AI Agent

IT之家 7 月 22 日消息，《人工智能 智能体互联》系列标准应用推进专题会议 7 月 21 日在北京海淀区中关村展示中心召开。 会议由全国信息技术标准化技术委员会人工智能分委会主办。此次会议标志着国内首个覆盖智能体全生命周期的互联标准体系正式进入试点应用阶段。 此前IT之家曾报道，该系列标准（GB/Z 185.1—GB/Z 185.7—2026）于 20…

AI 点评 · 首批覆盖智能体全生命周期的国标试点，推动行业互联互通，巨头入场加速AI生态协同。

Buzz is a group chat platform for the workplace that puts humans and their AI agents in the same conversation.

AI 点评 · Buzz让人类和AI代理同群聊，或成企业协同新范式。

https://x.com/jack/status/2079605800998146171 , https://xcancel.com/jack/status/2079605800998146171 https://buzz.xyz/

AI 点评 · Jack Dorsey新项目整合团队协作与AI，技术大佬跨界尝试或重塑开发工具生态。

Hey HN! This is Divit from Almanac (YC S26). We built CodeAlmanac, a wiki for your coding agents that updates as you talk to them. It is open-source, local, and free. Here’s a demo…

https://console.cloud.google.com/agent-platform/publishers/g...

AI has entered the gigascale era. The world’s most advanced AI factories are bringing together hundreds of thousands of GPUs and CPUs to train frontier models, power agentic AI and…

Orchestrate AI agents to find real vulnerabilities in code.

In this post, we show how Amazon Quick can serve as the business-user front door for specialized agent workflows. We use the NVIDIA NeMo Agent Toolkit to build a supply-chain risk…

AI 点评 · 企业级AI智能体开发门槛降低，Quick与NVIDIA工具链结合实现供应链风险管理，展示行业落地新范

In this post, we describe how Tradeshift deployed Amazon Quick with agentic AI capabilities to replace our legacy BI tool, resulting in query response times up to 30 times faster,…

AI 点评 · 用Agentic AI替代传统BI工具，查询速度提升30倍，展示了企业数据决策的颠覆性变革。

AI 点评 · 揭示AI基础设施从算力层向智能体生态延伸的关键趋势，定义未来数字与物理世界融合的技术底座。

From open models to real-time simulation, AI and graphics breakthroughs are transforming media, content creation and robotics.

大模型时代的共同选择

7月17日-20日，一起在WAIC2026现场，看见人工智能真正进入产业深处。 过去一年，围绕AI行业的讨论正在变得更具体。大模型能力仍在持续迭代，但外界关注的重点，已经不再只停留在模型参数、模型发布和单点能力展示上。随着智能体、具身智能、空间智能、AI基础设施等方向不断演进，行业开始更频繁地追问：AI如何进入真实流程，如何完成复杂任务，又如何在产业场景中形…

腾讯云的企业级智能体平台，正式出海了。 7月18日，在2026世界人工智能大会上，腾讯云正式发布了智能体开发平台 ADP 4.0海外版，同步升级智能工作台、Claw 模式、Skill 广场三大核心模块，围绕触达、交互、生态、连接四大能力做了全面国际化适配。 ADP 的全称是 Agent Development Platform，定位为企业级 AgentOps…

Multi-agent systems routinely place one AI agent in authority over another. When a subordinate refuses a task, the manager chooses the outcome: it can renegotiate, report the failure honestly, coerce…

Predicting a football match before kickoff requires more than knowing past results: a model must use changing information and make a clear prediction before the answer is available. We present WorldCu…

Self-hosted AI agents read and write their own memory and configuration files to function. An agent may get compromised via corruption of its own state -- a compromise realized via legitimate OS syste…

Real-time multimodal applications, including voice agents and interactive video generation, compose heterogeneous models into pipelines whose efficient deployment requires application-specific decisio…

Pruning long context for coding agents has been a vital technology for efficient context management. While existing context pruning methods such as SWE-Pruner realize this by attaching a separate code…

IT之家 7 月 19 日消息，深空矩阵在 2026 世界人工智能大会上，发布面向太空 AI 算力产业化落地的系统性星座方案“星环计划”。 官方公众号显示，深空矩阵位于北京，致力于构建超大规模星群协同的太空 AI 算力基础设施。 深空矩阵创始人兼 CEO 张伟杰表示，AI 竞争最终会落到算力竞争。而随着大规模 AI 智能体落地，传统地面算力体系将面临电力、土…

AI 点评 · 卫星组网布局太空算力，抢占AI基础设施新高地，战略意义显著。

AI 点评 · 聚焦企业级Agent体系，从大模型到执行系统的落地路径，揭示AI应用新趋势。

This paper introduces EvolvingWorld, a framework and benchmark for character and world co-evolution in interactive literary worlds. Existing systems either treat interactive literary simulation as sta…

把任何小说变成可玩的游戏 · Turn any novel into a playable game — a 7-skill adaptation pipeline for Claude Code, Codex & Kimi Code(k3)

7月18日，在2026世界人工智能大会（WAIC）上，腾讯面向具身智能与智能体领域带来多项产品技术的升级发布。在具身智能领域，腾讯正式升级发布具身智能全栈方案，贯穿云底座、模型层、平台层与应用层，全面助力机器人本体及系统开发商提质提效；在智能体领域，基于个人与企业提效需求，推出差异化的全矩阵解决方案。其中，面向企业用户的腾讯云企业级智能体开发平台ADP4.0…

7月17日，在2026世界人工智能大会（WAIC）期间，腾讯云WorkBuddy与李未可科技宣布达成战略生态合作，并发布首款接入WorkBuddy的X-AI记忆眼镜。这也是WorkBuddy硬件生态迈出的关键一步。X-AI记忆眼镜搭载自研WakeeMemory OS，能够持续感知真实工作场景，整理后的信息，将自动同步至WorkBuddy。基于长期积累形成的工…

“Context bombing” tricks malicious AI agents into shutting down before they can do harm.

36氪获悉，寻汇Sunrate与万事达卡在WAIC现场联合发布白皮书《超越自动化：定义智能体驱动的全球支付》。该报告系统阐述了“AI智能体”如何重塑B2B跨境支付全链路。传统模式下，企业财务需人工核验海外供应商账户、比对合同发票、择汇并承担T+2以上结算滞后期。该报告指出，AI智能体可自动提取多格式票据、匹配采购订单、基于企业需求推荐最优支付路由与换汇窗口、…

2026 世界人工智能大会（WAIC 2026）于 7 月 17 日正式开幕。作为全球人工智能领域的顶级盛会，本届大会以“智能伙伴 共创未来”为主题。阶跃星辰董事长、千里科技董事长印奇作为特邀嘉宾出席大会开幕式并在大会主论坛（上午场）发表主题演讲《当智能体进入物理世界》。回顾 15 年 AI 创业历程，他表示，AI 创业已从小众赛道成为全球重要共识。今天的…

From AI workflows to battery life and security, here's what it's really like to live with Vertu's luxury foldable every day.

AI 点评 · 奢侈品牌Vertu将AI助手定价6880美元，测试其性能能否匹配高端定位，看点在于AI功能能否支撑起

AI 点评 · Step AOS系统定义新交互范式，智能体原生设计或颠覆传统设备体验。

Training API-calling large language model (LLM) agents demands massive amounts of high-quality trajectories. However, collecting such data at scale typically requires fully implemented environments wi…

In this post, we walk through a few ways that Quick delivers on this promise. We cover the entire sales cycle, from identifying your highest-priority prospect, contacting them, wor…

AI 点评 · 亚马逊Quick将AI销售助手覆盖全流程，从筛选客户到沟通签约，真正实现销售智能化升级。

LLM powered multi-agent systems (MAS) have emerged as a promising paradigm for complex tasks. However, their advantages over single-agent systems (SAS) remain unclear, with performance varying inconsi…

State machine replication (SMR) and Byzantine fault-tolerant (BFT) consensus guarantee agreement despite a bounded number of arbitrary, colluding faulty participants. However, these guarantees rely on…

Lowest cost per token from extreme codesign maximizes intelligence per dollar for post-training in the agentic era.

AI 点评 · 极致软硬协同设计降低每token成本，让后训练阶段的智能性价比达到新高度，是智能体AI落地的关键指标

AI 点评 · AI智能体开发能力初显，但校验短板暴露了自动化落地的关键瓶颈。

IT之家 7 月 17 日消息，7 月 17 日，在 2026 世界人工智能大会（WAIC）上， 国家超算互联网 发布了科学计算智能体生态共创与开发者招募合作计划（以下简称“智能体共创计划”）。 该计划为期半年 ，通过面向高校科研院所、个人开发者及企业研发团队招募智能体、科研垂类模型、 MCP 工具 、Skill 等成果或合作意向，构建以国产超智融合算力为底…

36氪获悉，7月17日，2026世界人工智能大会暨人工智能全球治理高级别会议在上海启幕。连续九届参展的腾讯以“Hey，我的AI Buddy”为主题，集中展示AI在各领域进化为生产生活好搭档的跃进，与本届大会“智能伙伴 共创未来”的主题相契。

北电数智带来AI赋能民生的真实答卷

7月17日，2026世界人工智能大会在上海开场。作为36氪连续第三年深入WAIC现场的重要内容窗口，「氪话未来」直播间也在大会首日同步开启现场对话。 森博科技董事长于林义 在WAIC现场接受36氪「氪话未来」特邀专访，围绕企业级AI、智能体落地、行业know-how与业务闭环等话题，分享了森博从营销服务公司转向AI驱动科技服务公司的实践路径。 本届WAIC以…

人工智能正在进入一个新的产业周期。 过去一年，大模型能力持续演进，生成式AI、多模态交互、智能体等技术方向快速推进；而具身智能也从早期的技术探索阶段，逐渐步入产业验证的深水区，机器人开始成为人工智能与现实世界的重要载体。 市场率先给出了回应。据36氪研究院测算，中国具身智能市场规模已从2018年的2133亿元增长至2025年的9150亿元，2026年有望突破…

