每日简报

2026-10-03

← 历史归档

Panniantong/Agent-Reach

Python · ★ 88,625 · 🍴 7,800 · 📈 696 stars today

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.

中文介绍 为 AI Agent 提供跨平台联网能力,通过一个 CLI 即可读取和搜索 Twitter、Reddit、YouTube、GitHub、Bilibili、小红书等平台内容,无需申请 API 接口或支付调用费用,适合需要实时获取全网信息的自动化任务。

JuliusBrussee/caveman

Go · ★ 109,102 · 🍴 6,315 · 📈 209 stars today

🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.

中文介绍 面向编程 Agent 的 Token 压缩技能,通过让 Agent 以极简"穴居人式"语言风格输出,将 Token 消耗降低约 65%。基于代理机制接入各类 AI 编码助手,适合对推理成本敏感的批量开发场景。

obra/superpowers

Shell · ★ 294,460 · 🍴 26,318 · 📈 556 stars today

An agentic skills framework & software development methodology that works.

中文介绍 一套面向 AI Agent 的技能框架与软件开发方法论,将软件工程最佳实践封装为 Agent 可调用的能力模块,引导 Agent 按结构化流程完成开发任务。适合希望提升 AI 辅助开发可靠性的开发者。

DietrichGebert/ponytail

JavaScript · ★ 151,800 · 🍴 8,141 · 📈 1,435 stars today

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

中文介绍 编程 Agent 行为增强技能,让 Agent 以"最懒资深工程师"思维工作,优先选择不写代码的最优解。倡导最小变更原则,减少冗余输出,适合追求代码简洁性和维护性的日常开发场景。

pbakaus/impeccable

JavaScript · ★ 74,323 · 🍴 4,480 · 📈 722 stars today

The design language that makes your AI harness better at design.

中文介绍 一套设计语言规范,专门为提升 AI Agent 的设计产出质量而构建。通过结构化设计约束和最佳实践,使 Agent 在 UI/UX 生成任务中产出更专业、一致的设计方案,适用于前端和视觉开发工作流。

mattpocock/skills

Shell · ★ 274,704 · 🍴 23,063 · 📈 955 stars today

Skills for Real Engineers. Straight from my .agents directory.

中文介绍 由知名开发者 Matt Pocock 整理的实用工程技能集,源自其个人 .agents 目录,涵盖真实工程中常用的高价值技能。适合追求生产级开发质量的 AI 编程助手用户。

NVIDIA/OpenShell

Rust · ★ 14,431 · 🍴 1,660 · 📈 594 stars today

OpenShell is the safe, private runtime for autonomous AI agents.

中文介绍 NVIDIA 推出的自主 AI Agent 安全运行时环境,提供隔离、隐私保护的执行沙箱。确保 Agent 在隔离环境中安全运行,防止数据泄露和未授权操作,适合企业级 Agent 部署场景。

coreyhaines31/marketingskills

JavaScript · ★ 52,415 · 🍴 7,879 · 📈 140 stars today

Marketing skills for Claude Code and AI agents. CRO, copywriting, SEO, analytics, and growth engineering.

中文介绍 面向 Claude Code 等 AI Agent 的营销技能包,覆盖 CRO 转化优化、文案撰写、SEO、数据分析与增长工程等核心营销领域。适合需要 AI 辅助完成营销内容的运营和增长团队。

heygen-com/hyperframes

TypeScript · ★ 55,899 · 🍴 5,038 · 📈 580 stars today

Write HTML. Render video. Built for agents.

中文介绍 允许通过编写 HTML 直接渲染生成视频的工具,专为 AI Agent 设计接口。开发者可让 Agent 自动生成 HTML 页面并转换为视频输出,适用于自动化视频制作和 AI 驱动的内容生成流水线。

mksglu/context-mode

TypeScript · ★ 25,035 · 🍴 1,798 · 📈 282 stars today

Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory, and enforces routing across 17 platforms via MCP + hooks.

中文介绍 AI 编程 Agent 的上下文窗口优化工具,通过沙箱化工具输出将上下文消耗降低 98%,持久化会话记忆,并通过 MCP 和 hooks 机制在 17 个平台上强制路由,解决长会话中 Token 膨胀和上下文溢出问题。

google/skills

Python · ★ 20,747 · 🍴 1,721 · 📈 39 stars today

Agent Skills for Google products and technologies

中文介绍 Google 官方为 Google 产品和技术生态提供的 Agent 技能集,涵盖其各类开发工具和平台的能力封装。使用 Google 技术栈的开发者可通过此技能包让 AI Agent 更好地操作 Google 服务。

getsentry/sentry

Python · ★ 45,031 · 🍴 4,891 · 📈 16 stars today

Developer-first error tracking and performance monitoring

colbymchenry/codegraph

C · ★ 72,960 · 🍴 4,682 · 📈 98 stars today

Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, CoPilot, and Hermes Agent — fewer tokens, fewer tool calls, 100% local

中文介绍 预索引代码知识图谱工具,代码变更时自动同步更新索引,兼容 Claude Code、Codex、Gemini、Cursor、CoPilot 等主流 AI 编码助手。通过知识图谱减少 Token 消耗和工具调用次数,实现 100% 本地运行。

cursor/plugins

TypeScript · ★ 9,502 · 🍴 897 · 📈 163 stars today

Cursor plugin specification and official plugins

中文介绍 Cursor 编辑器的插件规范定义与官方插件集合,为开发者提供标准化的插件开发框架和官方实现参考。是 Cursor 插件生态的基础设施项目,插件开发者可基于此规范扩展编辑器功能。

mvschwarz/openrig

TypeScript · ★ 4,307 · 🍴 293 · 📈 683 stars today

Build your own network of agents from Claude Code, Codex and Pi: persistent teams with roles, shared context and owned work.

中文介绍 多 Agent 协作编排框架,可将 Claude Code、Codex、Pi 等 Agent 构建为具有持久角色、共享上下文和独立工作归属的团队网络。适合需要多 Agent 分工协作完成复杂开发任务的场景。

Effect-TS/effect

TypeScript · ★ 16,547 · 🍴 802 · 📈 80 stars today

Build production-ready applications in TypeScript

中文介绍 面向生产级 TypeScript 应用的函数式编程框架,提供强大的错误处理、依赖注入、并发控制和副作用管理机制。适合构建高可靠性、可测试的 Node.js 服务端应用。

pablostanley/yoinks

TypeScript · ★ 3,500 · 🍴 310 · 📈 623 stars today

yoink any video from your terminal. no shady ads.

