AAI Search
AI 资讯日报 · 2026-09-09

2026-09-09 · AI 资讯日报

今日新收录 99 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LibreOffice breaks download records after declaring it has no AI features— Hacker News
  2. 2.
    行业动态Tao: Open math problems being non-renewably mined by AI— Hacker News
  3. 3.
    行业动态We Must Return to the Office to Use AI in Person— Hacker News
  4. 4.
    行业动态Kimi K3 (2.8T) at 1 token/s on a MacBook Pro, streamed from four SSDs— Hacker News
  5. 5.
    模型发布/更新Introducing ChatGPT Images 2.5— OpenAI

    ChatGPT Images 2.5 helps turn your ideas, sketches, and reference photos into more personalized, polished images that better reflect your ideas.

  6. 6.
    模型发布/更新On the Navier–Stokes Millennium Prize Problem— OpenAI

    We’re sharing an AI-generated solution to the Navier–Stokes Millennium Prize Problem, including a writeup and a formal proof in Lean.

模型发布/更新6 条

  1. 1.
    Introducing ChatGPT Images 2.5— OpenAI

    ChatGPT Images 2.5 helps turn your ideas, sketches, and reference photos into more personalized, polished images that better reflect your ideas.

  2. 2.
    On the Navier–Stokes Millennium Prize Problem— OpenAI

    We’re sharing an AI-generated solution to the Navier–Stokes Millennium Prize Problem, including a writeup and a formal proof in Lean.

  3. 3.
    Funding grants for new research into AI and teen development— OpenAI

    Apply now for OpenAI’s $5 million grant program supporting independent research on how generative AI affects teen development, well-being, and safety.

  4. 4.
    How GPT-5.6 Sol helps run quantum computing experiments— OpenAI

    See how an MIT researcher uses GPT-5.6 Sol with Codex to autonomously run quantum computing experiments, analyze results, and calibrate qubits.

  5. 5.
    The Work Now Within Reach— OpenAI

    Explore how more capable, affordable AI can expand the work people and businesses can accomplish—and make growth more economical.

  6. 6.
    OpenAI expands initiatives to support journalism from classrooms to newsrooms— OpenAI

    OpenAI is expanding support for journalism with tools, training, and partnerships for students, educators, journalists, and news organizations.

产品发布/更新6 条

  1. 7.
    Amazon SageMaker Feature Store introduces UpdateRecord for feature-level writes— AWS ML

    AI 导读 · 细粒度更新减少读写开销，提升实时特征管理效率，对ML流水线优化意义重大。

  2. 8.
    Sonos 27 音频系统更新：配对增强环绕声道、预览外部 AI 接入等— IT之家

    AI 导读 · 智能音响跨界AI，音频生态迎来更多玩法升级。

  3. 9.
    苹果 2026 款 Apple TV 4K 曝光：升级 A19/Pro 芯片，支持 Siri AI— IT之家

    AI 导读 · A19芯片加持，Apple TV或补齐AI短板，成家庭智能中枢关键棋子。

  4. 10.
    扎克伯格推进“个人超级智能”：Meta 发布其首款个人智能体 Muse，内部测试喜忧参半— IT之家

    AI 导读 · 个人智能体落地关键在数据权限治理，成败或影响AI代理商用前景。

  5. 11.
    王云鹤创业后交出首个模型— 量子位

    把多模型执行经验用到了模型训练

  6. 12.
    派早报：华为举办 HarmonyOS 7 | HUAWEI Mate XT 2 及全场景新品发布会等— 少数派

    少数派的近期动态全新iPhone发布在即，来与少数派一起看Apple发布会口袋先知新版本1.3.3上线，你可以自定义任何你想展示的屏幕效果。了解更多能让AI助手通过自然语言指令直接与您的Quote/0 ... 查看全文

行业动态6 条

  1. 13.
    LibreOffice breaks download records after declaring it has no AI features— Hacker News
  2. 14.
    Tao: Open math problems being non-renewably mined by AI— Hacker News