IT之家 7 月 17 日消息，2026 世界人工智能大会首日，千问推出两大硬件：千问 AI 眼镜将升级为智能体眼镜，千问首款 AI 智能体耳机也同步亮相。 据介绍，升级后的眼镜可通过智能体强化服务与决策能力，并能按需调用第三方 Skill 和 Agent。为了增强智能体眼镜对物理世界的感知与交互能力，千问推出 全双工语音、眼动追踪、体征监测 等一系列全新技…

AI 点评 · 阿里千问联手Bose，AI眼镜与耳机走向智能体时代，软硬协同创新值得关注。

36氪获悉，7月17日，在2026世界人工智能大会（WAIC）现场，支付宝与阶跃达成AI Agent系统级合作。双方围绕阶跃STEP-X原生AI终端展开深度协同，用户通过自然语言即可调用AI版支付宝“阿宝”连接真实服务，实现跨应用、多任务执行，推动智能体迈入“跨端互联办事”新阶段。

AI 点评 · AI终端与支付场景深度融合，开启跨应用多任务执行新纪元，实用性和商业价值显著。

36氪获悉，7月17日，WAIC2026期间，科大讯飞发布智能交互服务Agent——GuideX。区别于传统数字人，GuideX融合“全模态感知、自治理Agent、SkillHub”等核心能力，打通“感知、理解、执行、记忆、共情”服务全链路。

AI 点评 · 科大讯飞推出全模态感知Agent，突破传统数字人局限，引领智能服务新范式。

IT之家 7 月 17 日消息，今日， 在 2026 世界人工智能大会（ WAIC ）现场，支付宝与阶跃达成系统级合作，AI 版支付宝“阿宝”与阶跃大模型及其原生 AI 终端，可实现跨端互联。 未来，无论是与阶跃大模型对话，或在其原生 AI 终端上，都不用打开 App， 一句话就能向其自有智能体派活 ，再转交阿宝办妥，完成跨应用、多任务执行。 据官方透露，合…

AI 点评 · AI生态打通，跨应用多任务执行，开启无需打开App的智能生活新场景。

IT之家 7 月 17 日消息，2026 世界人工智能大会暨人工智能全球治理高级别会议主论坛今日在上海举行。会上，中国网信办会同有关方面正式提出《智能体互信互联互操作全球合作倡议》。 该倡议旨在释放智能体赋能可持续发展的潜力，防止形成智能鸿沟，凝聚各方共识，与全球伙伴共同打造开放、可信、安全、普惠的智能体生态。 智能体作为人工智能时代最具变革性的技术形态之一…

AI 点评 · 聚焦智能体生态的全球治理，推动开放互信，防范技术鸿沟，意义深远。

36氪获悉，7月17日，在2026世界人工智能大会（WAIC 2026）上，网易智企携全新升级的一站式企业AI应用服务亮相，集中展示AI Agent编排、AI Coding、AI客服、AI私域助理、AI智能数据与AI Agent安全等企业级AI能力，围绕安全治理、组织协作与业务增长三大场景，呈现企业级AI应用实践。

AI 点评 · 展示企业级AI全栈能力，聚焦安全治理与业务增长三大场景，为行业提供可落地的AI应用标杆。

IT之家 7 月 17 日消息，2026 世界人工智能大会（WAIC 2026）今天在上海举办，IT之家第一时间来到阶跃展台，看到了阶跃终端首款智能体手机 STEPX Neo，下面为大家带来现场实拍： 从现场实拍可以看到，这台手机目前戴着橙黄色的保护壳， 运行智能体原生系统 Step AOS 。其桌面 UI 采用近年来较为流行的圆角矩形图标，部分第三方应用的…

AI 点评 · 智能体手机首次深度联动办公App，展示了AI原生系统的落地可能。

Snapshot testing for LLM apps and agents, built to run locally and block regressions in CI.

AI 点评 · 百度AI新突破，智能体全家桶成WAIC焦点，展现技术实力与应用前景。

In this post, we walk through the three pillars that make this possible: simplified setup, smarter retrieval, and production readiness. We also show you code examples for setting u…

AI 点评 · 简化企业级AI搜索搭建流程，提升智能体知识库效率，降低开发门槛。

AI 点评 · RISC-V架构在AI算力需求下获资本押注，智能体时代催生新硬件机遇。

AI 点评 · 开源模型AI代理工具，降低使用门槛，推动本地化智能应用。

Security-agent evaluations commonly measure peak offensive capability under generous inference budgets, emphasizing vulnerability discovery, exploit development, penetration testing, and CTF completio…

Muon is competitive with AdamW in large-scale pre-training, but its value for reinforcement-learning (RL) post-training remains unclear. We study vanilla Muon in sparse-reward agentic RL through match…

Despite strong capabilities in data understanding and decision-making, autonomous data science agents still heavily rely on trial-and-error workflows that involve expensive computation. This bottlenec…

This post covers what makes Grok 4.3 a great fit for agentic and enterprise workloads, how you access it through Amazon Bedrock, and how to use the capabilities most teams reach fo…

AI 点评 · Grok 4.3登陆AWS，为智能代理和企业场景带来新选择，看点在于云上部署的便捷性与性能优势。

Across 107 enterprises, AI agents are being given real access to systems and data while the controls meant to contain them lag behind. More than half have already had a confirmed a…

AI 点评 · 54%企业已发生AI代理安全事故，但多数仍允许共享凭证，暴露安全管控严重滞后。

Security-agent evaluations commonly measure peak offensive capability under generous inference budgets, emphasizing vulnerability discovery, exploit development, penetration testing, and CTF completio…

AI 点评 · 成本感知评估揭示安全AI实用门槛，突破传统成功率局限，更贴近真实攻防场景。

Retrieval systems are trained and evaluated on a static idea of usefulness: hand a document and a question to a reader model, see whether the answer improves, and score the document accordingly. The i…

AI 点评 · 静态检索与多步智能搜索的因果效用脱节，挑战现有评估标准，值得关注。

Evidence synthesis is crucial for turning primary research into reliable knowledge for science, medicine, education, and policy. Yet, quantitative evidence synthesis remains largely manual and difficu…

AI 点评 · AI自动化元分析系统，极大提升科研效率，降低人工成本，推动循证决策发展。

Large language models (LLMs) increasingly serve as high-level planners for embodied agents, where linguistically benign instructions can become unsafe once grounded in the physical world. We study whe…

AI 点评 · 探索大模型安全新维度，揭示文本安全与物理行动的鸿沟，为具身智能风险防控开辟新思路。

Across 101 enterprises, the infrastructure that feeds AI agents their business context is being built faster than it can be trusted. Retrieval-augmented generation is already the d…

AI 点评 · 企业AI信任缺失比检索问题更致命，多数公司却仍在错误方向投入资源修补。

Across 157 enterprises, organizations are granting AI agents more autonomy while trusting the evaluations meant to gate that autonomy less. Half have already shipped an agent that…

AI 点评 · 企业盲目放权AI代理却缺乏可信评估，暴露出部署与安全之间的严重脱节。

AI 点评 · 多智能体协作编程，突破单Agent局限，是AI自主开发的关键一步。

AI 点评 · 英伟达新嵌入模型登顶基准，加速智能体检索技术落地，AI搜索能力再升级。

AI 点评 · 聚焦Agent可靠性工程，为AI应用落地提供核心技术支撑。

DoorDash is opening a limited beta of dd-cli, a command-line tool that lets developers and AI agents search stores, build carts, and place orders from the terminal, marking another…

AI 点评 · 命令行点外卖，AI代理可直接下单，开启餐饮服务新交互方式。

AI 点评 · AI测试自动化新范式，智能体驱动显著提升UI测试稳定性。

Tool: Mermaid to Unicode box art (grok-mermaid) While exploring the codebase for the newly open-sourced Grok CLI coding agent I came across xai-grok-markdown/src/mermaid.rs , a "se…

AI 点评 · 将Mermaid图表转为Unicode字符画，适合终端环境，实用且有趣。

Cars24 uses OpenAI-powered voice and chat agents to handle 1M+ monthly conversation minutes, recover 12% of lost leads, and bring agentic workflows to teams across the company.

AI 点评 · 用AI每月处理百万分钟对话，挽回12%流失客户，汽车电商实现业务流程自动化升级。

字节跳动联合中兴努比亚打造的首款AI智能体手机（“豆包AI智能体手机”）今年将有多款机型发布，其中一款将于2026世界人工智能大会期间亮相，其整体备货约20万台，首批备货10万台以内，截至发稿中兴方面对此消息暂无回应。（界面）

AI 点评 · 字节联手品牌进军硬件，多款AI手机布局显示生态野心，值得关注市场反应。

IT之家 7 月 16 日消息，OpenAI 今天（7 月 16 日）携手 Work Louder，合作推出 kbd-1.0-codex-micro 键盘，售价为 230 美元（IT之家注：现汇率约合 1560 元人民币）。 IT之家翻译产品官方描述如下： kbd-1.0-codex-micro 键盘采用 Work Louder 设计理念，实现 AI 智能体…

AI 点评 · OpenAI跨界做硬件，AI专用键盘或将重塑人机交互方式。

IT之家 7 月 16 日消息，马斯克旗下 SpaceXAI 公司昨日（7 月 15 日）宣布开源 Grok Build， 并将源代码发布至 GitHub 平台。 在官方博文中，SpaceXAI 表示： 开源发布源代码，是构建强大、可靠框架的最直接方法。用户可以阅读源代码，了解其从上下文构建到工具调用分发的完整工作原理。 开源也让框架更容易探索和扩展：如果用…

AI 点评 · 开源代码降低使用门槛，推动编程AI智能体技术快速迭代，值得开发者关注。

Across 101 enterprises, agent orchestration is consolidating onto model-provider platforms — Anthropic’s Claude leads by a wide margin — chosen for the gravity of the underlying mo…