中文介绍 终端命令行视频下载工具,支持从各大平台快速下载视频,无广告和付费墙干扰。适合开发者和技术用户在终端环境下高效获取视频资源,注重隐私和简洁体验。

OneStreamer: Unifying Perception, Memory, and Proactive Response in Streaming Video Interaction

👍 146

Streaming video LLMs must retain evidence before its relevance to future tasks is known and respond when sufficient evidence becomes available. The challenge is to form reusable factual memory without compromising real-time perception. We introduce OneStreamer, which jointly learns query-independent

中文介绍 针对流式视频大模型需边感知边记忆且即时响应的问题,提出 OneStreamer,联合学习查询无关的可复用事实记忆与实时感知,在保留证据不损害低延迟响应的同时支持主动响应。

On-Policy or Off-Policy Learning? A Systematic Study of Distillation Dynamics

👍 114

On-policy learning has been argued to reduce catastrophic forgetting, produce sparser parameter updates, and improve generalisation. However, existing comparisons between supervised fine-tuning and reinforcement learning vary many factors simultaneously, making the contribution of rollout policy dif

中文介绍 系统研究蒸馏动力学中 On-Policy 与 Off-Policy 学习差异,指出以往 SFT 与 RL 对比同时改变过多因素,难以分离 rollout 策略影响,本文通过控制变量厘清其各自贡献。

E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models

👍 49

Masked diffusion models (MDMs) generate sequences by progressively unmasking several tokens per denoising step, but their reverse process is typically factorized over positions, limiting sample quality in the few-step regime where diffusion's speed advantage over autoregressive decoding matters most

中文介绍 针对 masked diffusion models 反过程按位置分解导致少步生成质量受限的问题,提出 E-MoE,以增强混合专家打破因子化约束,提升扩散语言模型在少步推理下的样本质量。

Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL

👍 42

Multi-reward reinforcement learning trains large language models to satisfy multiple behavioral objectives simultaneously. Reward-wise normalization, as used in GDPO, preserves reward-specific relative information within rollout groups, but different objectives can still exhibit uneven learning prog

中文介绍 多奖励强化学习训练中各目标学习进度不均。本文提出密度感知奖励聚合方法,对稀疏与密集奖励区别处理,使多条行为目标均衡优化,提升 LLM 多目标对齐效果。

Decentralized Master-Mind: Joint Action Refinement through Iterative Intent Denoising in Multi-Agent Pathfinding

👍 36

Decentralized multi-agent path finding (MAPF) with communication requires agents to reach individual goals without collisions under partial observability. Learnable policies trained on expert data provide an effective approach to this problem. However, when several coordinated joint actions are vali

中文介绍 针对去中心化多智能体路径规划中多组可行联合动作难以选定问题,提出 Decentralized Master-Mind,通过迭代意图去噪细化联合动作,提升通信受限下协作成功率。

InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation

👍 36

We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills, improving from its own attempts, and retaining what it learns, without retraining. Our key insight is that a broad controller already holds much of t

中文介绍 面向人形机器人 loco-manipulation,提出测试时进化框架 InterEvolve,复用已有控制器技能解决未训练任务,从自身尝试中改进并保留经验,无需重新训练即可提升成功率。

EgoTools: Towards Tool-Centric Reasoning in Real-World Egocentric Videos

👍 34

Real-world embodied tasks, from everyday activities to professional procedures, require agents to act under physical constraints while tracking evolving object and task states. Tool use sits at the heart of such tasks, as many everyday and professional activities are tool-mediated. Understanding the

中文介绍 面向真实第一人称视频中工具使用场景,提出 EgoTools,聚焦工具中心的推理任务,帮助智能体在物理约束下跟踪工具与任务状态变化,提升具身工具使用理解能力。

Fewer Tokens, Better Action: GPT-6 Astra Robot Agents with 14% Higher Success Rate but 65% Fewer Tokens

👍 24

Vision language model (VLM) agents can control robots through visual feedback and action primitives, but repeated model invocations and redundant observations incur substantial token overhead. We introduce PyRUA-Lean, an interactive code-execution framework that couples feedback-driven primitive com

中文介绍 VLM 机器人智能体因反复调用与冗余观测造成 token 浪费。提出 PyRUA-Lean 交互式代码执行框架,将反馈驱动的原始动作组合精简,成功率提升 14%,token 减少 65%。

Decoding Looped Transformers Better for (Almost) Free

👍 18

Looped Transformers achieve parameter efficiency by repeatedly executing a shared block across recurrent loops. Each loop yields an intermediate representation decodable for the same next token, yet standard decoding discards earlier states. Because earlier loops embody less computation, recurrence

中文介绍 Looped Transformers 通过循环复用共享模块实现参数高效,但标准解码丢弃早期循环中间表示。本文利用早期循环计算量更少、信息更粗的特点改进解码,几乎无额外成本提升效果。

SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation

👍 18

High-resolution 3D generation increasingly relies on voxel latents and multi-stage pipelines that first predict active structure and then synthesize local geometry. While effective, this design fragments continuous surfaces into many local tokens, inflates generation cost, and often weakens topologi

中文介绍 高分辨率 3D 生成中体素潜在将连续表面碎片化,增加成本并破坏拓扑。提出 SILSA 滑动窗口切片潜在,在保持拓扑的前提下提升生成效率与连续性,适用于高分辨率 3D 合成。

Multimodal Flow: Unified Flow Modeling of Language and Vision in Embedding Spaces

👍 17

We present Multimodal Flow, a fully continuous generative model of language and vision. Most unified multimodal models either model both language and quantized images as discrete tokens or combine discrete language prediction with continuous image generation. The former introduces a visual quantizat

中文介绍 提出 Multimodal Flow,将语言与视觉统一建模于连续嵌入空间的生成模型,避免离散化图像带来的视觉量化误差,实现语言与视觉的全连续联合生成。

Argo-Bench: Evaluating Data Agents on Enterprise-Scale Workflows

👍 17

Real-world enterprise data science and analytics workflows require reasoning across dozens of tables, performing statistical analyses, and acting on the results. Established text-to-SQL benchmarks evaluate query generation alone, and audits have found their answer keys frequently wrong. Because real

