TypeScript · ★ 103,976 · 🍴 22,810 · 📈 25,599 stars today
Reverse engineer anything with agents, from app behavior down to native binaries.
中文介绍 基于 AI Agent 的自动化逆向工程工具，能够端到端解析从应用顶层行为到原生二进制代码的完整逻辑。适用于安全研究员与逆向工程师，可在恶意代码分析、未知协议逆向及漏洞挖掘等场景中大幅提升分析效率。
C++ · ★ 30,089 · 🍴 2,583 · 📈 3,952 stars today
Tool for automatic PS5 executables porting to Linux and Windows
中文介绍 用于将 PlayStation 5 原生可执行文件自动移植到 Linux 和 Windows 平台的自动化工具。它通过解析与转换二进制依赖简化移植流程，主要面向主机游戏开发者与逆向工程爱好者，便于在 PC 环境下测试与运行 PS5 程序。
Rust · ★ 17,863 · 🍴 2,922 · 📈 2,957 stars today
ArtCraft is an intentional crafting engine for artists, designers, and filmmakers
中文介绍 面向艺术家、设计师与电影创作者的意图驱动型创作引擎，专注于辅助视觉叙事与艺术概念设计。它通过规范化工具链帮助创作者把控创作流程，适合影视概念设计、游戏美术及独立数字艺术创作团队使用。
HTML · ★ 50,324 · 🍴 3,196 · 📈 1,083 stars today
Editorial diagram design for Claude Code, Codex, GitHub Copilot, Cursor, Factory Droid, and Pi. 44 diagram types. Self-contained HTML + SVG. No shadows. No Mermaid slop.
中文介绍 专为 Claude Code、Cursor 等 AI 编码工具优化的图表设计规范库，内置 44 种图表类型，采用独立无依赖的 HTML 与 SVG 渲染。适合技术作者与开发者生成高质量架构图，替代样式粗糙的传统图表方案。
JavaScript · ★ 51,666 · 🍴 10,375 · 📈 886 stars today
A spy satellite simulator in your browser, except the data is real. Live open source spatial intelligence on a photorealistic 3D globe.
中文介绍 基于 Web 端的间谍卫星模拟器，在写实级 3D 地球仪上实时聚合与可视化真实的开源空间情报（OSINT）数据。适合地理空间分析师、开源情报研究人员与天文地理爱好者进行全球态势感知与卫星数据探索。
Python · ★ 104,482 · 🍴 32,117 · 📈 315 stars today
Tensors and Dynamic neural networks in Python with strong GPU acceleration
中文介绍 主流开源深度学习框架，支持灵活的动态计算图与高并发 GPU 加速张量计算。它由 Meta 主导维护，提供丰富的 Python 接口，是 AI 研究员与算法工程师在学术研究、计算机视觉、大模型训练及推理部署中的核心基础设施。
Jupyter Notebook · ★ 36,474 · 🍴 3,818 · 📈 548 stars today
AirLLM 70B inference with single 4GB GPU
中文介绍 面向极端受限硬件的大模型推理加速库，采用分层加载与内存置换技术，使单张 4GB 显存消费级显卡也能完整运行 70B 规模参数的大语言模型。适合预算有限的个人开发者与本地实验场景，实现零显存焦虑的本地推理。
Assembly · ★ 74,246 · 🍴 8,101 · 📈 466 stars today
Original Apollo 11 Guidance Computer (AGC) source code for the command and lunar modules.