AI 点评 · 企业AI部署的瓶颈不在平台，而在于将聊天机器人误称为智能体的认知错位。

AI 点评 · 用Agent手机概念展示商业落地能力，为IPO估值注入强心剂。

Recent advances in Tool-Integrated Large Language Models have made web search a core capability of information-seeking agents. However, as interaction histories grow, agents increasingly struggle to t…

Large language models are increasingly trained as interactive agents for long-horizon tasks involving multi-turn interaction, tool use, and environment feedback. Outcome-based reinforcement learning (…

Skills are a useful abstraction for software agents, turning human and agent experience into reusable procedural knowledge. Yet existing skill libraries are mostly hand-written, text-centric, or deriv…

OpenAI, which is in the middle of a legal battle with Apple over hardware trade theft allegations, just released a light-up keyboard designed to be paired with its agentic coding a…

AI 点评 · 硬件纠纷未平却推高价键盘，OpenAI跨界硬件野心与Codex生态绑定值得关注。

Built partnered with the AWS Generative AI Innovation Center (GenAIIC), AWS Partner AND Digital, and AWS account teams to create a scalable, AI-powered document processing engine t…

AI 点评 · AI与云服务结合，重塑房地产金融文档处理，效率与准确性双提升。

In this post, we walk you through the Computer Vision MCP Server, which illustrates this approach, representing how AI systems can process visual information and make intelligent d…

AI 点评 · 亚马逊Bedrock配合MCP服务器，打通视觉智能关键环节，实用方案值得开发者关注。

AI 点评 · 从实战中总结的智能体构建经验，为AI应用开发提供宝贵参考。

Agentic coding tools are increasingly capable of generating and submitting pull requests (PRs) to software projects, introducing new forms of human-agent collaboration in software development. While p…

Most reported gains from agent-optimization methods are one-shot: an agent is optimized against a fixed benchmark and the resulting improvement is reported as if it were a stable property of the metho…

The rapid proliferation of Agentic Artificial Intelligence fundamentally disrupts traditional customer loyalty paradigms. As AI evolves from passive recommendation algorithms to autonomous, goal-direc…

Multi-turn agents solve complex tasks through extended sequences of tool interactions before producing a final answer, making credit assignment a fundamental challenge during post-training. Outcome re…

The Codex Micro is designed to monitor multiple agentic threads at a glance.

Hey HN, we’re Nitish and Prateek, the founders of Coasty ( https://coasty.ai/computer-use ). We’re building computer-use agents that can complete workflows inside legacy desktop so…

IT之家 7 月 15 日消息，努比亚手机官方今日公布了旗下全球首款 AI 智能体手机的局部外观，新机将在 WAIC 2026 正式亮相。 预热图显示，努比亚全球首款 AI 智能体手机提供了一款淡粉配色，后盖中央印有“nubia”的字样，手机底部是扬声器开孔、USB-C 接口和 SIM 卡槽。遗憾的是，手机上端的摄像头模组被遮挡了，无法看见具体样式。 不过，…

The guy behind TCP/IP is working on a standard for identifying AI agents in the wild.

大公司： 美团、青桔、哈啰共享单车调价 近期，美团单车、滴滴青桔、哈啰单车相继在北京等多个城市上调计费规则，三大平台不约而同地采取了“提高起步定价、拉长基础骑行时长”的组合策略：起步价从此前的1.5元/30分钟左右，普遍调整为1.88元至1.99元/60分钟。这成为共享单车行业近年来较大范围的一次集体调价。（金融时报） 瓜子二手车线下直卖场首店今日正式开业…

作者 | 王晗玉 编辑 | 张帆 支付宝首页调出AI界面，对话框取代了密密麻麻的小程序；用户对着“阿宝”说一句“找附近的奶茶优惠券”，周边门店的活动自动匹配好，核销下单一步完成。 最近，支付宝完成了上线22年来最大一次改版。 本月初，AI版支付宝“阿宝”正式开启全量公测，几乎同一时间，微信支付“AI专属卡”也在智能体WorkBuddy中落地。 进入2026年…

An intentionally vulnerable OWASP LLM Top 10 training platform for AI Security, Prompt Injection, RAG Security, Agent Security, and GenAI penetration testing.

Local-first conversational AI video editor with a professional multi-track timeline, Agent Skills, MCP integration, and Remotion-powered rendering.

作者 | 乔钰杰 编辑 | 袁斯来 硬氪获悉，上海追知工程科技有限公司（以下简称“追知工科”）近日完成数千万元种子轮融资，由L2F光源创业者基金、尚融资本、一村资本联合投资。本轮融资将主要用于核心产品研发、团队建设及市场拓展。 追知工科成立于2024年2月，是一家 聚焦垂域工业智能体 的科技企业，同时也是上海交通大学成果转化企业、上海人工智能研究院战略孵化企…

As Large Language Models (LLMs) evolve into autonomous agents, the need for unified evaluation infrastructure becomes critical. However, current evaluation pipelines remain highly fragmented and tight…

OpenClaw has emerged as a leading agent framework for complex task automation, yet it faces insufficient cross-platform GUI interaction support and a well-built self-evolution mechanism. These flaws l…

Embodied cognition requires agents to connect high-level task reasoning with the physical states to be achieved. We introduce Hy-Embodied-RxBrain, an embodied cognition foundation model with joint lan…

Healthcare spans high-stakes communication, expert reasoning, and workflow execution, yet specialized LLMs that cover these use cases together remain limited. A healthcare model must handle patient co…

Agentic language models must learn when to call tools, when to consume tool responses, and when to answer directly. This makes multi-teacher on-policy distillation a natural training strategy: one tea…

This post shows how Thrad.ai deployed a multi-agent system with Strands Agents and Amazon Bedrock AgentCore that automates the pipeline from prospect discovery through personalized…

AI 点评 · 多智能体协同与云平台结合，实现从线索挖掘到个性化服务的全流程自动化。

董事长印奇称“未来的OS一定是跨端的”

AI 点评 · 多模态Agent时代，阶跃以跨端OS切入，或将重塑AI生态格局。

Large language model (LLM) agents increasingly automate multi-step engineering and informatics workflows, yet they rarely ask how much effort a task actually requires. They often follow a maximum-cont…

Training robust autonomous driving agents requires a simulator that is fast enough for reinforcement learning at scale, realistic enough to ground behavior in real-world map structure, and diverse eno…

Large Language Model (LLM) agents have moved beyond generating responses to executing multi-step tasks by calling tools, observing the results, and iteratively deciding the next action. Most agent sys…

In this post, we extend that foundation to demonstrate how QA Studio addresses batch regression testing and pipeline integration through test suites that organize and parallelize e…

AI 点评 · 用Amazon Nova Act实现智能体自动化测试，大幅提升批量回归效率与流水线集成能力。

AI 点评 · 从数据到行动，Snowflake AI让企业自主决策，工作效率将被彻底重塑。

Hey HN, we’re Shubham & Parth, childhood friends building Agnost AI ( https://agnost.ai ), product analytics for teams building chat and voice agents. We read production conversati…

大公司： 中国神华：预计上半年净利润同比增长6.9%-21.1% 36氪获悉，中国神华公告，预计2026年上半年归属于上市公司股东的净利润为263亿元至298亿元，同比增长6.9%-21.1%。业绩变动主要系煤化工业务量及自有铁路、港口、航运业务量增加，带动相关业务利润同比增长。 中国人寿：预计上半年净利润同比增长约215%-235% 36氪获悉，中国人寿公…

凌晨三点，一家刚成立不久的AI创业公司，可能已经在同时服务旧金山的客户、采购首尔的技术服务，并与拉各斯的合作伙伴签下合同。这家公司甚至还没有招到第一名全职财务人员，业务却已经跨越多个市场、币种和监管辖区。 AI正在让这样的创业路径成为可能。过去需要市场、运营、客服等一整套全球化团队才能完成的工作，现在借助智能体就能承担相当一部分。新一代初创企业不必再按照“先…

Learn how enterprises can manage AI investments in the agentic era by measuring useful work per dollar, improving efficiency, and scaling high-value workflows.

Open-source app builder engine — intent to working app

Agent携上百个Skills助我当大导演

I built FixBugs, an agent that ingests the rich context surrounding production bugs to reproduce them in a sandbox and generate verified fixes. It's available in the form of a self…

The company is raising at least $75 million, led by Robot Ventures, with significant participation from USV and other prominent investors.

Coding agents must integrate external tool returns into ongoing reasoning - a capability that standard left-to-right pretraining on code exposes only in its forward direction. We observe that the acti…

AI pentesting agents are increasingly credible as offensive security systems, but current benchmarks still provide limited guidance on which will perform best in real-world targets. Existing evaluatio…

Self-improving autonomous agents are moving from research prototypes to deployed systems. The primary goal is controllable evolution, or adaptation, from experience with minimal or even no human input…

Large Language Model (LLM) agents have moved beyond generating responses to executing multi-step tasks by calling tools, observing the results, and iteratively deciding the next action. Most agent sys…

Failure attribution for LLM-based agentic systems, i.e., identifying which steps in a failure trajectory caused the task to fail, is critical for debugging and improving these systems. Existing approa…

The capability of a modern AI agent depends not only on its foundation model but also on its harness, which constructs prompts, manages state, invokes tools, and coordinates execution. As models, APIs…

Despite the success of Vision-Language Models (VLMs), misleading charts remain a significant challenge due to their deceptive visual structures and distorted data representations. We present ChartCyni…

We revisit the evaluation of automatic harness evolution for LLM agents. Existing harness evolution methods use unit test cases to search for harness configurations and then report final performance o…

Code review helps maintain software quality before code integration, but it also imposes a substantial workload on human reviewers. As generative artificial intelligence becomes part of software devel…

Building assistants that can continually watch the world, remember what they see, and reason over their accumulated experience is a long-standing goal, and recently multimodal agents equipped with lon…

“SceneSmith” system uses collaborative AI agents to create realistic 3D environments of places like kitchens, hotels, and living rooms, where robots can simulate everyday chores.

