuration - Learning from Cheap, Optimizing Expensive
论文提出从廉价到优化的训练方法，自动配置LLM实验，大幅提升研究效率与成本控制。
arXiv:2605.11518v2 Announce Type: replace-cross Abstract: Effectively configuring scalable large language model (LLM) experiments, spanning architectu…
11
React Labs: View Transitions, Activity, and more
React官方发布View Transitions和Activity两项实验性功能，助力前端动画与活动管理，开发者必看！
In React Labs posts, we write about projects in active research and development. In this post, we're sharing two new experimental features that are re…
12
Prior Knowledge or Search? A Study of LLM Agents in Hardware-Aware Code Optimization
最新研究系统对比LLM智能体在硬件感知代码优化中依赖先验知识还是搜索策略，揭示两者优劣边界
arXiv:2605.19782v1 Announce Type: cross Abstract: LLM discovery and optimization systems are increasingly applied across domains, implementing a commo…
13
谷歌推出科研版 Gemini，能追踪论文、写实验代码
科研版Gemini助你追踪论文动态、自动编写实验代码，让研究效率翻倍
IT之家 5 月 20 日消息，在今日的 2026 谷歌 I/O 开发者大会上，谷歌宣布推出科研版 Gemini 模型 —— Gemini for Science 。 Gemini for Science 可以追踪最新论文、将研究目标转化为可用的代码、生成新的假设。 IT之家注意到，谷歌还提到了模拟…
14
Can machine learning for quantum-gas experiments be explainable?
探讨量子气体实验中的机器学习模型是否具备可解释性，揭示前沿交叉领域的关键挑战。
arXiv:2605.18689v1 Announce Type: cross Abstract: Virtually all aspects of many-body atomic physics are challenging: experiments are technically deman…
15
Build a Radio Wave Detector With Balls of Aluminum Foil!
用铝箔球自制无线电波探测器，边玩边学电磁波原理，有趣又有料！
Here’s how you can hack together a radio transmitter and receiver out of stuff you have at home—and explore the weirdness of wireless.
16
Designing Against the Gallery: A Two-Year Journey to a Layered Portfolio Experience
两年打磨个人作品集，如何用分层体验打破传统画廊思维，设计创意者的灵感实录。
A two-year journey to create a layered, engaging portfolio beyond the traditional gallery.
17
What would happen if we just gave people money?
深入探讨全民基本收入（UBI）的研究与乌托邦理想，揭示工作意义的核心困境。
This is a sharp and brilliant piece by Andrew Flowers. He looks at the current research behind universal basic income, where the government would effe…
18
Vector RAG vs LLM-Compiled Wiki: A Preregistered Comparison on a Small Multi-Domain Research
预注册实验对比向量RAG与LLM编译维基在小型研究语料库上的问答效果，透明严谨。
arXiv:2605.18490v1 Announce Type: new Abstract: We preregistered a comparison of two ways to help an LLM answer questions over a small research corpus…
19
How NVIDIA engineers and researchers build with Codex
NVIDIA团队如何利用Codex与GPT-5.5加速生产系统部署和转化研究想法。
Teams use Codex with GPT-5.5 to ship production systems and turn research ideas into runnable experiments.
20
Better Experiments with LLM Evals — A funnel, not a fork
Spotify分享如何用LLM评估提升A/B实验命中率，将测试漏斗与实验反馈循环结合，降低无效成本。
TL;DR LLM evals, automated judges that assess relevance, coherence, and quality at scale, are a powerful new... The post Better Experiments with LLM E…