中文介绍 阿波罗 11 号登月任务制导计算机（AGC）指令舱与登月舱的原始汇编开源代码。项目完整还原了 1969 年的人类登月控制系统逻辑，是计算机史学者与航空航天工程师研习早期高可靠嵌入式实时系统设计的经典参考。
👍 74
Reinforcement learning (RL) is the central training paradigm for advancing large foundation models towards self-improvement. This report introduces the MiMo-V2.6 series, an omni-modal family that pushes the frontier of model intelligence by scaling RL compute. Prior to RL, we conduct mid-training on
中文介绍 聚焦基础大模型的自我改进，发布全模态模型家族 MiMo-V2.6。该工作通过扩展强化学习（RL）计算规模推动模型智能前沿，并在执行大规模 RL 之前引入中期训练（mid-training），显著增强模型在多模态理解与自主迭代上的能力。
👍 52
Egocentric world models predict first-person observations conditioned on an agent's actions, but most focus on a single agent. Real embodied settings often involve multiple agents that act and interact within a shared environment. Existing multi-agent world models rely on coarse actions like locomot
中文介绍 针对现有第一人称世界模型局限于单智能体、且多智能体交互动作过于粗粒度的问题，提出支持细粒度具身交互的多智能体第一人称世界模型。该模型能在共享环境中基于多体动作联合预测第一视角观测，刻画复杂多智能体协作与交互动态。
👍 42
This is a book about large language models. As indicated by the title, it primarily focuses on foundational concepts rather than comprehensive coverage of all cutting-edge technologies. The book is structured into six main chapters, each exploring a key area: pre-training, generative models, prompti
中文介绍 系统梳理大语言模型的基础理论与核心概念。全书结构分为预训练、生成模型、提示工程等六大核心章节，摒弃碎片化追踪前沿技术的做法，专注提炼大模型底层原理与核心知识体系，为深入理解和研发大语言模型提供系统化的理论基石。
👍 39
Large language models are informing decisions with ever-higher stakes. As the consequences of their errors grow, a central question becomes harder to ignore: how much can we trust an individual answer? Yet recognizing when to defer remains difficult because language models can present incorrect conc
中文介绍 针对大语言模型在高风险决策中难以判断单次回答可靠性的难题，提出 U-Space 框架以揭示模型何时及为何产生不确定性。该方法剖析模型置信度与错误成因，帮助系统精准识别何时该推迟决策或拒绝回答，提升模型应用的安全性与可信度。
👍 38
Long-horizon agents compact their history to continue within a finite context window, but a textual summary alone may not support every subsequent decision. We introduce REMORY, a neural memory network that supplements the summary with a bounded sequence of soft memory tokens. Given the history and
中文介绍 针对长跨度智能体仅依靠文本摘要压缩历史容易丢失关键决策信息的问题，提出神经记忆网络 REMORY。该网络利用有界软记忆（soft memory）Token 序列补充文本摘要，保留历史关键残差信息，在有限上下文窗口内大幅提升长程决策质量。
👍 35
Learning to act in unfamiliar environments requires agents to infer how the world works and revise that understanding as new evidence arrives. Yet limited observations can support multiple world models that explain past interactions but predict different outcomes in unseen states. We introduce Memen
中文介绍 针对智能体在未知环境中因观测有限导致世界模型存在歧义的问题，提出 Memento 3 框架。该方法基于反思规则手册（Reflective Rulebooks）实现模型的递归自我改进，动态检验并修订对环境运作机制的认知，提高在未见状态下的预测与决策鲁棒性。
👍 34
Large language model (LLM) coding agents have advanced test generation across diverse programming tasks. However, the common practice of evaluating tests against a single reference solution overlooks alternative valid implementations and can overstate test quality. We introduce TestPrism, comprising
中文介绍 针对代码大模型生成测试用例时依赖单一参考实现容易高估测试质量、忽略多种有效实现的问题，提出评估框架 TestPrism。该框架跳出单参考局限，综合考虑多种有效实现方案，提供更严谨、更具泛化性的代码测试质量评估标准。
👍 29
Recent text-to-image generation models have achieved remarkable visual quality, but improving them through post-training remains challenging because no single reward signal captures the full range of human preference. In this work, we develop a simple and effective post-training recipe for open-doma