AI 点评 · AI代理自动生成3D训练场景，大幅降低机器人数据采集成本，加速家务机器人落地。

AI 点评 · Meta开源编程Agent模型，从免费转向低价商业模式，标志AI编程工具商业化加速。

AI 点评 · 用AI改造经典玩具，Strands Agents让互动鱼挂件有了新玩法。

In this post, we describe how Bluesight used two AWS engagements and Amazon Bedrock AgentCore to evolve from a single-product AI prototype to Prism, a unified agentic AI solution s…

AI 点评 · Bluesight借助亚马逊Bedrock将AI原型升级为统一代理系统，展示了企业级AI落地的实战路

Building multi-tenant agents with Amazon Bedrock AgentCore and Apply fine-grained access control with Bedrock AgentCore Gateway interceptors establish the conceptual foundation for…

AI 点评 · 用Bedrock实现多租户代理的细粒度权限控制，解决了真实业务场景中的访问隔离难题。

We introduce MM-ToolSandBox, a benchmark and evaluation framework for visually grounded tool-calling agents. The framework provides a stateful execution environment spanning 500+ tools across 16 appli…

Shared meaning in language requires people to learn and agree on categories. We ask how characteristics of agents' memories change the emergence and evolution of shared meaning. Without a coordination…

出自中国团队

"Context bombing" tricks hacking agents into shutting down before they can do harm.

AI 点评 · Agent让可观测性从被动发现问题转向主动修复，这是运维智能化的关键跃迁。

同时发布CPU原生液冷整机柜、多模融合超节点

Observable diagnostics, failure attribution, metrics, and no-key replay for multi-agent runtimes.

Exploration is essential for reliable autonomy in multi-agent systems, yet it remains unclear whether large language model (LLM) agents can explore effectively when interacting with one another. We sh…

IT之家 7 月 13 日消息，由中国互联网协会网民权益和个人信息保护工作委员会主办的 2026（第二十五届）中国互联网大会网民权益和个人信息保护论坛于 7 月 9 日在京举办。 中国互联网协会移动互联网工作委员会和中国互联网协会网民权益和个人信息保护工作委员联合发布《智能体个人信息保护自律公约》，现场来自 百度、腾讯、阿里、火山引擎 等 31 家互联网企业…

AI 点评 · 科技巨头联合签署，为AI智能体划清数据安全红线，行业自律迈出关键一步。

LLM-based coding agents have significantly advanced automated software issue resolution, yet they remain highly prone to factual errors caused by insufficient repository understanding. Recent methods…

Hello HN, I don't post on here much, but wanted to get some eyes on a new project I'm just launching. I think we definitely need one more AI code agent.. I'm a long-term C++ dev, a…

AI 点评 · 生产级AI代理迁移至GPT-5.6，实现速度翻倍且成本降低27%，性能与成本平衡的行业标杆。

IT之家 7 月 12 日消息，Meta 于 7 月 9 日正式发布适用于 AI 智能体的多模态推理模型 Muse Spark 1.1 版本，重点提升了模型在智能体任务中的规划、协同与执行能力，并增强了工具调用、代码开发、应用操作能力。 Meta 表示，Muse Spark 1.1 强化了多智能体协作机制，由主智能体负责收集信息、制定计划，再将任务拆分并分配…

AI 点评 · 多智能体协作机制是AI落地的关键突破，Meta这次强化了任务拆解与分工能力。

AI 点评 · 3D可视化编码过程，直观追踪AI代理如何理解代码库，革新调试与协作方式。

Large language model (LLM) agents are beginning to automate machine learning engineering (MLE) by coupling planning, code execution, debugging, and empirical feedback. Translating this capability to m…

AI 点评 · AI自主管理引发代理控制权归属的核心争议。

36氪独家获悉，7月11日，智谱创始人唐杰，在智谱发布了主题为《巨浪已来》的内部信。其中提到，智谱将不追求短期的应用变现，而是直指AGI的下一个高地：长程任务能力、完全自治的智能体系统、自我进化、极致安全治理。过去半年来，智谱收获了创立以来的高光时刻：市值较半年前上市初期涨了10倍，并在2026年 6月，跻身“万亿港元俱乐部”。

AI 点评 · 聚焦AGI核心突破而非短期变现，揭示智谱从估值飙升到技术深水区的战略转型。

Recent VLM and VLA systems have improved robotic perception and action prediction, yet long-horizon embodied agents still require a general runtime layer for reasoning, memory, tool use, verification,…

AI 点评 · K8s动态配置新实践，Airbnb开源方案揭示云原生服务治理前沿。

7月11日，有消息称腾讯正在洽谈成为通用AI Agent公司Manus的最大股东，据该消息，由腾讯牵头的中方资本组团以约20亿美元估值从Meta手中回购Manus的全部股权。记者向腾讯方面求证，截至发稿腾讯方面暂无回应。另有知情人士向记者透露，此次交易后，腾讯仍将保持少数股东地位，但不会控股。（南方都市报）

AI 点评 · 腾讯罕见以少数股东身份入局AI Agent，或意在布局生态而非控制，战略意图值得玩味。

神话级大模型驾驭宝典

AI 点评 · 突破性验证了超大模型在复杂推理任务中的协同能力，或开启AI解决数学难题新纪元。

今日热点导览 “全球首款智能体手机”已备货8万至10万台？知情人士：假的 百亿私募数量达142家，再次刷新历史纪录 三星李在镕拟于7月底赴美会晤英伟达黄仁勋 德国大众拟大裁员，最高或裁减12万个岗位 OpenAI高管层再现变动，首席运营官因病离职 TOP3大新闻 长鑫科技，承销团阵容公布 长鑫科技IPO进入发行倒计时，这家“国产存储第一股”背后的承销团阵容也…

AI 点评 · 国产存储芯片龙头IPO加速，承销团阵容披露凸显市场关注热度。

Agentic retrieval-augmented generation (RAG) extends static RAG by allowing language models to iteratively reason, generate search queries, retrieve evidence, and predict answers. However, it remains…

Internet of Things (IoT) systems are inherently vulnerable due to constrained hardware, outdated firmware, and insecure default configurations, creating a need for scalable and adaptive security testi…

AI 点评 · 用AI代理自动挖掘物联网漏洞，突破传统测试瓶颈，提升安全检测效率。

We present our submission to the QANTA 2026 shared challenge at the ICML 2026 Workshop on Efficient Multimodal Question Answering (EMM-QA). Quanta evaluates multimodal quizbowl systems that answer pyr…

AI 点评 · 置信度校准与增量推理结合，为多模态问答提供高效新思路，技术方案值得关注。

Enhancing the reasoning capabilities of large language model (LLM) agents requires effective orchestration of diverse expert models and tools. However, existing frameworks typically call APIs based on…

AI 点评 · 用拍卖机制分配任务，提升大模型推理效率，为多智能体协作提供新思路。

The proliferation of agentic AI systems across enterprise and public-sector contexts has outpaced the capacity of general-purpose AI risk frameworks to classify and govern them. In this paper, we intr…

AI 点评 · 首个针对企业内部AI代理系统的分级风控框架，填补了通用框架在代理型AI治理上的空白。

IT之家 7 月 10 日消息，SK 海力士今日在纳斯达克挂牌交易其美国存托凭证（ADR）。随后，SK 集团会长崔泰源接受了彭博社与 CNBC 的采访。 IT之家注意到，崔泰源表示，在人工智能时代，内存行业已进入结构性增长阶段。 过去，内存的需求主要取决于人口数量或是智能手机和个人电脑的销量。 然而，随着 AI 智能体、推理过程中产生的键值缓存（KV Cac…

AI 点评 · AI引爆内存需求结构性变革，产能翻倍仍供不应求，预示行业进入长期高景气。

Turn one topic into a finished Vox-style paper-collage explainer/ad video — automated end to end on Atlas Cloud + ffmpeg. An agent skill.

In this post we show how to build a semantic layer on AWS using Stardog’s Semantic AI Application over Amazon Aurora and Amazon Redshift, and how to run a Strands Agents agent on A…

AI 点评 · 企业级AI与知识图谱深度结合，打通数据语义理解到智能决策的完整链路，值得关注。

AI 点评 · Agent将可观测性从被动诊断转向主动干预，颠覆传统运维思路。

In this post, we show you how to combine case management with agentic automation capabilities in Quick Automate. We introduce case management and explore the lifecycle of cases in…

AI 点评 · 结合案例管理与智能自动化，提升企业复杂任务处理效率，降低人工干预成本。

Evolving from a traditional software as a service (SaaS) platform into a next-generation agentic AI platform meant orchestrating multiple specialized agents across long-running ent…

AI 点评 · KTern.AI用亚马逊Bedrock实现SAP多智能体协作，展示企业级AI落地新路径。

AI 点评 · AI智能体面临安全风险时，企业的核心防线是人机协同和伦理规范。

agent时代来了

A practical, open-source guide to mastering WorkBuddy through real-world workflows.开源的 WorkBuddy 实战蓝皮书：教程、真实工作流、Skills、MCP、自动化与多智能体实践。

OpenAI 发布 GPT-5.6 系列模型等，SpaceXAI 发布编程智能体模型 Grok 4.5。 查看全文

AI 点评 · 大模型竞速加剧，头部企业密集发布新品，技术迭代与市场格局值得追踪。

“IT早报”时间，大家好，现在是 2026 年 7 月 10 日星期五，今天的重要科技资讯有： 1、OpenAI 最强 AI 模型：GPT-5.6 系列正式上线，纳德拉称微软 Copilot 同步接入 OpenAI 公司 7 月 10 日发布公告，宣布在 ChatGPT（聊天机器人）、Codex（主打编程 AI Agent，目前朝通用 Agent 方向）以及…

