AAI Search
AI 资讯日报 · 2026-11-04

2026-11-04 · AI 资讯日报

今日新收录 172 条公开资讯，按模型 / 产品 / 行业 / 论文 / 观点 自动归类汇编（非 AI 生成，点击可溯源原文）。

今日精选6 条

  1. 1.
    论文研究Nearly Linear-Time Follow-the-Perturbed-Leader for Weakly Observable Graph Bandits— arXiv

    We study online learning with directed weakly observable feedback graphs and propose a Follow-the-Perturbed-Leader (FTPL) policy that achieves best-of-both-worlds guarantee with nearly linear expected…

  2. 2.
    论文研究FlowTranslator: Training-Free In-Image Machine Translation with Flow-Based Editing— arXiv

    In-image machine translation (IIMT) replaces text in an image with its translation while preserving the surrounding visual context. Existing task-specific models achieve convincing performance but req…

  3. 3.
    论文研究OSGR: Observation-Support Graph Reasoning for Earth Observation Question Answering— arXiv

    Answering geographic queries over Earth observation (EO) archives requires solving a joint retrieval-and-derivation problem under strict spatiotemporal constraints. Existing LLM agents fail to ground…

  4. 4.
    论文研究SEAM: Situated Executable Associative Memory for a Co-located Community— arXiv

    In groups, knowledge is distributed among individuals, and how it is brought together depends on social ties. Taking inspiration from how a single mind recalls and reacts to memories and current stimu…

  5. 5.
    论文研究Reciprocity Can Erase Direction:tructural Boundaries and Controlled Utility of Magnetic Graph Networks— arXiv

    Magnetic graph networks encode direction through phase, but a learned graph can erase the intended directional mechanism before prediction. We characterize this failure for support-based magnetic Lapl…

  6. 6.
    论文研究How Risky Is Your Data-Driven Policy?— arXiv

    Data-driven algorithms are ubiquitous across industries, and there is an extensive literature on analyzing their \emph{average} performance. The risk profile of such algorithms, while critical, is muc…

模型发布/更新6 条

  1. 1.
    Agentic cheminformatics with Google Antigravity— Google DeepMind

    Learn how Google Antigravity and our Cheminformatics Skills accelerate drug discovery workflows, search compounds, and propose novel rigidified macrocycles.

  2. 2.
    AI Is Changing How Semiconductor Fabs Work — and Engineers Are Moving Faster Because of It— NVIDIA

    Demand for advanced chips is rising. The processes to build them are getting more complex. And the margin for error is shrinking. For engineers, understanding how a process change…

  3. 3.
    Accelerating engineering for the physical world— OpenAI

    Discover how ChatGPT and Codex are helping engineers accelerate CAD, PCB design, simulation, and other workflows across the physical world.

  4. 4.
    How Cisco defends at machine speed with OpenAI— OpenAI

    Cisco combines network intelligence with OpenAI models to strengthen enterprise security. Learn how it is building more adaptive defenses.

  5. 5.
    Heimstaden is changing residential real estate management with AI— OpenAI

    Discover how Heimstaden uses OpenAI’s ChatGPT Enterprise and Codex to transform property management, cut costs, and empower employees across Europe.

  6. 6.
    Cities Worldwide Build and Scale AI Services With NVIDIA Physical and Agentic AI— NVIDIA

    Cities generate enormous amounts of data from cameras, sensors and operational systems, but turning it into useful action — for planning, maintenance and public services — remains…

产品发布/更新6 条

  1. 7.
    Anthropic's Message Threads cut time to first token by 35 percent for long conversations— The Decoder

    Anthropic released Message Threads for the Messages API. Instead of resending the entire conversation with every turn, agents now only transmit what's changed. For requests between…

  2. 8.
    Soniox STT v6 Adds Agent Text Input and Echo Filtering for Voice Agents— MarkTechPost

    AI 导读 · 语音智能体可边说边听，回声过滤解决自干扰痛点，值得关注。

  3. 9.
    OpenAI Introduces Pando: A Git Backend That Cuts Median Push Latency ~80%— MarkTechPost

    OpenAI built Pando, a Git-compatible backend for its monorepo. It handles branches independently, reuses stored packfiles across fetches, cut median push latency by nearly 80%, and…

  4. 10.
    beipo-ai/NorthCUA— GitHub

    Turn intent into work. Open-source Desktop/Web agent harness with long-horizon memory, live evaluation and recoverable batches.