中文介绍 针对单一奖励信号无法全面捕捉人类审美与需求偏好的难题，提出结合偏好奖励与评价细则奖励（Rubric Rewards）的文生图模型后训练方案。该方法通过组合多种细粒度奖励信号，对开放域前沿文生图模型进行对齐优化，显著提升生成图像的综合质感与指令遵循度。
👍 24
Single-image, multi-image, and video deep research require different visual operations but share a workflow of visual grounding, external retrieval, and fact composition. A key challenge is to preserve the dependencies linking localized visual anchors, entity relations, source-supported facts, and a
中文介绍 针对图像与视频深度研究中视觉操作割裂的问题，提出统一多模态研究智能体 OneSearch-VL。该系统统一了视觉定位、外部检索和事实整合的工作流，有效维持视觉定位锚点、实体关系与检索事实间的依赖链路，实现了跨图文与视频的端到端深度分析能力。
👍 19
Most robot policies keep a model in the control loop: a VLA maps observations to actions, and an Agent Harness, such as Agent-as-Policy or Harness VLA queries a VLM for decision making at run time. We propose a different view: the embodied world is an Embodied Turing Machine, whose tape is the robot
中文介绍 突破传统机器人在控制循环中频繁调用 VLA/VLM 的范式，提出具身图灵机（Embodied Turing Machines）。该方案将物理世界与机器人状态形式化为图灵机磁带与有状态代码，通过程序化状态维护实现机器人的递归自我改进与高效长期执行。
👍 17
Long-context LLM agents accumulate interaction histories that strain KV-cache memory and attention computation. Although sparse attention reduces these costs, heterogeneous cache representations and workflows hinder integration with existing inference engines, while prior sparse-serving abstractions
中文介绍 针对长上下文 LLM 智能体导致 KV 缓存爆炸和注意力计算受阻的痛点，构建稀疏优先的推理引擎 SparseEngine。该引擎解决了稀疏注意力机制中缓存表征异构与现有推理架构难以整合的问题，显著降低长程交互下的显存占用并提升吞吐效率。
👍 15
Multimodal large language models (MLLMs) are rapidly evolving toward continuous audio--visual reasoning, creating an urgent need for evaluations that expose their capability limits. Audio--visual captioning is an ideal diagnostic task, yet current benchmarks face a coupled trade-off: whole-caption s
中文介绍 为突破多模态大模型连续音视频推理能力的评测局限，提出深度结构化评测框架 OmniCapBench。该基准针对音视频字幕生成任务，打破了传统整句评分的粗粒度权衡，通过细粒度语义结构分解，精准诊断模型在跨模态时序感知与细节描述上的极限。
👍 14
Long-horizon coding agents need timely corrections, yet feedback can be ineffective or even harmful when it misjudges ongoing work or fails to address the underlying problem. Existing critics focus on evaluating trajectories and generating feedback, but rarely track what happens after feedback is de
中文介绍 针对长程代码智能体中评价反馈容易出现误判、甚至无法解决根本问题而导致反向干扰的困境，提出语言评判框架 Opera。该框架超越单纯生成轨迹反馈，主动跟踪反馈落地后的执行效果与纠错闭环，有效提升长周期编程任务的纠错成功率与稳定性。
👍 14
Large language models (LLMs) are increasingly used as interactive agents and simulators, yet it remains unclear whether they can recover latent sequential structure beyond surface action frequencies. This distinction is critical for behavioral simulation, where actions are often shaped by prior cont
中文介绍 探究大语言模型在交互模拟中是否真正理解潜在的时序结构，而非仅仅拟合表面动作频率。通过对推理与生成行为的受控对比实验，揭示模型在结合上下文依赖进行时序决策时的能力边界，为将 LLM 用作行为仿真器提供了关键的机理洞察。
👍 12
Reinforcement learning from verifiable rewards (RLVR) frequently reuses rollouts across multiple policy updates, increasing the mismatch between the current policy and the data-generating policy. We identify a sign-dependent gradient starvation problem in clipped policy optimization: clipping suppre
中文介绍 针对可验证奖励强化学习（RLVR）在多轮策略更新复用采样时，截断策略优化引发的符号依赖型梯度饥饿问题，提出重塑序列策略优化算法 ReSPO。该算法重新调整序列层面的优化更新，缓解策略漂移带来的梯度抑制，提升离策略学习的训练稳定性。
👍 8