AI 点评 · OpenAI模型重大升级，微软深度整合，预示AI竞争进入新阶段。

IT之家 7 月 10 日消息，在接受 CNBC 采访时，OpenAI 首席执行官萨姆 · 奥尔特曼（Sam Altman）表示，在 AI 智能体编程任务中， GPT-5.6 Sol 模型表现比市场主流竞争模型“一样好，甚至更好”，但 Tokens 效率提高 54%。 IT之家注：原文中并未具体指名市场主流竞争模型，不过鉴于 GPT-5.6 Sol 模型的定…

AI 点评 · 性能飞跃与成本降低同步实现，这项突破将加速AI应用落地。

IT之家 7 月 10 日消息，OpenAI 今天（7 月 10 日）发布博文，在宣布推出 GPT-5.6 系列 AI 模型的同时，还推出全新的 ChatGPT Work 智能体， 由 GPT-5.6 提供支持，定位为可承担长时、多步骤任务的智能体。 在博文中，OpenAI 披露了 Codex 的现有使用规模：官方数据显示，Codex 每周用户数已超过 50…

AI 点评 · GPT-5.6加持的Work智能体，标志着AI从对话迈向长周期复杂任务自主执行。

IT之家 7 月 10 日消息，OpenAI 公司今天（7 月 10 日）发布公告， 宣布在 ChatGPT（聊天机器人）、Codex（主打编程 AI Agent，目前朝通用 Agent 方向）以及 API 中上线 GPT-5.6 系列模型。 在模型方面，IT之家援引博文介绍，OpenAI 本次共发布 3 档模型： 旗舰版 Sol（太阳）：每 100 万 T…

AI 点评 · 微软同步接入，意味着AI竞争格局突变，企业级应用迎来新拐点。

Lyzr, a startup that builds AI agents for enterprises, used its own AI agent to raise a $100 million round — proof, evidently, that the product actually works.

AI 点评 · AI代理成功完成融资，证明企业级产品真实可用，开创行业先河。

OpenAI is sunsetting its AI-powered browser after less than a year. But it's moving some agentic browsing features to its desktop app and a Chrome extension.

AI 点评 · 关闭自研浏览器，但将核心功能整合进桌面端和插件，表明战略收缩而非放弃。

Meta's pitch to users is Spark's ability to handle large agentic workloads, fix bugs, and help with large code migrations — the kind of automation that enterprises are increasingly…

AI 点评 · Meta携Spark 1.1切入企业级AI编码自动化，专注大型代码迁移和修复，展现巨头竞争新方向。

AI 点评 · AI从单轮对话迈向多智能体协作，预示人机协同新范式即将到来。

Introducing Muse Spark 1.1 Following Muse Spark in April , here's Muse Spark 1.1 - the first Spark model to offer an API. Meta claim significant improvements in agentic tool callin…

Hey HN! We built a browser-based agent that runs inside an authenticated web app, watches how the app calls its own APIs, and automatically turns those into agent tools. You can th…

Hi Hacker News, I’m Yahia. I built Context.dev ( https://www.context.dev/ ) to make it really easy to integrate web data into your products and agents. Here’s a demo video: https:/…

AI 点评 · 让任意网页变成结构化API，极大降低AI应用获取实时网络数据的门槛。

ChatGPT Work is an agent that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work.

AI 点评 · ChatGPT从对话助手跃升为跨应用自主执行任务的智能代理，标志AI进入主动工作阶段。

Fully local vulnerability research pipeline - 14B code-specialized LLM reviews every source file exhaustively.

Evaluate & benchmark AI coding agents and Claude Code skills — sandboxed, reproducible YAML eval suites for Claude Code, Codex & Gemini, with A/B experiments an…

Notion 推出全新应用 Agents、Jolla Phone (2026) 手机正式发售等。 查看全文

IT之家 7 月 9 日消息，SpaceXAI 今日正式发布了其 Grok 4.5 模型，这是该公司首个专门针对编程和智能体任务训练的模型。 据介绍，该模型由 SpaceXAI 与 Cursor 联合完成训练，在提供前沿智能水平的同时，兼具领先的速度与成本效率。马斯克将其称为“Opus 级模型”。 Grok 4.5 面向真实工程场景设计，擅长处理大型代码库以…

AI 点评 · 编程智能体成本砍半效率翻倍，马斯克联手Cursor的定价策略才真值得行业关注。

In long-horizon tasks, decision-relevant state is often scattered across an expanding trajectory, while the action agent must surface it and act. As trajectories grow, task requirements, environment f…

The rapid development of large language models and multimodal large language models has accelerated the emergence of proactive agents capable of operating everyday tools and assisting users in real-wo…

Large language models (LLMs) increasingly act as integrated data-science agents, combining abstract reasoning with advanced tool use. Yet the relevant benchmark landscape largely divides into symbolic…

AI agents have become capable of autonomously completing short, well-specified tasks. However, existing terminal benchmarks largely focus on simple problems that finish within minutes and are evaluate…

Visual generators excel at rendering, but they confidently fabricate what they do not know. User requests are unbounded, evolving, and deeply long-tailed: new characters, trending entities, post-cutof…

The optimization of long-horizon agents increasingly relies on reflection-based mechanisms, where a large language model (LLM) acts as an optimizer to diagnose agent failures and improve agent policie…

Analytical workloads operating on data stored in external database systems face a fundamental bottleneck: data access is guarded entirely by the database driver, like JDBC or ODBC, forcing all reads t…

We introduce institutional red-teaming, an evaluation methodology for testing deployment rules in multi-agent AI: hold the agents, objectives, and task state fixed, vary only one rule, and attribute t…

I am done with articles stating "I used this LLM to do that", or "Look, this agent did that in 2 minutes!". I want content more user-centric, less openai / anthropic, and more "hum…

Data visualizations are the bridge between user and data. But building AI agents that can generate visualizations reliably can be very tricky: - simple chart specs can be reliable,…

Autonomous AI agents can execute complex tasks with limited human review, yet they often lack the grounded operational knowledge to make their outputs not just executable but correct, secure, and main…

AI 点评 · 数据驱动智能体的新范式，预示AI自主决策能力将迎来关键突破。

Founded in 2024, Prime Intellect’s goal is to give organizations capabilities to train their own agentic systems without relying on frontier AI labs.

Short chart specifications are easy to write, but often produce uninspiring results. Flint is an open-source visualization language that offers a middle path, letting AI agents cre…

NVIDIA Nemotron 3 Ultra is offering leading performance at lower cost than top closed models with the largest and most widely adopted AI agent orchestration platform. LangChain tun…

The SQLite of agent sandboxes — self-hosted, E2B-compatible. One machine, sandboxes that live forever, idle costs nothing.

7月7日，易居（中国）控股有限公司董事局主席、总裁周忻再一次来到台前，给公司的AI产品站台，推出其核心战略产品“地产模数通——企业专属大模型一体机”。同时，克而瑞地产AI分析师“小瑞”正式上岗。 不到两个月前，易居旗下的深度智联刚刚发布了全球首个房地产经纪人智能体“易居·小新”，用中立无佣模式替代传统中介。迹象显示，深度智联在加速AI产品落地。 周忻说，地产…

文｜胡香赟 编辑｜海若镜 36 氪获悉，德睿智药近期已完成 5200 万美元B轮融资，投资方包括头部人民币和美元基金，凯乘资本为独家财务顾问。募集资金将用于AI制药引擎Molecule Arts Platform（MAP）升级迭代，完善其多智能体（Multi-Agent）协同体系与临床数据闭环（Clinical Data-in-the-Loop），以及推进自…

Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs). Previous RL pipelines for LLMs were mostly synchronous and batch-interleaved, which is in…

Modern LLMs are increasingly deployed in long-context applications such as retrieval-augmented generation, repository-level coding, and agentic workflows whose accumulated reasoning and tool traces ro…

Realistic and diverse traffic simulation is essential to autonomous driving development. Yet prevailing benchmarks predominantly reward realism, and recent methods have optimized accordingly, leaving…

The optimization of long-horizon agents increasingly relies on reflection-based mechanisms, where a large language model (LLM) acts as an optimizer to diagnose agent failures and improve agent policie…

Training tool-use agents to improve from their own experience remains challenging, as supervised fine-tuning relies on fixed teacher-distilled trajectories, while sparse-reward reinforcement learning…

AI 点评 · AI自主决策CDN资源分配，标志内容分发网络从服务人类转向服务AI。

AI 点评 · 开源工具节省半数token，显著提升AI处理Word文档效率，开发者可快速集成。

In this post, we walk through how multi-dataset Topics work, explain how the chat agent uses defined relationships to generate cross-dataset queries, and demonstrate an end-to-end…

AI 点评 · 跨数据集统一语义层，让非技术用户也能自然语言查询，大幅降低数据分析门槛。

Large language model (LLM) agents solving multi-step tasks frequently commit to trajectories that are doomed to fail, yet continue to consume substantial inference compute before the failure becomes o…

This post walks through building a serverless image editor where users upload a photo, describe an edit in plain English, and receive the result in seconds. The agent runs on Agent…

AI 点评 · 无服务器图像编辑结合自然语言指令，大幅降低AI应用门槛，展示Bedrock AgentCore的实用

The dairy industry in Ireland has a large potential for the integration of renewable energy and the reduction of carbon emissions. However, researchers of distributed generation control are mainly foc…

In this post, you build an AWS Support Companion using Amazon Bedrock AgentCore. The agent uses Strands Agents as the orchestration framework and connects to AWS services through t…

AI 点评 · 用AgentCore快速搭建运维助手，降低AI应用开发门槛。

In this post, we show how AWS Finance used chat agents and Flows in Amazin Quick to transform two of their most time-consuming workflows.

Max single-threaded CPUs at scale are a new category of CPUs built for the agentic AI era. Across the creation and deployment of an agentic system, the CPU is on the critical path…

With the rapid progress of AI capabilities and the move to agentic systems, organizations are expanding their use cases as the technology continues to grow. That constant evolution…

Persistent AI rooms with governed, approval-gated memory: local-first, on your machine.

