AI 情报文章归档

TestingCatalog 称 Gemini 4 Argon 已开始推送，现在可在 GCP 的 Agent Studio 中使用并测试。作者预计 Google AI Studio 和 Gemini apps 随后跟进，截图显示模型列表中 gemini-4-argon 已标为 New。

如果你周五在旧金山，可别错过这个。听说会有很多惊喜！https://luma.com/OpenTogether

What if you could have an AI coding agent that reads your files, writes code, executes commands, and...

全新 ARC Prize 2026 - ARC-AGI-3 高分 68.91%，由 @tufalabs 取得 Tufa Labs 从 56% 跃升至 68%，重新夺回第一名

ANTHROPIC 🔥：Claude Fable 5.5 的痕迹已在 Azure 配置中被发现。 如果这个模型今天发布，它将在许多领域成为新的 SOTA。 快了？👀

Ramp 推出 Glass，在任意模型前加一层共享、可编辑的公司知识层，包含定义、已保存 Skills，以及对接 Linear、Notion、Google、Slack 和 Gong 的连接。它按设计可换模型，同时支持开源与闭源模型，Ramp 的立场是更好模型一发布就切换。文中举例称，模型能正确回答英国客户数量，是因为财务部门数月前修正的定义被所有员工会话共享读取。

Character.ai 已申请破产并挂牌出售，交易预计 2026 年 12 月初前完成，期间平台和订阅照常运行，无买家则服务关停。公司归因于涉儿童伤害的诉讼、监管压力和运营问题；Disney 以 2000 万美元收购部分资产并获得永久非独占技术授权。

不要独自开发，申请加入我们的 Grok Bot X 社区 https://x.ai/bot/community！我们精心策划了以下群聊： • GTM • 产品 • 营销 • 工程 • 招聘 • 设计 在这里你可以分享你正在做的事情，并向其他使用 Grok Bot 的人学习

ScaleDown 推出 Classify、Extract、Summarize 三款小型 AI 模型，均支持 100 万 token 输入、文本或图像，可通过 ScaleDown API 及 OpenRouter、Vercel AI Gateway、LiteLLM、LangChain 调用。

Every restart erased our agents' memory. Every naive relaunch made them redo work they had already done, emails included. Here's how we made them remember everything and relaunch only when it's safe.

Multilingual dense retrievers place all languages in one shared space so that a question finds relevant documents, regardless of which language they are written in. We show that the same question, asked in different languages, retrieves su…

据 Bloomberg 报道，Epic Games 将用 AI 生成代码和美术来开发一款未公布的 Disney 射击游戏，该游戏已从原定的 3 月发售窗口延期。

Task planning for robots in off-site construction prefabrication is primarily manual, leading to planning inefficiencies and reluctance to adopt robotic solutions. Existing approaches provide limited support for scalable, modular automated…

Reusable skills let LLM agents load task-specific instructions on demand, but they make execution depend on implicit routing over skill metadata. We test whether this design scales by holding SkillsBench tasks and their golden skills fixed…

Studio 更新：🎛️ 我们一直在看你们的反馈。更精准的时序、片段拉伸、淡入淡出曲线和弹出式窗口现已上线。 以下是新内容 🧵

Autoregressive language models produce, at every prediction step, a vocabulary-sized logit vector whose structure determines the conditional next-token distribution. We empirically identify that the ordered vocabulary-head logits follow a …

Knowledge distillation trains student models on teacher-generated responses to transfer task capability, but those responses may also carry behavioral tendencies of their source. We investigate whether the source teacher can shape student …

Post-training pruning enables efficient deployment, but conventional methods do not explicitly preserve internal-gradient alignment between sparse and dense models. We study novel connections between downstream performance of compressed mo…

Large language models (LLMs) increasingly act as personal assistants; however, little is known about how they capture national-level preferences on a global scale. Given that "time" is a universal and representative concept, we examine LLM…

Motivated by applications to cognitive radio networks, multi-player multi-armed bandits (MMAB) have attracted considerable attention in recent years. The asynchronous variant, in which players are active only at a fraction of timesteps, be…

Buildings cause nearly 40\% of carbon emissions in the US. Of total building emissions, HVAC systems are responsible for approximately 40-60\% and account for 15\% of global energy consumption. Despite advances in artificial intelligence, …

The safety of LLM-based agents currently rests on alignment training and safety classifiers, neither of which can rule out that a sophisticated indirect prompt injection (IPI) goes through and causes significant harm. Recent work therefore…

Clinical trial protocols serve as the source of truth for study design, but are mostly written and used as unstructured PDF documents. Digitizing protocols turns static content into structured, machine-readable data that clinical software …

如果 AI 能生成任何东西，真正的问题是：你会构建怎样的世界？ Kavan Cardoza — Kavan the Kid @Kavanthekid — 加入 PixLight 2026 评审团，负责 Pilot + Screenplay 单元。 构建只有你能构建的世界。 投稿截止 10 月 31 日。 链接见下。

在空间计算中让梵高的《夜晚的露天咖啡座》活起来。 我们结合了 Gaussian Splatting、AI 生成的 3D 资产和 Apple Vision Pro，打造了这个交互式环境。 下方是技术栈拆解。

Theory 发布 2026 AI in Practice Survey 结果，455 名资深技术从业者参与，安全与治理从第六位升至首要痛点，占 43%，评测以 22% 居次。

Recently, Morris et al. introduced TinyLoRA, a strategy for LLM fine-tuning which introduces arbitrarily few parameters, and demonstrated accuracy improvements using less than $10$ parameters rather than the ${\gtrsim 10,000}$ typically de…

AI is everywhere. I don't know how to start this off really haha. Though to be fair, everyone is...

Egocentric video understanding spans diverse capabilities, modalities, and evaluation formats, yet these are often studied in isolation despite their tight coupling in real-world interaction. We propose the Ego-Action-World interaction loo…

Autoformalization (AF) requires more than compilation: formal statements must preserve the meaning of their informal sources, a property known as faithfulness. An emerging frontier is the automated evaluation of this property, yet most exi…

Tool-call judges evaluate argument values proposed by LLM agents. We find that they accept a correct value less often when their input holds no reference for it, that is, no text that says what the value should be. This happens when the va…

In order to execute quantum algorithms on bosonic quantum computers, it is necessary to encode their inputs into a desired target quantum state, with a process called state preparation. In real settings, state preparation methods operate i…

Objective digital measures of depression severity could complement conventional symptom assessments and facilitate more timely clinical evaluation. Here, we investigated whether task-evoked eye movements provide interpretable behavioral ma…

新的 HomeView，但真的不想让弱智 Siri 来当 AI Assistant 🫠 这正是我们需要的

Muse 用下来完成度第一 但为啥 Grokbot 留存这么低呢？

据 X.PIN 援引消息，阿里巴巴推出基于 Qwen 的智能体 Qwen Go，面向搜索、比价和客服协商任务。演示视频显示它为延误航班拿到 $10 赔偿，并通过淘宝价保追回约 $7。该智能体可查订单、联系人工客服并核验优惠，通过与淘宝和支付宝集成，把个人 AI 助手直接接入购物和支付系统。