Text-to-music systems produce increasingly convincing audio, yet evaluation reveals little about whether the result matches user intent. A global text-audio relevance score can overlook the implicit intent in underspecified prompts and mask failures in specific requirements, such as instrumentation,
中文介绍 针对文本生成音乐中全局相关性评分难以反映用户隐式意图、易忽略乐器编排等具体需求的弊端，提出音乐意图细化智能体 MIRA。该方法通过交互式意图细化与约束分解，精细对齐文本提示与音乐特征，显著改善音乐生成对复杂需求的遵从度。
👍 8
Recent AI systems have made rapid progress in scientific discovery when given well-defined metrics, but whether they can autonomously undertake open-ended scientific discovery remains unclear. We investigate AI's ability to tackle open-ended tasks in Station, an open-world environment in which multi
中文介绍 探究 AI 智能体在脱离明确指标时能否自主进行开放式科学发现。研究构建了多智能体开放世界环境 Station，系统评估智能体在开放式假设提出、实验设计及知识演化中的表现，实证揭示了当前自主智能体在开放科学探索中的潜力与核心瓶颈。
👍 8
Black-box optimization (BBO) arises in many scientific and engineering problems where objective evaluations are expensive and limited. Recent large language model (LLM) agents offer a new way to approach BBO by combining task semantics, computation, optimization tools, and feedback-driven decision m
中文介绍 针对昂贵目标函数下的黑盒优化（BBO）问题，提出面向大模型智能体的系统化基准 Agentic BBO。该基准全面评估智能体融合领域语义、计算工具及反馈驱动决策的能力，深入量化分析了 LLM 智能体在复杂科学与工程黑盒寻优中的表现与策略优势。
👍 8
When retrieved evidence contradicts an agent's prior beliefs, does it revise its answer, acknowledge uncertainty, or persist with an incorrect conclusion? Existing evaluations of agentic systems focus primarily on task success, offering limited insight into how agents handle such conflicts. We propo
中文介绍 探讨大语言模型智能体在检索证据与先验信念冲突时的认知谦逊（Epistemic Humility）表现。针对传统评测仅侧重任务成功率的缺陷，提出专门的冲突评估框架，深入探究智能体在面对矛盾信息时是修正认知、表达不确定性还是盲目固执。
👍 8
Spatial reasoning remains a persistent weakness of vision-language models (VLMs), because RGB inputs do not directly provide geometric evidence. Existing remedies either inject 3D into the model at inference, paying architecture and latency costs, or train with outcome rewards that supervise only th
中文介绍 针对纯 RGB 输入缺乏几何信息导致视觉语言模型（VLM）空间推理能力不足的问题，提出路由式 3D 特权蒸馏方法。该方法在训练时引入 3D 专家知识并路由蒸馏至模型中，无需在推理时增加 3D 模态架构与延迟开销，显著提升复杂空间推理表现。
@satyanadella · 9.6M 粉丝 · 448.2K 阅 · 1.3K 赞 · 183 转
As traditional software systems were being deployed across the economy over the last few decades, we had the tools and capability to trace behaviors to a specific code path. That same kind of
中文介绍 微软 CEO Satya Nadella 探讨超级智能时代的安全治理。他指出传统软件行为可通过代码路径追溯，而黑盒前沿模型更接近「内部人风险」，呼吁业界构建针对模型自主行为的新型追踪与审计体系。
@sermakarevich · 1.0K 粉丝 · 30.6K 阅 · 504 赞 · 80 转
An article for engineers who ship agents built on large language models (LLMs): systems that call tools, change records, and act on a customer's behalf over several steps. After reading it you will be
中文介绍 面向 LLM 落地工程师的 Agent 评测指南。针对具备工具调用、记录修改与多步骤代理执行能力的复杂智能体系统，解析如何建立评测指标，验证 Agent 任务完成度及其在多轮交互中的行为稳定性。
Tomorrow and Tuesday. If you see this you’re in!
中文介绍 Latent Space面向纽约本地订阅用户举办专属线下交流见面会，活动定于周一及周二举行，仅对订阅会员开放参与。
Inside Standard Bots’ AI stack, pretrained models learn factory tasks from demonstrations and improve through corrections from real deployments.
中文介绍 工业机器人公司Standard Bots探讨AI在制造业的执行可靠性，其预训练模型通过示范学习工厂任务，并在实际产线反馈中持续迭代优化。
Wow.
中文介绍 AI行业最新数据显示，TypeSafe/Jev在产品上线仅3周后，年度经常性收入（ARR）已突破1亿美元，公司估值达到75亿美元。
From the Bitter Lesson of AI scaling to the unsolved mysteries of protein folding, Google DeepMind’s Pushmeet Kohli and Biohub’s Sal Candido are rethinking what it takes to build AI that truly understands biology.
