NLNEXT LAB RADAR
AI 资讯日报 · 2026-08-12

2026-08-12 · AI 资讯日报

今日新收录 148 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.
    行业动态AI is removing the middle class of software engineering?— Hacker News
  2. 2.
    论文研究Surgical WAM: A World-Action Model for Data-Efficient Surgical Robot Learning— arXiv

    Learning reliable surgical manipulation policies is bottlenecked by the scarcity of action-labeled demonstrations: teleoperated surgical robot (e.g., dVRK) trajectories with synchronized kinematics ar…

  3. 3.
    论文研究ConVAWG: A Retrieval-Grounded Framework for Controlled Synthetic Dialogue Generation in Violence Against Women and Girls— arXiv

    Synthetic dialogue generation offers a way to study conversational dynamics in sensitive domains where real data are difficult to access, release, or annotate. The underlying abuse may occur online or…

  4. 4.
    论文研究Beyond a Bag of Features: Set-Level Instability in Sparse Autoencoders— arXiv

    Shani et al. (2026) show that LLM representations broadly recover human category boundaries, while failing to reflect fine-grained typicality structure. Their analysis uses cosine similarity over dens…

  5. 5.
    论文研究Long-Horizon AI Research for Grothendieck Constant: A Case Study in Human-AI Mathematical Collaboration— arXiv

    AI agents are increasingly used in mathematics research, but it is often unclear how to use them effectively. Towards this, we present an extensive case study of how AI was used to improve bounds on t…

  6. 6.
    论文研究Test-Time Self-Evolving GUI Visual Grounding via Reflection-Guided On-Policy Self-Distillation— arXiv

    GUI Visual Grounding is a fundamental capability for GUI agents. Existing models typically freeze their parameters after deployment, limiting their ability to adapt to unseen interfaces. Although rece…

模型发布/更新6 条

  1. 1.
    How well do job retraining programs work?— Anthropic

    AI 导读 · 再培训成效存疑，AI时代就业政策或需根本性反思。

  2. 2.
    DeepSeek V4 Pro 0813— Hacker News

    AI 导读 · 模型迭代速度惊人，性能或再破纪录，值得开发者紧盯。

  3. 3.
    Putting sign language AI into users’ hands— Google DeepMind

    Introducing sign-language-to-text (SL2T), our breakthrough model powering new sign language features for Deaf and hard of hearing users.

  4. 4.
    NVIDIA AI Factory Compute Is Becoming an Investable Asset Class— NVIDIA

    We announced partnerships with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR to establish independent financing platforms designed to mobilize over $500 billion…

  5. 5.
    Claude for Business Development Representatives— Anthropic

    Claude for Business Development Representatives Anthropic

  6. 6.
    Grok is now an AI ‘teammate’ you can assign work— The Verge

    SpaceXAI has introduced Grok Bot, an always-on AI agent service designed to behave like independent "AI teammates" that can do your work for you. The bots share their own cloud-bas…

产品发布/更新6 条

  1. 7.
    fuxicodex/Fuxi— GitHub

    FuXi is a fast, self-contained AI coding agent that lives in your terminal — edit code, run commands, and drive tools, with cost-aware routing across LLM provid…

  2. 8.
    shixinnt/codex-image-context-runtime— GitHub

    A Codex plugin and local MCP runtime for context-bounded image generation and inspection.

  3. 9.
    Launch HN: Keet (YC S24) – An app to create video courses on anything— Hacker News

    Hi HN! We’re Zack and Tommy the Co-Founders of Keet ( https://trykeet.com ). We are building a mobile app that generates courses on any topic, with short videos for explanation and…

  4. 10.
    全球首款机器人手机荣耀 Robot Phone 正式发布，天马宣布独家供屏— IT之家

    IT之家 8 月 12 日消息，在今晚的荣耀 Robot Phone 全球新品发布会上， 全球首款机器人手机 —— 荣耀 Robot Phone 正式发布 ， 定价 9999 元起 。 天马微电子宣布， 为本款机型独家供应屏幕，深度定制天马天工屏高端 OLED 显示方案 。天马表示，这是天工屏首次应用于具备具身交互能力的 AI 终端，适配机器人手机的创作、随…

  5. 11.
    国产具身智能创全球新纪录！以30%成本跑赢 Figure AI 45%效率，聪明的具身大脑成关键— 量子位

    具身模型一小时狂拣1816件异形包裹

  6. 12.
    紫东太初推出GMC核心集剪枝方法，少80%Token仍满血保真多模态能力— 量子位

    免训练、开箱即用！

行业动态6 条

  1. 13.
    AI is removing the middle class of software engineering?— Hacker News
  2. 14.
    Grok 4.6— Hacker News

    https://artificialanalysis.ai/articles/grok-4-6-benchmarks-a...