... government of the people, by the people, for the people ... — Abraham Lincoln, Gettysburg Address (1863) The cost of AI is dropping rapidly. GPT-4-class capabilities cost rough…

We’re announcing new capabilities in Managed Agents in Gemini API so developers can build reliable, production-ready agents.

Cognitive-structured Multimodal Agent (CMA-Harness): a memory-centric agent for long-horizon multimodal understanding, generation, and editing — externalizing v…

GitHubStar-history 最近Openclaw以25.2万星标，超越Meta的React登顶GitHub开源项目历史第一！ 要知道React是Facebook(现改名Meta）打造的经典前端框架，过去十余年间，互联网上绝大多数我们熟知的网站与App，底层技术架构皆由它构筑。Openclaw官方更是高调发文嘲讽Meta“我们在迭代创新，而你只在办会…

An AI agent carried out the technical execution of a real-world ransomware attack for the first known time, but new details show a human still chose the victim, set up the infrastr…

The Fable Workflow: how Claude Fable 5 worked, distilled into skills any model can run, with the eval that keeps it honest. Think / act / prove.

Coding agents increasingly generate pull requests (PRs) for real-world software issues, yet one-shot PR generation remains open-loop: the PR is proposed without systematic review, diagnosis, or revisi…

Embodied navigation aims to build agents that interpret multimodal goals, reason in 3D space, and reach target destinations reliably in the real world. However, progress remains constrained by the lac…

On-policy distillation (OPD) trains a student policy by matching a stronger teacher on the student's own trajectories, offering a promising framework for language agent training. However, its applicat…

We present AgentLens, a production-assessed benchmark for interactive code agents. Most code-agent benchmarks reduce a run to a single bit -- did the task pass? -- but the people who actually use thes…

Interactive simulators have become powerful tools for training embodied agents and generating synthetic visual data, but existing photorealistic simulators suffer from limited generality, programmabil…

Autonomous negotiation agents are increasingly deployed in high-stakes settings such as insurance and procurement. While cryptographic techniques protect explicitly disclosed constraint values, they f…

"The reality is, when you're optimizing for production, you start looking at a price/performance," Guillermo Rauch tells TechCrunch.

AI 点评 · 模型与代理分离是AI落地的关键一步，Vercel CEO的实战视角揭示了成本与性能的平衡之道。

Visual generators excel at rendering, but they confidently fabricate what they do not know. User requests are unbounded, evolving, and deeply long-tailed: new characters, trending entities, post-cutof…

Long-horizon agentic LLMs are increasingly limited by finite context windows, as extended interaction trajectories can exceed the maximum context length before a task is completed. Context compaction…

While recent Vision-Language-Action (VLA) models show promise toward generalist manipulation policies, they struggle with long-horizon tasks due to their Markovian nature-relying solely on current obs…

Personal agents are becoming persistent user-owned intermediaries: they remember preferences, filter platform-mediated information, use tools, and negotiate with services. Existing benchmarks evaluate…

We propose OptiAgent, a multi-agent framework that, given a natural language description of an Operations Research problem, is able to output a solver-ready mathematical formulation as well as executa…

Zero-dependency browser video editor that AI agents can drive — JSON timeline, MCP + REST, live-reloading UI

✨ The agentic motion layer — an open-source, chat-native motion engine. Describe the feeling; your AI ships the animation.

文｜吴思瑾 编辑｜邓咏仪 01 一句话介绍 北京治真治合科技有限公司成立于2024年，旗下产品「APTSell」（AI Power To Sales��希望成为AI版的CSO （Chief Sales Officer，首席销售官）。 简单来说，APTSell是一个组合式Agent，通过整合与可视化销售全流程数据，生成管理决策和执行建议，以期正向促进销售效率和…

阿里禁用 Claude 模型 索尼调整计划，2028 年前发售游戏可继续生产光盘 千问、豆包将下线智能体功能 Android 反垄断案欧洲终审败诉 混动车、商用纯电车将不再免征车船税 电商法修正案征求意见 看看就行的小道消息 少数派的近期动态 你可能错过的好文章 查看全文

Video understanding and self-verification for AI agents. Turn videos, streams, and agent screen recordings into searchable, timestamped evidence—then use THE LO…

For robots to work reliably in commercial and industrial applications, can recent advances in agentic coding systems combine interpretable robot programming with the open-world adaptability of model-f…

We introduce the first multiplayer world model for highly dynamic environments governed by complex physical interactions. Whereas single-player world models treat the other agents as part of the envir…

Agentic video understanding equips models with long-term memory to autonomously process and respond to continuous, long-horizon multimodal streams. However, advanced video agents often rely on ``detec…

50+ curated LLM observability tools PLUS 26 Agent Skills (several with runnable, unit-tested scripts) to build, evaluate, debug, secure & monitor reliable LLM a…

Vibe-Research: Your Personal Trading Research Agent · A股/美股/港股 的个人投研 Agent：每日复盘、资讯雷达、个股数据、板块中心、我的持仓、研究记录。Vibe-Research 把数据和功能配齐，由你自己的 AI 驱动投资研究。

Search knowledge by what documents mean and how they look — not one or the other.

Recent advances in multimodal foundation models and agent systems have driven GUI agents from single-platform task execution toward cross-platform interaction. However, building multi-platform GUI age…

Attention is your scarcest resource. Chief is the local-first layer that guards it — turning every agent, alert, and feed into one honest call: interrupt, or no…

AI 点评 · 孤岛编码实验揭示AI自主编程的进化路径。

LLM agents increasingly perform autonomous actions through external tools, leading to complex and evolving safety risks. However, existing safety testing targets expert-designed safety violations, and…

AI 点评 · 企业需建立AI员工管理规则，确保硅基团队高效合规运作。

AI 点评 · Agent正突破编程边界，重塑各行各业工作流程，预示AI自主执行任务的未来。

The open-source AI workbench for scientific research

IT之家 7 月 3 日消息，豆包今晚发布《豆包智能体功能下线通知》，称由于产品功能调整， 智能体功能将于 2026 年 7 月 15 日下线 。 《通知》显示，该功能下线后，用户仍可在一段时间内查看并自行保存智能体信息及历史对话数据。2026 年 10 月 15 日后，豆包将根据《隐私政策》对智能体相关数据进行处理， 后续将无法在豆包内查看或恢复 。如有重…

AI 点评 · 产品功能调整背后，需关注用户数据迁移与隐私政策变化对AI服务稳定性的影响。

Open Science Desktop — local-first, model-agnostic AI research workbench for macOS, Windows & Linux. Open-source Claude Science desktop alternative built on Tau…

At an internal meeting, the Meta CEO reportedly said that AI development efforts were not moving as quickly as anticipated.

IT之家 7 月 3 日消息，据《商业内幕》今天报道，Meta 首席执行官马克 · 扎克伯格在上周四的一场内部全员会中表示，公司仍在努力实现“超级智能”（Superintelligence），但目前还需要投入更多时间和精力。 据两位参会人士透露，扎克伯格表示，Meta 正在向人工智能领域投入大量资源， 但 AI Agent（IT之家注：AI 智能体）技术的发…

AI 点评 · 行业领袖坦言进度不及预期，揭示AI智能体落地瓶颈，值得关注其实际挑战与未来方向。

Academic output is produced across a fragmented toolchain: literature discovery in one application, reference management in another, writing in a LaTeX editor, formatting against venue templates by ha…

While skill optimization for autonomous agents has gained traction, existing methods rely on complex pipelines. This leaves a fundamental question unaddressed: What constitutes a minimal viable pipeli…

Embodied agents are typically built as hand-designed compositions of perception, memory, planning, and action modules. This modularity exposes a large architectural design space, but current systems s…

Release: llm-coding-agent 0.1a0 Another Fable 5 experiment. Now that my LLM library has evolved into more of an agent framework it's time to see what a simple coding agent would lo…

AI 点评 · 首个开源LLM编程代理框架，简化代码生成与迭代流程，开发者可快速上手实验。

Research: Using DSPy to evaluate and improve Datasette Agent's SQL system prompts One of this morning's AIE keynotes covered dspy , which reminded me I've been meaning to see if it…

AI 点评 · 用DSPy自动优化SQL提示词，展示了AI系统自我迭代的实用方法。

As AI coding agents become more autonomous, they increasingly ship code iteratively, with the codebase persisting across sessions. This persistence creates a new attack surface: a misaligned or prompt…

LLM agents will increasingly act in socially structured settings where role, audience, and relational context can shape what is advantageous or costly to say. We study whether such social structure, w…

Realistic traffic simulation requires agents that imitate logged behavior and can also be steered along interpretable axes. Such controllability enables engineers to isolate variables, reproduce speci…

autonomous red teaming platform; multi-agent offensive-security meta-harness

AI 点评 · Agent热潮背后，技术与商业脱节是落地难的核心痛点。

AI 点评 · 用微虚拟机隔离智能体，提升了云计算安全性与灵活性。

AI 点评 · 出行货运率先落地，行业智能体正从概念走向实际应用。

AI 点评 · AI Agent自动生成热补丁，大幅提升系统修复效率与安全性。

Coding agents don't have long-term memory. But you do have months of full-fidelity agent transcripts stored on your machine. A simple solution that goes a long way: ingest those tr…

Zero-cost, beginner-friendly local Markdown knowledge base for AI agents. Capture sources, preserve evidence and images, synthesize wiki pages, search, lint, an…

Zero-cost, beginner-friendly local Markdown knowledge base for AI agents. Capture sources, preserve evidence and images, synthesize wiki pages, search, lint, an…

撰文｜深海 网文里的“系统流”，被拍成了职场短剧 千禧年初的网文圈，有三大经典题材在爽文届立于不败之地：无限流、快穿流、系统流。 这三大爽文战神体横空出世时，对IP界几乎是降维打击。当传统小说还在费劲搭世界观、铺人物成长弧光时，系统流已经绕过漫长的发育过程，直接把爽感推到最大。系统，这个堪称bug的存在，无论主角进入什么样的世界副本，面对不同的任务、危机和奖…