中文介绍 Google DeepMind的Pushmeet Kohli与Biohub的Sal Candido展开对话，探讨AlphaFold局限与蛋白质折叠未解难题，反思生物学AI发展路径。
中文介绍 艾伦人工智能研究所（Allen AI）在Hugging Face博客探讨GPU集群的高效调度策略，旨在优化算力资源分配并提高模型训练与推理效率。
Ever since people first seriously contemplated giving machines an intelligence modeled on our own, there has never been any question that they would, like us, be able to say no. The sci-fi canon is full of stories of robotic disobedience. Most of these capers are, of course, cautionary. But recently
中文介绍 《麻省理工科技评论》发文探讨人类对AI拒绝能力的盲目信任。文章指出人们常假定系统能如人类般拒绝不合理指令，忽视了AI对合规执行的底层局限。
Discover how Sophos uses OpenAI’s Daybreak to cut cyber-threat investigation time by 96% and automate 52% of MDR cases while preserving human oversight.
中文介绍 网络安全公司Sophos接入OpenAI的Daybreak技术，将网络威胁调查时间大幅缩减96%，并在人工监督下实现了52%托管检测与响应案例的自动化。
Using GPT-6 Astra in Codex, Asana made its browser agent 76x cheaper and 5x faster in tests to offer customers more capable models.
中文介绍 协同办公平台Asana在浏览器智能体测试中引入OpenAI模型，将运行成本大幅压缩至原先的七十六分之一（76倍降幅），同时实现5倍的运行速度提升。
Friday, October 16, 2026 Can AI design new life forms? In 2025, Stanford University PhD student Samuel King came up with a preliminary answer when he used a generative AI model to propose genetic blueprints for microscopic viruses. It isn’t yet an example of AI-generated life, but that could be next
中文介绍 《麻省理工科技评论》对话斯坦福大学博士生Samuel King。King此前利用生成式AI为微观病毒设计基因蓝图，探讨AI能否创造全新生命形式及前沿生物风险。
中文介绍 TLDR AI汇总前沿技术动态，重点关注OpenAI发布的GPT-6.1 Ultrafast、Gemini通用智能体代理，以及加速推理的投机解码技术进展。
a quiet day.
中文介绍 Latent Space发布最新AI行业简报，称当日人工智能领域整体动态相对平淡，未出现重磅产品发布或突破性行业事件。
A special Science pod and Engineering pod crossover.. with Forward Deployed Engineering kicker!
中文介绍 Periodic Labs联合创始人Liam Fedus与Ekin Dogus Cubuk探讨AI驱动的物质合成超级智能，涵盖从半导体到超导材料领域的研发创新与工程应用。
Across recruiting, engineering, and operations, Oracle turns specialist knowledge into fast, repeatable workflows with ChatGPT Work and Codex.
中文介绍 甲骨文（Oracle）在招聘、工程与运营等业务中深度部署ChatGPT Work与Codex，将数天的专业流程缩短至数分钟，实现高效可复用的自动化工作流。
With GPT-5.6, GPT-6 Astra, and GPT‑Image‑2.5, Pollo AI helps creators turn bold ideas into detailed images and cinematic video ads.
中文介绍 Pollo AI整合OpenAI的GPT-5.6、GPT-6 Astra及GPT-Image-2.5模型，助力创作者将创意快速转化为高清图片与电影级视频广告活动。
LegalOn cut estimated daily Codex costs by 65% while maintaining development speed. It matched Astra, Sol, and Luna to tasks and managed budgets strategically.