  3. 15.
    Grok 4.6 scores 61 on the Artificial Analysis Intelligence Index— Hacker News

    AI 导读 · 超越GPT-4o的硬核成绩，AI评测新标杆诞生。

  4. 16.
    Company Offering '100% Human-Written, Never AI' Medical Research Is 100% AI— Hacker News
  5. 17.
    Someone is running mass vulnerability scans, spoofing AI bots like ClaudeBot— Hacker News
  6. 18.
    llama.cpp— Hacker News

论文研究6 条

  1. 19.
    Surgical WAM: A World-Action Model for Data-Efficient Surgical Robot Learning— arXiv

    Learning reliable surgical manipulation policies is bottlenecked by the scarcity of action-labeled demonstrations: teleoperated surgical robot (e.g., dVRK) trajectories with synchronized kinematics ar…

  2. 20.
    ConVAWG: A Retrieval-Grounded Framework for Controlled Synthetic Dialogue Generation in Violence Against Women and Girls— arXiv

    Synthetic dialogue generation offers a way to study conversational dynamics in sensitive domains where real data are difficult to access, release, or annotate. The underlying abuse may occur online or…

  3. 21.
    Beyond a Bag of Features: Set-Level Instability in Sparse Autoencoders— arXiv

    Shani et al. (2026) show that LLM representations broadly recover human category boundaries, while failing to reflect fine-grained typicality structure. Their analysis uses cosine similarity over dens…

  4. 22.
    Long-Horizon AI Research for Grothendieck Constant: A Case Study in Human-AI Mathematical Collaboration— arXiv

    AI agents are increasingly used in mathematics research, but it is often unclear how to use them effectively. Towards this, we present an extensive case study of how AI was used to improve bounds on t…

  5. 23.
    Test-Time Self-Evolving GUI Visual Grounding via Reflection-Guided On-Policy Self-Distillation— arXiv

    GUI Visual Grounding is a fundamental capability for GUI agents. Existing models typically freeze their parameters after deployment, limiting their ability to adapt to unseen interfaces. Although rece…

  6. 24.
    How to Verify Consistency of Probabilistic Claims— arXiv

    When a probabilistic predictor answers many conditional-probability queries, are its answers self-consistent, and can this be verified in polynomial time? This problem is of interest for AI safety, wh…

技巧与观点6 条

  1. 25.
    Introducing OlmoEarth embeddings: Custom embedding exports from OlmoEarth Studio for downstream analysis— HuggingFace Blog

    AI 导读 · 开放定制嵌入导出，打通地理空间数据与下游分析链路。

  2. 26.
    LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge— HuggingFace Blog
  3. 27.
    The White House Is Going to Expand Its AI Policy— Wired

    AI 导读 · 白宫转向开源模型监管，预示AI政策风向生变，值得关注。

  4. 28.
    Part 2: Amazon Bedrock cost attribution with Amazon Athena and CUDOS— AWS ML

    AI 导读 · 打通成本归因与可视化，让AI支出透明可控，是企业上云精细化管理的关键一步。

  5. 29.
    Quoting Florian Herrengt— Simon Willison

    But then users start to report a weird bug. It's the 4th time your team has been trying to fix it. I mean... asking AI to fix it. Unfortunately, it seems like not even Fable can fi…

  6. 30.
    How OneAdvanced deployed over 50 AI agents on UK-sovereign AWS— AWS ML

    Learn how OneAdvanced, a UK enterprise software provider, built a UK-sovereign AI platform by self-hosting Llama 4 Maverick and Llama Guard 4 on Amazon SageMaker AI, with a RAG pip…

快讯

日报生成时间：2026-08-16
← 2026-08-11最新日报2026-08-13 →