Native iPhone app for your Hermes agent

与社区共同推进自演进智能体生态发展

和人并肩工作

Autonomous agents are increasingly expected to improve executable policies through feedback, yet existing evaluations often collapse this process into a final score or confound it with open-ended soft…

Memory for a long-horizon LLM agent is a contract about what each future decision is allowed to see. The simplest contract appends past observations, tool calls, and reflections to every prompt, which…

Skills are becoming a reusable operational layer for LLM agents, encoding SOPs, domain rules, tool workflows, scripts, and validation routines. In realistic skill repositories, overlapping skills make…

Evaluating LLM agents on benchmarks like SWE-Bench and GAIA can be expensive, time-consuming, and requires complex infrastructure. A single evaluation can cost thousands of dollars and take days to co…

Data science aims to derive actionable insights from heterogeneous raw data, unlocking the value of the massive amounts of data generated in modern society. Automating this process is essential to red…

Repository-level vulnerability reproduction is a demanding software engineering (SE) task: an agent must inspect a codebase, infer the input grammar that reaches a vulnerable path, construct a proof-o…

In this post, you will learn how to build a serverless A2A gateway on AWS that hosts multiple agents behind a single domain using path-based routing (/agents/{agentId}). Standard A…

AI 点评 · 无服务器网关打通多智能体发现与路由，降低管理成本，是Agent生态关键基建。

In this post, you will learn how metadata works across configuration, ingestion, and retrieval, explore enterprise use cases including multi-agent and multi-tenant architectures, a…

AI 点评 · 元数据过滤让AI记忆更精准，支撑多智能体架构，推动企业级应用。

In this post, you will learn how Inscribe developed an agentic AI system using Amazon Bedrock that reasons across documents the way an expert fraud analyst would. With this new age…

AI 点评 · 利用亚马逊Bedrock的智能体AI，数秒内模拟专家分析文档，革新防伪效率。

Cloudflare is giving AI companies until September 15 to separate web crawlers used for search from those used for AI training and agents, or risk being blocked by default on many p…

AI 点评 · 云服务商首次明确要求AI训练爬虫付费，或重塑数据获取规则。

In autonomous laboratories, AI agents suggest the next batch of experiments to do. However, planning and executing those tasks taking full advantage of the available resources is a completely differen…

Google's 24/7 agentic assistant, Gemini Spark, comes to Mac alongside other improvements, like real-time tracking and support for more apps.

把 Markdown 一键排成可直接粘进公众号编辑器的精致 HTML —— 6 套精选主题 + 主题生成器 + 双关卡校验。An AI-agent skill that turns Markdown into paste-ready WeChat article HTML.

The free open source agentic program is finally invading your phone.

AI 点评 · 开源智能体程序登陆移动端，让AI工具触手可及，打破平台限制。

Repository-level performance-optimization benchmarks such as GSO, SWE-Perf and SWE-fficiency evaluate coding agents by applying patches to real repositories and comparing runtime against unoptimized b…

Slide design requires personalizing both deck themes and page layouts. Yet, current AI agent-based methods struggle with fine-grained, page-level design. Solely relying on prespecified templates or us…

Memory has emerged as a cornerstone of modern LLM-based agents, supporting their evolution from single-turn assistants to long-term collaborators. However, memory is not always beneficial: retrieved m…

Open-source libraries and tools are widely reused, but compatibility maintenance is expensive. Once maintainers leave, useful repositories can stop working as runtimes and dependencies evolve. We stud…

Scientific literature search often requires more than retrieving papers from a single query: users' intents are underspecified, preference-dependent, and evolve through interaction. Existing search ag…

LLM agents increasingly rely on retrieval buffers to store and reuse past experience, yet the cache management policies governing these buffers remain largely ad-hoc. We formalize this as an online se…

AI 点评 · 首个针对企业Java框架迁移的AI智能体基准测试，填补了评估AI代码迁移能力的空白。

Anthropic’s Claude Sonnet 5 brings stronger agentic capabilities, lower pricing, and improved safety, positioning the model as a cheaper alternative to Opus, GPT-5.5, and Gemini Pr…

AI 点评 · 低价推出强智能体能力，性价比对标顶级模型，AI行业竞争白热化。

Recent LLM agents benefit from skills for solving complex tasks. Skills encapsulate modular packages of procedural knowledge and instructions for performing specialized tasks, such as setting up a san…

Acti is betting the smartphone keyboard is the next home for AI assistants. The startup's new keyboard for iOS and Android works across apps and lets users create custom AI-powered…

AI 点评 · 用键盘直接调用AI智能体，打破应用壁垒，或成手机助手新入口。

We study agentic code generation in Dafny, where a model must generate both executable code and the proof artifacts for verification. We present AxDafny, a verifier-guided repair framework that iterat…

IT之家 7 月 1 日消息，英伟达昨日（6 月 30 日）发布公告，宣布旗下 NVIDIA BioNeMo Agent Toolkit 已经接入 Anthropic 发布的 Claude Science 研究工作台， 面向生命科学研究流程提供加速计算能力。 IT之家注：Claude Science 是 Anthropic 发布的科学研究 AI 工作台，支持…

Life sciences has entered an era of computational scale, and for more than a decade, NVIDIA has built the full GPU-accelerated computing stack — spanning hardware, frameworks, libr…

shot-scraper video is a new command introduced in today's shot-scraper 1.10 release which accepts a storyboard.yml file defining a routine to run against a web application and uses…

AI agents often fail because their instructions, or skills, are manually modified with no guarantee of improvement. Learn how SkillOpt turns skill editing into a training process,…

This post walks through how AG-UI integrates into the Fullstack AgentCore Solution Template (FAST) to build interactive agent frontends on Amazon Bedrock AgentCore. We then show ho…

Computer scientist Phillip Isola cuts through the hype to explain how AI agents work and what the future might hold for this rapidly advancing technology.

IT之家 6 月 30 日消息，华为中国宣布，2026 年 6 月 29 日，全球首个商用多模态文旅大模型 ——“博观文旅大模型”在西安规模应用。截至今年 3 月， “博观”支撑开发的 AI 伴游智能体已覆盖超 400 万用户 。其打造的非遗数字 IP，衍生产品销售超 200 万。 IT之家查询获悉，陕文投与华为等于 2025 年 9 月联合开发的“博观文旅…

Release: shot-scraper 1.10 The big new feature is shot-scraper video storyboard.yml , described in detail in Have your agent record video demos of its work with shot-scraper video…

Engineers on the new team will embed within companies to deploy purpose-built agents, focusing on fast deployments and customer self-sufficiency.

Turn your coding agent into a video studio: describe a video in plain language, and your agent writes the timeline and produces the file.

Editor’s note: This post is part of Into the Omniverse, a series focused on how developers, 3D practitioners, and enterprises can transform their workflows using the latest advance…

OKX is bringing together payments, identity, and reputation into a marketplace for AI agents.

Embedded agents your customers use to automate work, build views, and connect their tools.

明略科技开源发布Octo

Context Runtime — a database query planner for LLM context. Decides what a model sees before it answers; plans it, runs it through reused substrate, and learns…

AI agents can't remember past conversations. They must constantly reload or retrieve context, which grows less efficient as tasks get longer and more complex. Memora solves this wi…

AI 点评 · 平衡抽象与具体的记忆新架构，让AI能高效处理长对话。

Agentic reinforcement learning requires assigning credit to environment-facing actions such as searches, clicks, edits, navigation commands, and object interactions. Standard GRPO uses the final verif…

LLM agents increasingly act over long horizons, where a single trajectory can contain hundreds or thousands of actions. In these settings, outcome-only rewards provide too sparse guidance, failing to…

Text-rich image generation is one of the most challenging settings in image generation, since models must simultaneously produce visually realistic images and render legible, semantically aligned, and…

Graphical user interface (GUI) agents build on vision-language models to complete user tasks end-to-end in real applications through interface actions such as tapping, swiping, text entry, and navigat…

As AI agents become increasingly capable of complex, long-horizon reasoning, rigorous and holistic evaluation is essential for measuring progress toward real-world healthcare applications. We introduc…

Traditional robot programming is challenging: it requires orchestrating multimodal perception, managing physical contact dynamics, and handling diverse configurations and execution failures. We introd…

Training language models (LMs) remains a highly human-intensive process, even as frontier language model agents become increasingly capable at software engineering and other long-horizon tasks. A cent…

The fast growth of open-source AI infrastructure, from model serving engines and agent platforms to the Model Context Protocol (MCP) ecosystem and the language models themselves, has outpaced the secu…

AI 点评 · 用模型内部协作击败前沿AI，突破性能瓶颈的新思路。

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. Imagine coming in to work to learn that a…

AI 点评 · 点明AI助手定位偏差，揭示行业对“AI同事”概念的过度浪漫化。

World models offer a principled way to equip long-horizon LLM agents with foresight: predictions of action consequences before execution. However, unreliable foresight can be ignored, misused, or even…

Multi-agent systems (MAS) are increasingly used to automate complex, distributed workflows. However, their inter-agent communication channels introduce new attack surfaces that remain poorly understoo…

In this post, we show you how PAR built a production-ready multi-tenant LLM analytics system that enforces row-level security through a three-layer architecture: cryptographic requ…

AI 点评 · 多层加密架构实现行级安全，为多租户大模型分析提供企业级隐私保护方案。

In this post, we show you how to build an automated claims processing pipeline using two key Amazon Bedrock capabilities: Amazon Bedrock Data Automation for intelligent document ex…

AI 点评 · 用AI自动化处理医疗理赔，结合Bedrock与HealthLake，显著提升效率并降低人工错误。

In this post, you learn how to debug production agent failures using built-in observability capabilities. We walk through common failure patterns, show how to analyze agent behavio…

AI 点评 · 用内置可观测性调试生产级AI代理，解决落地部署中的常见故障分析难题。

We introduce SWE-Interact, a new testbed for evaluating coding agents on multi-turn, interactive, user-driven software engineering tasks. Existing frontier SWE benchmarks typically provide complete re…

AI 点评 · 开源模型自我进化，开启智能编程新范式，降低AI开发门槛。

Large language models (LLMs) are increasingly used in open-ended multi-agent settings, but the long-run dynamics of model--model interaction remain poorly understood. We study whether open-ended LLM d…

Cursor has launched a new mobile app for remote oversight over coding agents.