中文介绍 法律科技公司LegalOn通过精准匹配Astra、Sol与Luna模型及预算管理，在保持开发速度的同时，将Codex预估日支出大幅缩减了65%。
全球核心大类资产呈现结构性分化行情。美股主要指数与龙头资产维持强劲上行形态，SPY 与 QQQ 价格分别报 778.57 与 751.27，均接近 52 周高点并呈现标准均线多头排列，同期 VIX 恐慌指数下行至 14.84 处于空头排列。加密市场方面，全网加密总市值报 2.82T USD（24h 下跌 2.71%），BTC 主导率达 59.6%，加密恐慌贪婪指数维持在 69（贪婪）；BTC 保持在 83774.24 美元且均线多头排列，而 ETH 与 SOL 短期动能回调，呈现高位整理。中概资产如 BABA 与 PDD 迎来单日 3% 至 5% 的反弹并伴随 MACD 低位金叉，但受制于长期均线压制，大级别仍处空头排列。商品与外汇领域，黄金期货（4228.5）受均线空头压制，美元人民币（6.68）走弱探底，美元指数则在 102.15 保持多头格局。
当前价格 535.07 美元，日内上涨 2.38%，5 日累涨 3.39%。价格稳固运行于 SMA20（510.10）、SMA50（500.45）与 SMA200（434.00）上方，均线维持多头排列。MACD 指标达 10.64 且高于信号线 9.13，RSI14 读数为 67.5 展现强动能，距离 52 周高点仅差 3.37%，技术面多头特征显著。
当前汇率报 6.68，单日下行 0.28%，距离 52 周低点仅 0.13%。价格持续承压于 SMA20（6.70）、SMA50（6.72）与 SMA200（6.82）下方，均线系统呈现典型空头排列。MACD 读数 -0.0062 低于信号线 -0.0059 延续死叉状态，RSI14 跌至 22.5 进入超卖区，下行趋势主导特征明确。
以太坊当前报价 2538.61 美元，日内反弹 1.38%，但 5 日回撤 5.89%。价格虽守在 SMA200（2133.89）上方，但跌破 SMA20（2650.80）与 SMA50（2556.24），趋势定性为中性震荡。MACD 跌至 -1.75 且显著落后于信号线 31.58，RSI14 处于 44.9 的中间水平，指标呈现多空博弈与修正状态。
VIX 恐慌指数
10Y 美债收益率 (%)
美元指数 DXY
S&P 500 ETF
Nasdaq 100 ETF
Apple
Microsoft
Nvidia
Alphabet
Tesla
Meta
Bitcoin
Ethereum
Solana
阿里巴巴 (BABA)
拼多多 (PDD)
京东 (JD)
腾讯控股 (0700.HK)
黄金期货
WTI 原油期货
美元 / 人民币
本报告内容仅供技术指标解读参考，不构成任何投资建议或交易指引。技术指标仅反映历史价格与动能的客观统计状态，存在固有时滞性与局限性，过去走势不代表未来表现，请理性评估市场波动风险。
Felipe VI’s first visit to Ceuta and Melilla as king brings Spain’s territorial dispute with Morocco back into focus.
中文摘要 西班牙国王费利佩六世首次以国王身份访问休达和梅利利亚两座自治市，使西班牙与摩洛哥之间长期存在的领土主权争议再次成为焦点。摩洛哥方面将休达称为塞卜泰，并坚称对该地拥有主权。
Iran says US has accepted some of its peace terms but calls the reply 'inadequate' as the Riyadh airport is hit again.
中文摘要 伊朗官方表示，美国已接受其提出的部分和平条件，但认为美方的整体回应依然「不充分」。与此同时，沙特阿拉伯首都利雅得的国际机场再度遭遇袭击，中东地区安全局势持续紧张。
Follow the day’s news live Get our breaking news email, free app or daily news podcast Labor not watering down 30% tax on discretionary trusts, says assistant treasurer The assistant treasurer, Daniel Mulino, says Labor is not watering down proposed changes to put a minimum 30% tax on discretionary
中文摘要 澳大利亚助理财政部长丹尼尔·穆利诺明确表示，工党政府不会削弱针对全权委托信托征收最低30%税率的拟议改革方案。与此同时，两党就消费税改革提议及维多利亚州选举投票倾向展开激烈辩论。
At least one person was killed in the Saudi capital, Riyadh, kingdom authorities said. The Houthis said Saudi aircraft hit the Yemeni capital’s airport.
中文摘要 沙特阿拉伯与也门两地机场在新一轮相互打击中受损。沙特官方证实，首都利雅得遭遇袭击并导致至少1人死亡；也门胡塞武装则指责沙特军机空袭了也门首都萨那的机场设施。
Saudi authorities say a Pakistani national was killed and several people injured in a second strike on Riyadh’s Airport.
中文摘要 沙特阿拉伯当局通报，也门胡塞武装对首都利雅得国际机场发动了第二次袭击，造成一名巴基斯坦籍公民死亡，另有多人受伤。这是该机场短期内遭遇的第二轮致命越境打击。
Decision to livestream Nidal Hasan execution has led to harsh criticism and warnings that it could encourage violence.