中文介绍 现有 text-to-SQL 基准仅评查询生成且答案键常有误。提出 Argo-Bench,评测数据智能体在企业级流程中的多表推理、统计分析与结果执行能力,覆盖真实数据科学工作流。

CorrGRPO: Correlation-Normalized GRPO for Multi-Reward Learning

👍 16

Group Relative Policy Optimization (GRPO) is widely used to train reasoning language models, where it computes advantages by centering and normalizing rewards across rollouts of the same prompt. For multiple rewards, GRPO sums the reward components and normalizes the total reward by its within-group

中文介绍 GRPO 在多奖励场景下将各奖励求和后组内归一化,会混淆不同奖励相关性。提出 CorrGRPO,按相关性归一化处理多奖励,更公平地估计各奖励优势,改善多目标推理模型训练。

X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization

👍 12

Multi-step agents are trained on flat action streams: SFT and RLVR weight every token uniformly and ignore the sub-procedures that recur across tasks, the hierarchy that lets humans plan top-down from reusable routines. This structure sits unused, and flat training uses each scarce trajectory less f

中文介绍 多步智能体训练通常对扁平动作流均匀加权,忽视跨任务复用的子流程层级。提出 X-Tree,将可复用经验结构化为树,让 SFT/RLVR 高效利用每条轨迹并支持自上而下规划泛化。

PhysVista: Benchmarking Physical Intelligence in VLMs via a Perception-Reasoning-Assessment Loop

👍 12

Vision-Language Models (VLMs) have shown strong multimodal reasoning capabilities, yet whether they truly capture the physical consistency underlying real-world dynamics remains unclear. Existing benchmark paradigms often suffer from fragmented evaluation, focusing on isolated cognitive stages while

中文介绍 针对 VLM 是否真正掌握物理一致性存疑的问题,提出 PhysVista 基准,通过感知-推理-评估闭环考察物理智能,避免孤立认知阶段评测的碎片化缺陷。

Architect-Ant: Editable Automatic Furnishing of Architectural Floor Plans

👍 11

Furnished floor plans support real-estate visualization, interior design, and architectural workflows, yet automatic furnishing remains challenged by limited real-world data and the need to satisfy interacting geometric and functional constraints. We ask whether professional furnishing knowledge can

中文介绍 面向建筑平面图自动陈设,受真实数据有限与几何功能约束交织困扰。提出 Architect-Ant,将专业陈设知识编码为可编辑生成流程,在满足约束下自动生成可修改的家具布局。

Omni-Embed-Mini: Binding Modalities Without Forgetting via Dense Distillation

👍 10

Extending a text embedding model to new modalities typically degrades text retrieval quality, and existing omni-modal embedders compensate with multi-billion parameters. We present Omni-Embed-Mini, a 0.9B-parameter model that maps text, speech, audio, images, video, and visually-rich documents into

中文介绍 将文本嵌入模型扩展到多模态常损害文本检索质量。提出 0.9B 的 Omni-Embed-Mini,通过密集蒸馏统一文本、语音、图像、视频与富视觉文档到同一空间,扩展模态不遗忘原能力。

Where-OPD: Spatially Guided On-Policy Self-Distillation of MLLMs with Synthetic Scenes

👍 7

On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supervising students with a frozen or EMA version of themselves that receives privileged information. Its application to multimodal large language models (MLLMs), however, remains larg

中文介绍 将 on-policy 自蒸馏扩展到多模态大模型场景,提出 Where-OPD,用合成场景提供空间引导信息作为特权监督,提升 MLLM 在多步视觉推理中的表现。

Better Supervision Is Nearby: Neighborhood On-Policy Self-Distillation

👍 7

On-policy self-distillation (OPSD) trains mathematical reasoning models using a privileged teacher that sees a reference solution and supervises student-sampled prefixes. Standard OPSD uses one fixed parameter setting at every state, but nearby settings may offer additional supervision. We find that

中文介绍 标准 on-policy 自蒸馏在每个状态仅用固定参数教师。本文发现邻近参数设置可提供额外监督,提出 Neighborhood OPSD,聚合近邻教师信号增强数学推理模型训练效果。

Benchmarking and Enhancing Skill-Level Memory for Partially Observable Robotic Manipulation

👍 7

Recent advances in robot learning have enabled manipulation policies to perform increasingly diverse tasks and generalize across environments. However, reliable execution often depends on hidden task states that cannot be determined from current observations alone, making interaction history essenti

中文介绍 机器人操作策略在部分可观测环境下依赖历史隐状态。本文构建技能级记忆基准并改进机制,使策略能利用交互历史记忆,提升多任务、跨环境操作的可靠性。

The Rise of the Automation Engineer

@jakeserval · 1.1K 粉丝 · 406.0K 阅 · 505 赞 · 1.1K 转

There is a lot of talk about the jobs AI takes away and very little about the ones it creates. While pessimism might be easier (and certainly gets more clicks), I’m more optimistic. From where I sit,

中文介绍 探讨 AI 催生的新兴岗位「自动化工程师」,作者从从业者视角反驳「AI 只消灭工作」的悲观叙事,认为真正懂 prompt 工程与流程编排的人会获得更多价值,是 AI 时代的新核心角色。

Why LLMs can't write well and how Sherpa fixes it

@RohanNayak2 · 13.7K 粉丝 · 230.7K 阅 · 514 赞 · 142 转

As we pivoted to AI-written shows, countless people told us this would kill Pocket FM. After six long months, I almost believed them. But then our ecosystem exploded, and not despite AI writing but

中文介绍 Pocket FM 负责人披露:转向 AI 写有声书后历经 6 个月低谷,生态系统反而爆发增长。提出 Sherpa 方案解决 LLM 写作平庸问题,证明 AI 批量创作可显著提升内容供给与质量。

Safeguarding the AI Era of Biology: Watermarking the Building Blocks of Life

@pushmeet · 29.0K 粉丝 · 209.0K 阅 · 502 赞 · 88 转

Today, we announced SynthID Bio, a method to watermark and detect AI-designed proteins, the building blocks of life. We believe this is the first time humanity has successfully synthesised AI-designed

中文介绍 Google 发布 SynthID Bio,首次在 AI 设计蛋白质中嵌入可检测水印,用于标记与溯源 AI 生成生物分子。这是人类首次成功对 AI 设计生命基础单元做合成后验证,涉及合成生物学安全新边界。

History Rhymes: Large Language Models Off to a Bad Start?