    AI 导读 · AI正将数学难题当矿藏开采，资源枯竭警示科研模式变革。

  3. 15.
    We Must Return to the Office to Use AI in Person— Hacker News
  4. 16.
    Kimi K3 (2.8T) at 1 token/s on a MacBook Pro, streamed from four SSDs— Hacker News

    AI 导读 · 本地跑大模型突破显存瓶颈，SSD直读新思路，性能代价成看点。

  5. 17.
    Hackers are stealing Claude tokens from subscribers— TechCrunch

    AI 导读 · 订阅账户被黑暴露AI服务安全短板，用户需警惕凭证泄露风险。

  6. 18.
    Cognition hits $48B valuation, signaling investors believe AI coding is far from a winner-take-all market— TechCrunch

    AI 导读 · AI编程市场多点开花，资本押注非垄断格局。

论文研究6 条

  1. 19.
    Unlocking Lossless Speedups in LLMs via Discrete Diffusion— HuggingFace Papers

    Large Language Models (LLMs) owe much of their success to next-token prediction (NTP), but their autoregressive (AR) structure requires slow, sequential token generation. To overcome this bottleneck,…

  2. 20.
    FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning Experience— HuggingFace Papers

    A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers provide reliable yet sparse supervision, while dense same-model guidance can reinfor…

  3. 21.
    ENEAS: Embedding-guided Neural Ensemble for Adaptive Segmentation— HuggingFace Papers

    We present ENEAS, a unified, text-promptable method for instance tracking and semantic discovery. Text-promptable segmentation models, including the latest foundation models such as SAM 3, still suffe…

  4. 22.
    OpenAI researcher allegedly pressured mathematician to drop Anthropic co-author from math breakthrough paper— The Decoder

    AI 导读 · 学术霸凌争议升级，揭露AI研究合作中的权力失衡与利益冲突。

  5. 23.
    Causal Foundation Models— HuggingFace Papers

    Causal inference is the practice of estimating the effect of a treatment or intervention from data. It traditionally requires a bespoke pipeline for every new problem: first proposing a causal mechani…

  6. 24.
    Unfold The World: Factorize 4D Properties in Reinforcing Spatial Reasoning— HuggingFace Papers

    Despite the remarkable prowess of Vision-Language Models (VLMs) in general multimodal tasks, they remain fundamentally ``flat'' when reasoning about the physical world. We argue that this spatial bott…

技巧与观点6 条

  1. 25.
    Pathway’s brain-inspired architecture development on Amazon SageMaker HyperPod— AWS ML

    AI 导读 · 脑启发架构跳出思维链范式，在潜在空间推理，或为AI效率突破提供新路径。

  2. 26.
    How HPE Zerto built an agentic troubleshooting system with Amazon Bedrock— AWS ML

    HPE Zerto built an agentic troubleshooting system powered by Amazon Bedrock that runs on-premises inside the customer environment. This post describes the multi-agent architecture,…

  3. 27.
    How DiDi built intelligent contact center QA with Amazon Bedrock— AWS ML

    DiDi built a transparent, self-owned contact center quality assurance (QA) system on Amazon Bedrock, replacing an opaque third-party tool. Intent verification accuracy rose from 38…

  4. 28.
    Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic— HuggingFace Blog
  5. 29.
    Automated agent evaluation with Amazon Bedrock AgentCore and GitHub Actions— AWS ML

    Wire Amazon Bedrock AgentCore Evaluations into a GitHub Actions pipeline: deploy an AI agent and an OAuth-protected MCP server to AgentCore runtime, invoke the agent with test prom…

  6. 30.
    Benchmarking small LLM inference on SageMaker AI: G7 vs G5 and G6— AWS ML

    Benchmark two 30B Mixture-of-Experts models, Qwen3-Coder-30B and NVIDIA Nemotron-3-Nano-30B, across G5, G6, G6e, and G7 GPU instances on Amazon SageMaker AI. Compare throughput, la…

快讯

日报生成时间：2026-09-28
← 2026-09-08最新日报2026-09-10 →