  5. 11.
    Jumbomuffin777/Frontier-UI— GitHub

    ChatGPT-style generative UI for Open WebUI: photo galleries, live maps with real routes, multi-day trips, budgets, calculators, live weather & stocks, link prev…

  6. 12.
    uclanlp/OpenVLHarness— GitHub

    Official Code for OpenVLHarness: Real-World Visual Agentic Tool‑Use with Multimodal Memory

行业动态6 条

  1. 13.
    Postman silently auto-installs AI plugins and attempts to share all API data— Hacker News
  2. 14.
    Google will turn off search for schools that reject AI— Hacker News

    https://archive.ph/RuwsB https://blog.google/products-and-platforms/products/educatio...

  3. 15.
    Building connections across the research infrastructure— MIT News

    AI 导读 · 共享科研设施如何跨机构互联，是提升芯片研发效率的关键看点。

  4. 16.
    OpenAI seems to be silently re-enabling model training on Codex chats— Hacker News

    My company is pretty paranoid about leaking IP to the frontier labs through prompts, so we've all been diligent about disabling data sharing in Codex/Claude Code/etc. But yesterday…

  5. 17.
    US tech giants' hidden AI debt swells 90% to $3.2T in 6 months— Hacker News

    AI 导读 · 科技巨头借债扩张AI，债务半年激增九成，暗藏算力泡沫与财务风险。

  6. 18.
    The Little Book of Generative AI Foundations: An Intuitive Mathematical Primer— Hacker News

    AI 导读 · 用极简数学直觉拆解生成式AI底层原理，适合想真正理解而非只会调包的读者。

论文研究6 条

  1. 19.
    Nearly Linear-Time Follow-the-Perturbed-Leader for Weakly Observable Graph Bandits— arXiv

    We study online learning with directed weakly observable feedback graphs and propose a Follow-the-Perturbed-Leader (FTPL) policy that achieves best-of-both-worlds guarantee with nearly linear expected…

  2. 20.
    FlowTranslator: Training-Free In-Image Machine Translation with Flow-Based Editing— arXiv

    In-image machine translation (IIMT) replaces text in an image with its translation while preserving the surrounding visual context. Existing task-specific models achieve convincing performance but req…

  3. 21.
    OSGR: Observation-Support Graph Reasoning for Earth Observation Question Answering— arXiv

    Answering geographic queries over Earth observation (EO) archives requires solving a joint retrieval-and-derivation problem under strict spatiotemporal constraints. Existing LLM agents fail to ground…

  4. 22.
    SEAM: Situated Executable Associative Memory for a Co-located Community— arXiv

    In groups, knowledge is distributed among individuals, and how it is brought together depends on social ties. Taking inspiration from how a single mind recalls and reacts to memories and current stimu…

  5. 23.
    Reciprocity Can Erase Direction:tructural Boundaries and Controlled Utility of Magnetic Graph Networks— arXiv

    Magnetic graph networks encode direction through phase, but a learned graph can erase the intended directional mechanism before prediction. We characterize this failure for support-based magnetic Lapl…

  6. 24.
    How Risky Is Your Data-Driven Policy?— arXiv

    Data-driven algorithms are ubiquitous across industries, and there is an extensive literature on analyzing their \emph{average} performance. The risk profile of such algorithms, while critical, is muc…

技巧与观点6 条

  1. 25.
    StateBench: Testing how language models track state changes over long documents— HuggingFace Blog
  2. 26.
    Palette: teaching a small model to build decks in your design language— HuggingFace Blog
  3. 27.
    CWE-bench v2: More coverage, quality, and transparency— HuggingFace Blog
  4. 28.
    Transformers Model Integration Contracts— HuggingFace Blog
  5. 29.
    Show HN: HackerTV – HN's most-recommended Youtube videos as 24/7 TV channels— Hacker News

    I saw HackerTube ( https://news.ycombinator.com/item?id=50319510 ) a few days ago, the YouTube videos HN keeps recommending, pulled from 19 years of comments. It made me want to fl…

  6. 30.
    Bring structured and unstructured data together with Amazon Quick— AWS ML

    Amazon Quick brings structured and unstructured data together in one experience. In this post, you connect Amazon S3 as a knowledge base, enrich a dataset, build a Space, and creat…

快讯

日报生成时间：2026-11-07
← 2026-11-03最新日报2026-11-05 →