@michaeljburry · 2.1M 粉丝 · 120.2K 阅 · 519 赞 · 84 转

This 145-year-old case study – presented at the Smithsonian Institute no less - provides a potentially devastating critique of today’s Large Language Models and the spending behind them. Michael Burry

中文介绍 Michael Burry 分享一份 145 年前在美国史密森尼学会发表的案例研究,认为其对当时计算方法的批评,映射到今天 LLM 的投入产出问题,质疑当前大模型烧钱模式的可持续性。

here's how to make viral videos with Seedance 2.5

@0xROAS · 54.6K 粉丝 · 118.6K 阅 · 502 赞 · 56 转

Seedance 2.5 is one of the craziest "nano banana moments" of AI VIDEO in 2026. It's the model behind every viral video that you are seeing right now... and it's pretty simple to generate those video,

Own your AI, don’t let it own you

@0xSigil · 56.0K 粉丝 · 79.1K 阅 · 509 赞 · 72 转

6 months ago, the #1 AI was Claude Opus 4.6. Our model, Underdog 27B beats it while running on your Mac or iPhone. Most AI will run on your device, not in data centers. Intelligence that recently

中文介绍 Underdog 27B 模型发布:声称在本地设备(Mac/iPhone)即可运行,性能超越半年前榜首 Claude Opus 4.6,主打端侧推理,让 AI 智能从数据中心回归个人设备,强调数据主权。

We can and must solve alignment

@eric_ho · 4.4K 粉丝 · 66.2K 阅 · 505 赞 · 72 转

In San Francisco, it feels like the eve of the singularity. Yet, a walk through the city looks surprisingly mundane, with driverless cars and rolling fog equal parts of the scenery. The world feels

中文介绍 Anthropic 研究员 Eric Ho 撰文谈 AI 对齐问题,以旧金山街头景象类比当前 AI 发展状态——看似奇点已至,实则表面平常。主张对齐是可解的,行业必须正视而非回避。

how to build agentic systems for knowledge work

@arscontexta · 13.6K 粉丝 · 41.4K 阅 · 510 赞 · 45 转

the goal is to build software around your agents that gives them the context, tools and methods for a particular kind of work @garrytan put it like this: but what are people really trying to achieve

中文介绍 如何为知识工作构建 Agentic 系统的思路分享:核心是给 Agent 配备上下文、工具和方法论,围绕具体工作流封装。引用 Garry Tan 观点,强调 Agent 要服务于人的真实业务目标而非技术炫技。

Why Instinct feels Magical

@yashhsm · 27.2K 粉丝 · 35.7K 阅 · 516 赞 · 24 转

how instinct, a $10b startup works - by reverse-engineering its harness: single-threaded chat, memory, computer use, background tasks etc One month ago, i got an invite for instinct my first

中文介绍 对 $100 亿估值 AI 初创 Instinct 做逆向工程分析,拆解其核心机制:单线程对话、记忆系统、计算机操作、后台任务等模块如何协作,解释其交互为何让用户感觉像「魔法」。

DeepSeek Harness for Desktop Is Here

@DeepSeekHarness · 989 粉丝 · 35.4K 阅 · 1.0K 赞 · 69 转

Today, we’re releasing the DeepSeek Harness v0.2 preview, with a desktop app available for macOS and Windows. You can install it here and follow the instructions to get started. This release comes

中文介绍 DeepSeek Harness v0.2 桌面版发布,支持 macOS 与 Windows,提供本地安装指引。该工具旨在简化 DeepSeek 模型的开发部署流程,降低开发者接入 DeepSeek 生态的门槛。

Creative prompting with Gemini Omni in Google Flow

@FlowbyGoogle · 73.2K 粉丝 · 23.0K 阅 · 539 赞 · 51 转

To help get you started exploring Gemini Omni Flash in Google Flow, here is a comprehensive guide featuring prompt templates and detailed ideas for your next creative generations. Whether you are

中文介绍 Google 官方发布 Gemini Omni Flash 在 Google Flow 中的创意提示词指南,包含多组 prompt 模板与详细创意方向,帮助创作者快速上手 AI 图像生成,覆盖不同风格与应用场景。

September at OptimAI: Building the Next Layer of the Agentic Internet

@OptimaiNetwork · 98.4K 粉丝 · 10.7K 阅 · 513 赞 · 138 转

September marked an important step in OptimAI's evolution. Over the past month, we have been expanding the products people interact with while continuing to strengthen the infrastructure underneath

中文介绍 OptimAI 九月月报:在 Agentic Internet 基础设施层持续迭代,扩展用户侧产品矩阵的同时加固底层架构,定位为企业级 AI Agent 调度与编排平台,强调基础设施可靠性优先。

OpenAI Withheld GPT-6.1 Astra. It Exposed AI’s Permission Problem.

@ActionModelAI · 60.2K 粉丝 · 6.8K 阅 · 501 赞 · 439 转

OpenAI has chosen not to release GPT-6.1 Astra after the model failed to meet its safety standards. To be clear, GPT-6.1 Astra is not GPT-6.1 Sol, the model OpenAI released at DevDay. GPT-6.1 Sol has

中文介绍 爆料 OpenAI 未发布 GPT-6.1 Astra,因该版本未通过内部安全标准。Astra 不同于 DevDay 发布的 GPT-6.1 Sol,事件暴露了前沿 AI 在能力与安全性之间取舍的核心矛盾。

A model guide for the GPT-6 family

Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.

中文介绍 OpenAI发布GPT-6系列模型选择指南,面向初创企业提供模型选型建议,涵盖推理力度调节、提示词与技能优化、工具协同以及生产级工作流准备等实践方法。

Redefining enterprise intelligence with autonomous AI

Enterprise AI is no longer a future ambition. It is in full operational flight. Model capabilities are advancing faster than most organizations can absorb, while the cost of performance continues to fall. Globally, AI investment is set to reach $2.5 trillion in 2026, up 44% from the previous year. F

中文介绍 MIT科技评论报道,企业AI已全面进入落地运营阶段,模型能力提升速度快于组织吸收能力,性能成本持续下降。预计2026年全球AI投资将达2.5万亿美元,同比增长44%。

Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience

After leading Meta’s Llama models, Ahmad Al-Dahle is now transforming Airbnb with AI — from how its teams develop products to how it serves guests.