中文摘要 针对公开网络直播处决胡德堡枪击案凶手尼达尔·哈桑引发的广泛争议，美国国土安全部部长表态称该举措不存在安全风险。此前外界普遍批评此举极不妥当，警告可能诱发模仿暴力。
CCTV footage shows the moment of a Russian strike on a residential street in Kharkiv, Ukraine.
中文摘要 乌克兰哈尔科夫的监控画面记录下了俄罗斯对居民区发动打击的瞬间。袭击直接击中了当地一条居民街道上的杂货店设施，直观反映了前线城市民用基础设施遭空袭破坏的情况。
The Saudi capital has been hit by multiple attacks from the Iran-backed Yemeni group over the weekend.
中文摘要 沙特阿拉伯首都利雅得机场在周末两天内遭遇也门胡塞武装发动的第二次袭击，已造成1人死亡、11人受伤。受伊朗支持的也门胡塞武装近期对沙特本土目标明显加大了打击频次。
Second boy, 16, also charged in relation to the stabbing death of a 30-year-old man in Mount Druitt Get our breaking news email, free app or daily news podcast A 13-year-old boy has been charged with murder after a man was stabbed to death in Sydney’s west on Friday. Emergency services were called t
中文摘要 澳大利亚悉尼西部德鲁伊特山发生持刀杀人案，一名30岁男子周五被刺身亡。新南威尔士州警方随后逮捕涉案人员，正式指控一名13岁男孩犯有谋杀罪，另一名16岁男孩亦遭到起诉。
Unidentified projectile strikes King Khalid airport, killing one and injuring 11, as escalating attacks threaten to draw more countries into Yemen conflict Analysis: attack will force rethink of Trump strategy A projectile hit a passenger terminal at the airport in Riyadh on Sunday, the day after a
中文摘要 一枚不明发射物击中沙特首都利雅得哈立德国王机场客运航站楼，造成1死11伤。不断升级的越境袭击加剧了更多国家被卷入也门冲突的风险，并可能迫使各方重新审视美国的地区战略。
CCTV footage released by Al Arabiya shows the moment of a Houthi strike on Riyadh’s international airport on Saturday.
中文摘要 阿拉伯电视台公布的监控录像记录了也门胡塞武装周六袭击沙特利雅得国际机场的具体瞬间。该起袭击直接击中了民航枢纽区域，造成了人员伤亡并导致机场运营严重受阻。
Zelensky says Kyiv would support such a truce. Moscow has yet to comment.
中文摘要 乌克兰总统泽连斯基对美国前总统特朗普提出的「能源停火」主张表示欢迎，明确称基辅方面愿意支持此类休战方案。截至目前，俄罗斯官方尚未对此项停火提议作出任何公开回应。
As attacks continue to mount on shipping vessels transiting the Strait of Hormuz, a seafarer's union official says some crews are being offered big rewards for making the crossing.
中文摘要 随着霍尔木兹海峡商船遇袭事件持续频发，海员工会官员透露，多家航运机构正为愿意穿越该关键海域的船员提供巨额高额报酬，以重金激励船员在巨大安全风险下维持国际航运畅通。
Madi Bolanos and Anna Sussman's new podcast We Are Jane Doe, by KQED and Snap Studios, investigates rampant sex abuse at the Central California Women's Facility in Chowchilla, Ca.
中文摘要 由KQED与Snap Studios联合推出的播客调查节目披露，位于加利福尼亚州乔奇拉的中加州女子监狱内长期存在猖獗的性虐待与暴力行为，通过多方第一手陈述揭露了监管系统的严重失职。
NPR's Joel Rose speaks with Ukrainian Minister of Parliament, Oleksandr Merezhko, about the U.S.'s diesel deal with Russia will impact Ukraine.
中文摘要 乌克兰议员亚历山大·梅列日科在接受采访时，将美俄之间达成的柴油交易直接斥为对乌克兰的「背叛」，指出此类能源合作将对乌克兰抗俄努力产生严重的负面冲击与战略挫伤。
A rally in South Korean bonds may prove fleeting as the effect of government-led measures to quell volatility fades and expectations of further monetary policy tightening build, according to strategists.
