AI 情报文章归档

作者引用高盛研究称，总 token 需求从 2025 年 12 月到 2026 年 9 月增长约 18 倍，而前沿模型 token 需求同指数下仅增长约 8-9 倍，7 月前沿 token 需求增速放缓。

Flash 0.5.5 turns a local-first terminal agent into something you can use from any screen in your house, without sending a single line of code to someone else's server.

https://github.com/567-labs/instructor/discussions/2706

Originally published on Medium. I maintain bcb-br-mcp, an open-source MCP server that gives AI...

By Andrew Pyle The first time I tried to run two coding agents at once on the same project, it went...

有人想用 jev 搭配 instructor 吗？

一位 Haskell 程序员提出，用 LLM 做规划、调研等枯燥工作，而把写代码留给自己，以避免技能退化和代码库变成只有 coding agent 能维护的"LLM 荒原"。他建议让 LLM 把专家对话转成可执行 todo、用 markdown frontmatter 跟踪计划项，但关键决策必须由人来做，调研时自己也应并行搜索、至少大致掌握 agent 所知的一切。

Geoffrey Hinton 在 CNN 采访中主张对 AI 实行更强政府监管，认为仅靠标准不够，大型 AI 公司口头支持监管却反对任何具体方案。他提出 AI 应至少达到药品监管强度：新药须经 FDA 证明安全才能上市，AI 也应如此。

一个巴掌拍不响，第二天

账户通过 Plaid 安全连接。访问为只读，Grok Bot 永远不会看到你的登录信息，你可以随时取消关联。 在市场（marketplace）中搜索“Finance”即可开始。 试用 Grok Bot：https://x.ai/bot

Grok Bot 推出 Finance 集成，用户可关联银行、卡和投资账户，让 Bot 协助管理支出和投资等。配图演示了查询订阅的示例，Bot 从 Chase 和 Amex 拉取近 12 个月循环扣款，找出 14 个订阅每月共 $286，其中两个自春季以来涨价。

Hugging Face 联创 Thomas Wolf 称"手写代码的美好时代已经结束"，软件工程师将转向"专业造物者"的新职业，去指挥此前只存在于科幻中的智能。他引用 @scottstts 的演讲表示，每个软件工程师乃至每个人都该听听，保留美好回忆、正视现实、对未来保持兴奋。

巴西 AI 实验室 Supersonic Labs 发布 144.3M 参数的开源决策模型 Julia 1，权重以 Apache 2.0 协议托管在 Hugging Face，可在普通 CPU 上运行，不生成文本，只对 2 到 20 个候选答案输出选择和概率。

Astra killed Claude. Claude killed Gemini. Gemini killed Google. Google committed suicide, and...

突发 🔥：OpenAI 预告将在下周的 DevDay 活动上推出“o”常驻智能体。 要来了 👀

An agent needs a boundary between proposing an action and taking it. I would put that boundary in...

距离 OpenAI DevDay 还有 72 小时。 我们一直在打磨。是时候展示成果了。

An on-call agent that answers "no prior incidents" might be telling you the archive is empty. It...

Chamath 讲述他如何在个人生活与工作中使用定制化的 AI 智能体和技能。 - 用于高风险谈判的纳什均衡 AI 智能体 - 用于重大人生决策的个人 AI 世界模型 - 他实际上围绕基础模型构建了一套持久化的微服务 ---- 来自 "Chamath Palihapitiya" YouTube 频道，（链接见评论）

原来你不需要IMU 🤯 [引用 @aditya_bhatt]：在我最新的博士论文中，我们向"堆传感器"宣战。 首次实现抗推挤的人形机器人行走，仅靠关节编码器！无需IMU，无需力/力矩传感器。 🥁 隆重介绍 Blind Dexterity 🧵 👇 与 @OKaidanov @liu_puze @Jan_R_Peters 合作，来自 @DFKI @ias_tudarmstadt

乌克兰前国防部长Mykhailo Fedorov在IT Arena 2026上宣布推动战场全面机器人化的私营计划，称无人机已占所有目标交战的95%以上。乌克兰现有超过700家无人机制造商，其AI作战数据已成出口产品，英国是首个买家。该计划将投资防务科技公司并与军方规模化测试系统，同时兼具威慑俄罗斯和吸引投资的心理战意图。

OpenAI 的 Jalapeno 负责人回应 NVIDIA 首席科学家在计算机博物馆座谈会 YouTube 评论区的质疑，称 Jalapeno 已在 SemiAnalysis InferenceX 基准上运行非 OpenAI 模型且速度很快，甚至超过 NVIDIA 的 July Rubin 芯片。

Table of Contents The problem The architecture The Implementation Creating an...

Two checks in my build were green for months while testing nothing. One was switched off by a guard clause, the other by a missing space in a regex.

Coding agents such as Claude Code and Codex run with your shell and your credentials. Between October...

当前 AI 热点，聚合 3 个独立信源、0 条信号。

Jev 正迅速成为 OpenRouter 上分类请求的首选。 它占据了该类别每周请求量的 27%，几乎是此前位居榜首的 DeepSeek V4 Flash 份额的两倍

Learning transferable representations for dynamic text-attributed graphs (DyTAGs) requires models to capture underlying interaction dynamics that persist across domains. However, existing methods tend to overfit to domain-specific structur…

We propose sliced Orlicz-Wasserstein (SOW) distance which is a generalization of sliced Wasserstein (SW) distance. SOW replaces the $L^p$ norm in SW with a Luxemburg norm cost induced by an Orlicz function $φ$. First, we prove that SOW dis…

Multimodal contrastive learning is a dominant paradigm for learning transferable representations from unlabeled data, but standard objectives primarily capture information that is redundant between modalities. Partial Information Decomposi…

Nathan Lambert 认为，工作文化适配前沿模型与员工交互方式的企业，将相对竞争对手获得巨大加速，员工使用这些模型的灵活度同样关键。他引用 @signulll 的观点：若给 Opus 5.5 完整接入公司邮件、Slack、文档、浏览器、数据库、日历、内部工具、权限与组织记忆，并配合可靠的计算机使用与验证循环，几乎所有白领工作已可无需人类完成。

Accurate assessment of Schistosoma japonicum-associated liver fibrosis is essential for disease management and long-term follow-up in endemic regions. Ultrasound provides non-invasive imaging, but complex local echogenic patterns and anato…

Recommender systems rely heavily on heterogeneous behavioral feedback to infer user preference. Although abundant, these signals are imperfect measurements: the same observed behavior can arise from different underlying states, such as gen…

Predicting smooth dynamics and collisions requires modeling continuous evolution and abrupt state changes. We introduce HamiFormer, a dual-expert diffusion field combining whole-window denoising with residual-corrected Hamiltonian propagat…

Ultrasound (US) acquisition depends on the operator's ability to interpret anatomy and anticipate how the view will change with probe motion. Many robotic US navigation methods select actions without explicitly predicting these anatomical …

Radiomap blind prediction aims to infer radiomaps from observable representations of the propagation environment and base station configuration without field measurements. In practice, the observable representations are inherently incomple…

Runway 即将登陆 Muse。

As AI agents are becoming widely adopted in the financial services industry, careful measurement is essential to understand where they can be reliably deployed and where oversight and professional review remain necessary. Such measurement,…

This is a submission for the Sanity Challenge, Path Two: Vibe-Code Something Strange What I...

A trained neural network can be represented by a parameter vector in high dimensions. Learning distributions over these vectors enables the generation of new models across various tasks and architectures. A central challenge is permutation…