中文介绍 曾领导Meta Llama系列模型研发的Ahmad Al-Dahle如今在Airbnb推动AI转型,涉及团队产品开发流程优化以及面向旅客的服务体验提升,被称为内部重塑与用户端双管齐下的AI战略。

Don’t be fooled—LLMs don’t reason

On an afternoon in Seoul in March 2016, I watched a program I helped build put a stone on the fifth line of a Go board in what looked like a gift to its human opponent. Move 37 in game two of the five-game match looked so absurd that some commentators thought it was a…

中文介绍 MIT科技评论文章以2016年AlphaGo与人类对弈为例,论述大语言模型并不具备真正的推理能力,提醒读者不要误认为LLM拥有类人逻辑思维。

[AINews] Pi 1.0, Pi Durable, and AIE NYC

the minimalist harness goes stable... and TypeScript!

中文介绍 Latent Space的AINews栏目汇总近期AI动态,涵盖极简框架Pi 1.0稳定版发布、Pi Durable项目进展,以及纽约AIE(AI企业峰会)相关活动信息,其中Pi正式支持TypeScript。

Academia is for Ambition — Alex Zhang, MIT

We catch up with RLM first author Alex Zhang, MIT PhD, on Jev, PhD masxing, and the future of harnesses.

中文介绍 Latent Space专访MIT博士Alex Zhang,讨论其主导的RLM(推理语言模型)项目,内容涉及学术追求、博士生身份认同(PhD masxing)以及AI框架的未来发展方向。

Chatham scales its capital markets expertise with OpenAI

Chatham Financial uses Codex and GPT-5.6 to build technology and redesign workflows, cutting trade validation from 30 minutes to under 4.

中文介绍 英国金融科技公司Chatham采用OpenAI的Codex与GPT-5.6模型优化资本市场监管技术,将交易验证流程从30分钟压缩至4分钟以内,显著提升运营效率。

The eternal complement

Advanced AI may matter most for the routine work behind breakthrough ideas. Explore why execution could shape the next economy and the pace of progress.

中文介绍 OpenAI发布观点文章,探讨高级AI对执行性常规工作的价值,认为推动突破性想法落地所需的执行能力可能对未来经济格局与进步节奏产生深远影响。

How Albertsons Companies is reimagining retail from the inside out

Albertsons Cos. is using ChatGPT Enterprise and the OpenAI API to help teams work faster and make grocery shopping easier for millions of customers.

中文介绍 美国零售巨头Albertsons公司引入ChatGPT Enterprise及OpenAI API,协助内部团队提升工作效率,并通过AI优化超市购物体验,服务数百万顾客。

[AINews] Gemini 4 Argon: GDM’s answer to Astra/Fable, with 1M output

... but you can’t try it yet unless you are “government users and trusted cyber defenders in the Fairwind Program”

中文介绍 Latent Space报道谷歌Gemini 4 Argon模型发布,该模型支持百万级token输出,被视为Google DeepMind对Anthropic Astra及Meta Fable的回应,目前仅向Fairwind计划下的政府用户和网络安全防御者开放。

The Den frees up 10-15 hours a week to grow with ChatGPT Work

As it opens a new location, the social club prepares grant applications in 2 hours instead of 3 days and liquor-license materials in 3 hours instead of 4 days.

中文介绍 社交俱乐部The Den使用ChatGPT Work平台大幅提效,每周节省10至15小时,新建地点的资助申请从3天缩短至2小时,酒类许可证材料从4天压缩至3小时。

市场总览

美股端:SPY 769.64 与 QQQ 749.58 均录得多头排列且逼近 52 周高(SPY -1.25%、QQQ -0.66%),QQQ 的 MACD 红柱扩张至 7.58,RSI 61.5;个股层面 MSFT 触发 MACD 金叉、NVDA 近 5 日 +4.17% 领涨,而 TSLA、GOOGL 因 MACD 死叉转入中性。加密端:恐慌贪婪指数 67(贪婪),总市值 2.89T(24h -3.22%),BTC 主导率 58.7%、ETH 11.3%;BTC 84601 与 ETH 2674、SOL 119 均维持 SMA20/50/200 多头排列,但三者 MACD 同步死叉、绿柱微扩,短线动量收敛。中概端:BABA、PDD、JD、0700.HK 集体走熊,RSI 全部落在 35-41,BABA 与 0700.HK 呈空头排列并触发 MACD 死叉,距 52 周高 40-46%,距 52 周低不足 15%,方向明显偏弱。商品外汇:黄金 GC=F 空头排列、RSI 34.7 接近超卖;原油 CL=F 多头排列但 MACD 死叉、红柱回落;USDCNY=X 报 6.70 触及 52 周低并呈空头排列,美元指数 101.92 与 10Y 美债收益率 5.28% 双双 RSI 超买并逼近 52 周高,美元—长端利率偏强挤压黄金与中概风险资产。

今日关注

MSFT Microsoft (MSFT)
偏上行

现价 517.53,近 1 日 +0.9%、近 5 日 +3.94%,价格站上 SMA20 501.30、SMA50 485.01、SMA200 432.62,呈多头排列;MACD 7.42 高于信号线 7.36 触发金叉,柱状值转红;RSI 14 报 60.2,未超买,动能扩张方向向上。

^TNX 10Y 美债收益率 (^TNX)
偏上行

收益率升至 5.28%,近 1 日 +0.76%、近 5 日 +1.79%,逼近 52 周高(-1.22%);SMA20 5.05、SMA50 4.83、SMA200 4.45 逐级抬升呈多头排列;RSI 14 报 74.6 已进入超买区,MACD 柱状值 0.134 仍在信号线之上,动量偏强但短线过热。