中文摘要 策略师指出，韩国国债的反弹走势或难以持久。随着政府此前平抑市场波动的干预措施效果逐渐消退，交易员对央行进一步收紧货币政策的预期不断增强，导致债市前景承压。
Lou Crandall, the Wrightson ICAP chief economist whose dispatches on Treasury financing and Federal Reserve operations became required reading across trading desks for nearly half a century, announced he’s ending his weekly commentary.
中文摘要 Wrightson ICAP首席经济学家卢·克兰德尔（Lou Crandall）宣布结束其撰写了44年的每周市场评论。近半个世纪以来，他对美国国债融资和美联储公开市场操作的专业解读一直是交易员的必读之作。
One in 10 scams reported to Citizens Advice relate to fraudsters charging for work not done, or done badly.
中文摘要 英国公民咨询局（Citizens Advice）数据显示，每十起诈骗报案中就有一起来自屋顶等房屋维修骗局。不法分子针对未完成或劣质施工收取高额费用，导致房主损失数千英镑。
People are increasingly falling behind on credit card repayments, so here's the best way to get debt advice.
中文摘要 由于信用卡逾期违约人群持续增加，市场上不规范的债务咨询方案频现。劣质建议可能令借款人陷入更多债务纠纷与利息成本中，寻求专业合规的财务顾问指引尤为关键。
People are increasingly falling behind on credit card repayments, so here's the best way to get debt advice.
中文摘要 由于信用卡逾期违约人群持续增加，市场上不规范的债务咨询方案频现。劣质建议可能令借款人陷入更多债务纠纷与利息成本中，寻求专业合规的财务顾问指引尤为关键。
Both Republican and Democratic lawmakers have criticised the deal, which was announced weeks before the US midterm elections.
中文摘要 特朗普政府计划自俄罗斯进口柴油的举措引发广泛争议。该交易在美国中期选举前数周公布，招致了美国国会共和党与民主党两党议员的强烈反对和批评。
New Zealand’s ruling National Party says the nation’s debt burden and budget deficit mean now is not the time to cut taxes.
中文摘要 新西兰执政党国家党明确表示，鉴于当前国家债务负担较重以及持续的财政赤字，目前尚不是推行减税的时机。政府将推迟实施减税措施，直至财政预算实现盈余。
Brad Smith calls for global rules on AI, and says the UK is in the "best position" to lead the charge.
中文摘要 微软总裁布拉德·史密斯（Brad Smith）公开表示，人工智能目前的监管力度甚至低于牛奶行业，呼吁制定全球统一的AI监管规则，并认为英国最有优势引领这一规则的建立。
The impact of the Iran war poses significant risks for the government’s spending plans, the Institute for Fiscal Studies says.
中文摘要 英国智库财政研究所（IFS）发出警告，伊朗战争的相关冲击将对英国政府的支出计划构成重大风险，进而对国防大臣希利（Healey）的预算方案实施造成严重威胁。
Prime minister and chancellor to meet chief executives and lobby groups after complaints about lack of detail on plans for economy
中文摘要 由于英国商界对预算案可能加税充满担忧并抱怨经济计划缺乏细节，英国首相与财政大臣伯纳姆（Burnham）计划加强与商界领袖的沟通，安排会见多家企业首席执行官及游说团体。
Prime minister’s pledge is backed by tech groups but hedge funds and banks say it will damage competitiveness
中文摘要 英国首相伯纳姆拟取消员工竞业禁止条款的计划遭到伦敦金融城强烈批评。尽管科技行业表示支持，但对冲基金与银行业警告称，该举措将严重削弱金融机构的竞争优势。
Automatic enrolment worked for pensions — a similar scheme could pool the risk of heavy costs from illness and infirmity in later life
中文摘要 针对社会护理资金难题，《金融时报》探讨可借鉴养老金自动加入机制。通过推行类似的强制或自动加入方案，有效汇集并分摊老年群体因患病或身体虚弱所面临的高额护理成本风险。
2 回复 · 程序员 节点
19 回复 · 程序员 节点
2 回复 · 程序员 节点
6 回复 · Apple 节点
17 回复 · Apple 节点
4 回复 · 程序员 节点
69 回复 · Linux 节点
5 回复 · Apple 节点
33 回复 · 程序员 节点
30 回复 · Apple 节点
该源今日无内容。