DX-Y.NYB 美元指数 DXY
偏上行

指数报 101.92,近 1 日 -0.17% 但近 5 日 +0.94%,逼近 52 周高(-0.28%);SMA20 100.36、SMA50 100.00、SMA200 99.27 多头排列;RSI 14 达 71.6 超买,MACD 0.579 高于信号 0.388 且红柱扩张,美元短线动能偏强。

0700.HK 腾讯控股 (0700.HK)
偏下行

现价 421.20,近 1 日 -2.27%、近 5 日 -3.92%,距 52 周高 -38.29%、距 52 周低仅 +2.48%;价格低于 SMA20 433.70、SMA50 447.51、SMA200 498.88 呈空头排列;MACD -3.995 低于信号 -3.954 出现死叉,RSI 14 仅 40.6 显示动量偏弱。

GC=F 黄金期货
偏下行

金价报 4172.10,近 1 日 -0.72%、近 5 日 -3.45%,距 52 周高 -25.31%;价格显著低于 SMA20 4334.82、SMA50 4366.47、SMA200 4553.41 呈空头排列;MACD -62.72 远低于信号 -38.47 绿柱放大,RSI 14 仅 34.7 接近超卖但趋势仍偏空。

全部资产

^VIX

VIX 恐慌指数

$15.31 -6.59%
5 日
+2.96%
距 52w 高
-56.6%
RSI(14)
47.6
趋势
空头
SMA 20 / 50 / 200
15.92 / 15.78 / 18.06
MACD / 信号
0.005 / -0.027
MACD 金叉 (3 天前)空头排列

^TNX

10Y 美债收益率 (%)

$5.28 +0.76%
5 日
+1.79%
距 52w 高
-1.2%
RSI(14)
74.6
趋势
多头
SMA 20 / 50 / 200
5.05 / 4.83 / 4.45
MACD / 信号
0.134 / 0.116
RSI 超买接近 52 周高多头排列

DX-Y.NYB

美元指数 DXY

$101.92 -0.17%
5 日
+0.94%
距 52w 高
-0.3%
RSI(14)
71.6
趋势
多头
SMA 20 / 50 / 200
100.36 / 100.00 / 99.27
MACD / 信号
0.579 / 0.388
RSI 超买接近 52 周高多头排列

SPY

S&P 500 ETF

$769.64 +0.92%
5 日
+0.32%
距 52w 高
-1.2%
RSI(14)
49.2
趋势
多头
SMA 20 / 50 / 200
765.00 / 763.07 / 720.03
MACD / 信号
0.536 / 0.907
MACD 死叉 (1 天前)接近 52 周高多头排列

QQQ

Nasdaq 100 ETF

$749.58 +1.33%
5 日
+1.14%
距 52w 高
-0.7%
RSI(14)
61.5
趋势
多头
SMA 20 / 50 / 200
726.18 / 715.66 / 667.91
MACD / 信号
7.583 / 6.058
接近 52 周高多头排列

AAPL

Apple

$333.69 +0.20%
5 日
-0.66%
距 52w 高
-3.4%
RSI(14)
51.4
趋势
多头
SMA 20 / 50 / 200
331.62 / 322.18 / 288.86
MACD / 信号
4.083 / 5.073
MACD 死叉 (2 天前)多头排列

MSFT

Microsoft

$517.53 +0.90%
5 日
+3.94%
距 52w 高
-6.5%
RSI(14)
60.2
趋势
多头
SMA 20 / 50 / 200
501.30 / 485.01 / 432.62
MACD / 信号
7.416 / 7.356
MACD 金叉 (今天)多头排列

NVDA

Nvidia

$233.95 +2.44%
5 日
+4.17%
距 52w 高
-1.6%
RSI(14)
60.1
趋势
多头
SMA 20 / 50 / 200
223.51 / 217.77 / 200.43
MACD / 信号
3.069 / 2.455
接近 52 周高多头排列

GOOGL

Alphabet

$343.50 -0.17%
5 日
+0.33%
距 52w 高
-15.9%
RSI(14)
45.4
趋势
中性
SMA 20 / 50 / 200
342.57 / 343.85 / 338.74
MACD / 信号
-0.648 / -0.360
MACD 死叉 (今天)

TSLA

Tesla

$370.59 +4.45%
5 日
-1.94%
距 52w 高
-25.7%
RSI(14)
45.6
趋势
中性
SMA 20 / 50 / 200
365.15 / 346.57 / 393.71
MACD / 信号
1.278 / 3.707
MACD 死叉 (3 天前)

META

Meta

$728.08 +0.40%
5 日
-6.37%
距 52w 高
-6.6%
RSI(14)
61.1
趋势
中性
SMA 20 / 50 / 200
690.05 / 622.60 / 627.98
MACD / 信号
36.037 / 35.275
加密恐慌贪婪
67
贪婪
加密总市值
$2.89 T
-3.22% / 24h
BTC 主导率
58.7%
ETH 11.3%
24h 成交量
$122.3 B
活跃币 21,783

BTC-USD

Bitcoin

$84,601.66 -0.30%
5 日
+0.17%
距 52w 高
-33.0%
RSI(14)
63.4
趋势
多头
SMA 20 / 50 / 200
82,050.81 / 78,124.99 / 71,371.76
MACD / 信号
1,999.544 / 2,133.722
MACD 死叉 (3 天前)多头排列

ETH-USD

Ethereum

$2,674.42 -1.15%
5 日
-0.47%
距 52w 高
-43.8%
RSI(14)
59.6
趋势
多头
SMA 20 / 50 / 200
2,627.14 / 2,468.08 / 2,114.00
MACD / 信号
70.702 / 81.077
MACD 死叉 (4 天前)多头排列

SOL-USD

Solana

$119.05 +0.55%
5 日
-2.46%
距 52w 高
-49.8%
RSI(14)
63.0
趋势
多头
SMA 20 / 50 / 200
113.08 / 102.89 / 85.67
MACD / 信号
5.234 / 5.542
MACD 死叉 (1 天前)多头排列

BABA

阿里巴巴 (BABA)

$105.85 -1.57%
5 日
-4.32%
距 52w 高
-44.6%
RSI(14)
38.4
趋势
空头
SMA 20 / 50 / 200
110.37 / 117.04 / 131.16
MACD / 信号
-2.239 / -2.132
MACD 死叉 (1 天前)空头排列

PDD

拼多多 (PDD)

$75.38 -3.28%
5 日
-3.61%
距 52w 高
-45.9%
RSI(14)
36.1
趋势
空头
SMA 20 / 50 / 200
78.79 / 83.93 / 94.68
MACD / 信号
-1.767 / -1.832
空头排列

JD

京东 (JD)

$25.78 -3.08%
5 日
-3.63%
距 52w 高
-28.8%
RSI(14)
35.3
趋势
中性
SMA 20 / 50 / 200
27.03 / 29.09 / 28.90
MACD / 信号
-0.627 / -0.660

0700.HK

腾讯控股 (0700.HK)

HK$421.20 -2.27%
5 日
-3.92%
距 52w 高
-38.3%
RSI(14)
40.6
趋势
空头
SMA 20 / 50 / 200
433.70 / 447.51 / 498.88
MACD / 信号
-3.995 / -3.954
MACD 死叉 (今天)接近 52 周低空头排列

GC=F

黄金期货

$4,172.10 -0.72%
5 日
-3.45%
距 52w 高
-25.3%
RSI(14)
34.7
趋势
空头
SMA 20 / 50 / 200
4,334.82 / 4,366.47 / 4,553.41
MACD / 信号
-62.715 / -38.467
空头排列

CL=F

WTI 原油期货

$91.26 -1.73%
5 日
-1.24%
距 52w 高
-23.6%
RSI(14)
47.7
趋势
多头
SMA 20 / 50 / 200
96.05 / 88.58 / 82.39
MACD / 信号
0.564 / 1.823
多头排列

USDCNY=X

美元 / 人民币

¥6.70 -0.09%
5 日
-0.21%
距 52w 高
-6.8%
RSI(14)
34.5
趋势
空头
SMA 20 / 50 / 200
6.71 / 6.73 / 6.82
MACD / 信号
-0.006 / -0.007
接近 52 周低空头排列
风险提示

本报告仅基于公开行情数据的量化技术指标读数进行客观描述,不构成任何投资建议。技术指标反映的是历史价格行为,过去走势不代表未来表现,亦不能预测短期涨跌,仅供技术指标解读参考。

Magnitude 4.8 earthquake hits Victoria with tremors felt across Melbourne

The epicentre was at Leongatha, 130km south-east of Melbourne, with residents reporting houses shaking just after 2am Get our breaking news email, free app or daily news podcast A magnitude 4.8 earthquake struck Victoria in the early hours of Saturday morning, shaking Melbourne and waking people acr

中文摘要 2026年10月2日凌晨,澳大利亚维多利亚州发生4.8级地震,震中位于雷昂加塔,距墨尔本东南约130公里。墨尔本多地居民在凌晨2时左右感到房屋摇晃。

How FlyDubai Passengers and Crew Averted Disaster After Pilot Was Stabbed During Flight

A badly wounded Indian pilot managed to unlock the cockpit door. Four Israeli men teamed up to overwhelm an attacker and avert disaster.

中文摘要 飞迪拜一航班在印度籍飞行员被刺伤后,重伤的飞行员设法打开驾驶舱门。四名以色列男子合力制服袭击者,避免了机舱内灾难性后果。

Iran war live: Fighting intensifies in Yemen, hundreds killed or injured

Fighting intensifies across Yemen as the army says 474 attacks in 24 hours killed or wounded 1,540 Houthi fighters.

中文摘要 伊朗与胡塞武装冲突升级,也门方面24小时内发生474起袭击,造成约1540名胡塞武装人员死亡或受伤,战斗在也门多地持续。

Cornell frat house rape accuser 'under siege' online, says lawyer

A lawyer for Jane Doe says she is "not doing well" amid attempts to uncover her identity.

中文摘要 美国康奈尔大学兄弟会性侵案原告Jane Doe的律师表示,当事人正遭受网络上大规模人肉搜索与攻击,精神状态不佳,身份仍被多方试图挖掘。

Dutch royal house helped slave trade to survive, study finds

In-depth study also says the royal house made huge profits from Dutch colonial policies in past centuries.

中文摘要 一项研究指出,荷兰王室过去曾依靠奴隶贸易维持存续,并在数世纪的荷兰殖民政策中获取巨大利润,揭露王室与殖民剥削的历史关联。

Could El Niño mean there are no Atlantic hurricanes this year?

For the first time in three decades there has not been a single Atlantic hurricane in September, normally the most active month.

中文摘要 受厄尔尼诺影响,2026年9月三十年以来首次没有出现大西洋飓风,而9月通常是该海域最活跃的月份。BBC讨论今年大西洋飓风季是否将持续异常平静。

Iran Jails Hard-Line Lawmaker, Exposing Deep Rifts in Its Leadership

Hamid Rasaei, a vocal opponent of a diplomatic deal with the United States to end the war, was convicted of publishing false information with the intent to harm the Parliament speaker.

中文摘要 伊朗强硬派议员哈米德·拉萨伊因反对与美国就结束战争达成外交协议被判公布虚假信息罪,入狱事件暴露伊朗领导层内部深层分歧。

Women given shorts at Oktoberfest to prevent upskirting

The step at the Munich festival comes after thousands of voyeuristic videos emerged on social media.

中文摘要 针对社交媒体上出现的大量偷拍视频,德国慕尼黑啤酒节主办方今年向女性游客发放短裤,以预防偷拍裙底行为。

The vanishing world of India's street magicians

India's street performers are struggling to survive as audiences shrink and public spaces become harder to access.

中文摘要 BBC报道印度街头魔术师群体正逐步消失,受众减少、公共空间准入难度上升,令这一传统表演生计日益艰难。

Flydubai pilot recounts cockpit stabbing in call with Indian PM Modi

Smit Machchhar recalls midair disaster after co-pilot's attack on Flydubai flight to Tel Aviv.

中文摘要 飞迪拜航班飞行员斯米特·马查卡尔在与印度总理莫迪通话时,回忆了同僚驾驶副驾遭袭击的空中惊魂事件,称自己险些酿成重大灾难。

Why Carney’s economic overhaul is clashing with Canada’s unions

Unions say proposed changes to Canada's labour laws could weaken workers’ right to strike.

中文摘要 加拿大总理卡尼推出的经济改革与工会组织产生冲突,工会指控拟议的劳动法修改可能削弱工人的罢工权,引发劳资关系紧张。

Bridal Shop Owner Who Put Child Onto a Coat Hook Cleared by Court

The “horseplay” was misguided, said the Canadian judge, but not malicious.

中文摘要 加拿大纽芬兰一婚纱店店主将男孩放到衣钩上的事件获法院宣判无罪,法官称此举属不当的「闹着玩」,但不具恶意。

Stocks Rise As Jobs Report Eases Fed-Hike Worries

Bloomberg Television brings you the latest news and analysis leading up to the final minutes and seconds before and after the closing bell on Wall Street. Today's guests are Chris Phelan CEA Chairman, Robert Reich, Former US Labor Secretary, Alex Straton, Morgan Stanley Equity Research Managing Dire

中文摘要 彭博收盘:美国就业数据缓解加息担忧,推动股市上涨。嘉宾包括经济顾问委员会主席Chris Phelan、前劳工部长Robert Reich及大摩分析师Alex Straton。

Wall Street Week | Your Brain on AI, Rare Earth Race, College Enrollment Cliff

This week, National Economic Council Director Kevin Hassett says AI’s productivity boost may be much larger than official economic data currently indicate. And, AI companies are largely in agreement that the technology needs guardrails, but the ethical implications of AI could also create new opport

中文摘要 彭博《华尔街一周》:国家经济委员会主任Kevin Hassett称AI生产力提升远超现有经济数据所反映;AI企业认同需护栏但伦理争议大。专题涵盖稀土竞赛与大学入学悬崖。

Former CEA Chair on Jobs Report, GDP Growth

Former CEA Chair Cecilia Rouse believes that we may be 'entering a new normal' of the jobs report, with the break even potentially settling in around 50,000. She speaks with Romaine Bostick on Bloomberg's "The Close." (Source: Bloomberg)

中文摘要 前经济顾问委员会主席Cecilia Rouse认为就业报告正进入「新常态」,月增就业盈亏平衡点可能稳定在5万左右。她做客彭博《The Close》节目讨论GDP增长。

Morgan Stanley's Nike Analyst on quarterly earnings

Morgan Stanley's Alex Straton sees Nike sales and earnings softening as the company deals with inventory gluts in North America and China. She speaks with Romaine Bostick on Bloomberg's "The Close." (Source: Bloomberg)

中文摘要 摩根士丹利分析师Alex Straton预计耐克在北美和中国库存积压问题持续拖累下,季度销售额与盈利将走弱。她做客彭博《The Close》评论财报。

Pam Bondi’s Secret DOJ Email Address Revealed

A FOIA request has uncovered another little-known feature of life at the top of the Justice Department: attorneys general have used alternate email addresses to conduct government business for over a decade. This time around, Bloomberg News Senior Investigative Reporter Jason Leopold obtained the al

中文摘要 一项信息公开申请披露,美国司法部长使用非公开备用邮箱处理公务已逾十年。彭博调查记者Jason Leopold解读Pam Bondi的秘密DOJ邮箱事件。

EM Assets Find Relief as US Jobs Data Eases Rate Fears

An index tracking emerging-market currencies climbed to a session high after weaker-than-expected US jobs data prompted traders to scale back bets on another Federal Reserve interest-rate hike this month. Nonfarm payrolls increased by 29,000 in September after downward revisions to the prior two mon

中文摘要 美国9月非农就业仅增2.9万且前值遭下修,市场缩减对本月美联储加息押注,新兴市场货币指数升至盘中高点。

CEA Chairman: Inflation coming down sufficiently fast

Chris Phelan CEA Chairman, sees inflation clearly easing and thinks it's doing so quickly enough. He speaks with Romaine Bostick on "The Close." (Source: Bloomberg)

中文摘要 美国经济顾问委员会主席Chris Phelan表示通胀正在明显缓解,且下行速度足够快。他做客彭博《The Close》节目阐述观点。

Lukoil Loses Bid to Dismiss Suit Over Franchisee Fuel Pricing

The owners of almost 30 Lukoil PJSC filling stations in Pennsylvania and New Jersey can proceed with a lawsuit accusing the Russian oil giant of overcharging franchisees.

中文摘要 卢克石油公司驳回诉讼的努力失败。宾夕法尼亚州和新泽西州近30座加油站业主可继续起诉这家俄罗斯石油巨头,指控其向加盟商收取过高燃油价格。

Former Labor Secretary Rob Reich on Sept. Jobs Report

Robert Reich, former US Labor Secretary, discusses persistent inflation, very slow hiring, and wages that trail prices. He speaks with Romaine Bostick on Bloomberg's "The Close." (Source: Bloomberg)

中文摘要 前美国劳工部长Robert Reich在彭博《The Close》节目上讨论9月就业报告,指出通胀顽固、招聘极度缓慢,且工资增速落后于物价上涨。

Using Decision Science in Investing | Masters in Business with Omar Aguilar

Barry speaks with Omar Aguilar, CEO and CIO of Schwab Asset Management. They discuss his time at Financial Engines, where he worked under Nobel laureate William Sharpe before joining Schwab Asset Management in 2011. They discuss how his doctorate in decision sciences helps him build trust with clien

中文摘要 彭博专访嘉信资产管理CEO兼CIO Omar Aguilar。他曾在Financial Engines任职于诺奖得主William Sharpe门下,2011年加入嘉信,探讨决策科学在投资中的应用。

Wall Street Tries to Live With 5% Yields as Market Cracks Grow

Wall Street has spent weeks trying to make peace with the great bond selloff. Friday offered some short-lived relief — along with a warning about the damage from stubbornly high yields across investment strategies of all stripes.

中文摘要 华尔街正艰难适应5%的高收益率环境。债券持续抛售引发市场裂痕扩大,周五短暂喘息也提醒顽固高收益率正在损害各类投资策略。

Treasury Yields Climb, Adding to Weekly Gains | Closing Bell

Comprehensive cross-platform coverage of the U.S. market close on Bloomberg Television, Bloomberg Radio, and YouTube with Romaine Bostick, Katie Greifeld, Carol Massar and Tim Stenovec. (Source: Bloomberg)

中文摘要 彭博收盘报道:美国国债收益率攀升,延续本周涨势。Romaine Bostick、Katie Greifeld、Carol Massar与Tim Stenovec跨平台评论美股收盘。