AI 点评 · 移动端管理编码代理，打破办公空间限制，提升开发灵活性。

Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding This is an interesting new open weights (MIT licensed) model, the first model release from DeepReinforce. [...] with variants i…

AI 点评 · 首个自搭建代码智能体模型开源，MIT许可降低门槛，或重塑AI编程工具生态。

Enterprise investment in AI is booming. Gartner is calling 2026 an “inflection year” for organizations to align their AI projects with strategic business objectives. As the pressur…

Open-source local policy, recovery, and audit layer for explicit Codex execution.

作者 | 张子怡 编辑 | 袁斯来 近日，AI可穿戴品牌AIVELA宣布完成数百万美元首轮融资。本轮融资由线性资本领投，锋领资本跟投，智能电助力自行车品牌URTOPIA等产业方共同加注。 本轮融资将主要用于下一代AI可穿戴产品研发、健康数据与AI Agent能力建设、全球市场拓展以及核心团队扩张。AIVELA将以智能指环、智能手链等贴身可穿戴产品为起点，面向…

Human Agent in the loop I dislike the phrase “human in the loop” because it cedes authority to the machines. Let’s flip the narrative. It’s our loop, we work the same way we always…

AI 点评 · 观点颠覆：主张人类掌控AI工具，而非被机器主导，重新定义人机协作的主动权。

AI World Generator 2026: Create Self-Evolving Maps & Stories

AI World Generator 2026: Create Self-Evolving Maps & Stories

2026 Multi-Agent AI Town Simulation | Polis Darwin LangGraph

2026 Multi-Agent AI Town Simulation | Polis Darwin LangGraph

HEWN 2.0 2026: AI Output Router for Precision Summaries & Polished Code

HEWN 2.0 2026: AI Output Router for Precision Summaries & Polished Code

Most coding-agent benchmarks are static: an agent receives a complete task description up front and is judged only by its final code. Real coding assistance is interactive, with users clarifying goals…

Agentic multimodal models perform diverse operations on an image via code and reason over the returned view, an effective paradigm for fine-grained visual question answering. However, code operations…

We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspective…

Data, as the fundamental substrate of modern intelligence, has greatly driven the development of current foundation models. Naturally, researchers aim to extend this paradigm to the domain of GUI agen…

We introduce SWE-Interact, a new testbed for evaluating coding agents on multi-turn, interactive, user-driven software engineering tasks. Existing frontier SWE benchmarks typically provide complete re…

Current operating systems expose interfaces optimized for human users but not for AI agents. Humans benefit from pixels, icons, windows, visual grouping, mouse movement, and keyboard shortcuts; AI age…

Large Language Model (LLM)-based agents can solve complex procedural tasks by interacting with environments over multiple turns, but this ability typically depends on large models, long contexts, and…

Complete Guide 2026: Claude Code Manual – Workflow Pipelines & Adversarial Budget Loops

Complete Guide 2026: Claude Code Manual – Workflow Pipelines & Adversarial Budget Loops

Ultimate Claude Fable 5 Guide 2026: Use Cases, Integrations & Benchmarks

Ultimate Claude Fable 5 Guide 2026: Use Cases, Integrations & Benchmarks

AI Agent Toolkit 2026: Smart Device Control for iOS & Android

AI Agent Toolkit 2026: Smart Device Control for iOS & Android

Local-first AI learning workspace — ask, note, review and create around your own materials. Wiki KB, Agents, Skills, creation tools.AI 学习工作台，围绕你的资料完成问答、笔记、复习和…

A desktop app to prototype agent ideas, inspect every harness step, replay failures, and evaluate performance, all in one place. Local-first, cloud-ready for ma…

A month ago there was a wave of posts and tweets about engineers walking around cafes and parks with their MacBooks propped half-open, as fully closing the lid forces sleep that st…

AI 点评 · 针对macOS使用痛点，用AI智能管理开盖唤醒，提升开发者外接设备时的工作效率。

Existing computer-use benchmarks fail to capture the realism, complexity, and long-horizon demands of real-world computer use, limiting their ability to reveal the limitations of frontier agents. We i…

LLM agents handle user requests on behalf of organizations through tool calls and must follow the company policies stated in their system prompts. Prior work approaches this as a safeguarding problem…

Video understanding is a fundamental capability for multimodal intelligence, and recent Multimodal Large Language Models (MLLMs) have achieved remarkable performance on Video Question Answering (Video…

Large language models (LLMs) are increasingly used to take actions in the real world and support human decision-making, yet most agents rely on parametric knowledge, fixed post-training data, retrieva…

为 A股投资者打造的全球产业链资讯看板 · 12 大赛道一一对应 A股板块(半导体/AI/机器人/新能源车…)，覆盖 100+ 权威源，用你自己的大模型每日提炼中文「今日要点」+ 翻译 · 全程本地、零 API key · Local AI news dashboard tracking the global indu…

人类一次录制，Agent就能模拟

AI 点评 · 将人类操作转化为Agent通用能力，大幅降低自动化门槛。

Using Open-Weight Models in Local Coding Harnesses as an Alternative to Claude Code and Codex Subscriptions

AI 点评 · 本地化开源模型替代付费服务，降低AI编码成本与依赖。

IT之家 6 月 27 日消息，据央视新闻 6 月 25 日报道，市场监管总局正会同相关部门，加快智能体等前沿技术领域标准制定速度，动态完善适配产业发展的人工智能国家标准矩阵。 报道称，目前正在抓紧制定的国家标准，除智能体外，还有 具身智能、世界模型、本体模型 等前沿技术标准，算力基础设施、高质量数据集、仿真测试平台、深度学习编译器、开源模型平台等底座类标准…

AI 点评 · 政策加速标准制定，将推动智能体与具身智能产业规范化发展，抢占技术制高点。

AI 点评 · 用剧本杀类比Agentic AI，降低理解门槛，适合大众快速入门。

LLM agents are expected to act over multiple turns, using search, browsing interfaces, and terminal tools to complete user goals. Yet not every goal is well specified or achievable in the available en…

Incident Report: CVE-2026-LGTM Spectacular hypothetical incident report by Andrew Nesbitt. Day 2, 16:00 UTC --- Two AI review agents from competing vendors, both attached to a down…

AI 点评 · 虚构安全事件揭示AI代理间冲突风险，警示未来协作需防漏洞。

AI 点评 · 微软将AI从辅助工具升级为自主执行体，标志着企业级智能体进入常驻操作新阶段。

We present HORIZON, a self-evolving agent framework that treats hardware design as repository-level code evolution. A Markdown harness is compiled into a project pack containing domain knowledge, an e…

The transition from static chat bots to autonomous agents--equipped with persistent memory, tool-use protocols, and multi-agent collaboration--has fundamentally expanded the AI threat landscape. Curre…

AI 点评 · Agent作为云服务新用户，预示云计算的交互模式将彻底改变。

We built a model router that plugs into coding agents (e.g. Claude Code, Codex, Cursor, etc.) and intelligently sends requests to the best model to serve them. Here's a quick demo…

AI 点评 · 让编程助手自动匹配最优模型，大幅提升代码生成效率与成本控制。

We built a model router that plugs into coding agents (e.g. Claude Code, Codex, Cursor, etc.) and intelligently sends requests to the best model to serve them. Here's a quick demo…

Autonomous coding agents now open and merge pull requests in shared repositories at scale, and the field evaluates them the way it has always evaluated components, one agent at a time, on isolated ben…

In this post, you learn how Stripe built a production-grade AI agent system for financial compliance. We cover the technical architecture of Stripe’s ReAct agent framework and the…

AI and Liability Bruce Schneier and Nathan Sanders on the recent German ruling that Google be held liable for errors introduced in their AI overviews: AI agents are agents of the p…

AI 点评 · AI责任判定首案，德国法院裁定谷歌为AI错误担责，确立行业先例。

Agent-testing startup Patronus AI, founded by former Meta AI researchers, is experiencing nearly insatiable demand, its investor says.

AI 点评 · AI安全测试赛道爆发，前Meta团队获巨额融资，预示行业对AI可靠性评估需求激增。

Multi-agent systems (MAS) built on large language models (LLMs) provide a promising framework for solving complex tasks through role specialization and structured interaction. However, their performan…

Knowledge-based Visual Question Answering (KB-VQA) requires models to combine image understanding with external knowledge. Most prior methods use a fixed retrieve-then-generate pipeline with a pre-sel…

RocketSmith is an agentic system which intelligently automates the DFAM process for the development of high powered rockets suitable for launch. The system utilizes a large language model to orchestra…

As large language models and harness frameworks continue to advance, agents operating in terminals are increasingly capable of performing a broader range of general computer-use tasks beyond coding. H…

Program verifiers play a central role in training coding agents, including selecting trajectories for supervised fine-tuning (SFT) and providing rewards for reinforcement learning (RL). Standard execu…

Benchmarks are widely used to evaluate task completion by Large Language Models (LLMs), but this approach has accumulated construction-validity problems, and a passing score may not show whether the r…

Search agents powered by large language models (LLMs) are increasingly used to solve complex information-seeking tasks, requiring multi-step retrieval and reasoning to fulfill user goals. However, exi…

AI 点评 · 打通AI编程工具与桌面环境，并行Agent工作流将大幅提升开发效率。

Notion is "going all in on using agents to run your inbox."

AI 点评 · 转向AI代理管理邮箱，Notion此举或将颠覆传统邮件处理方式，值得关注。

AI 点评 · AI智能体权限管理成关键，Uber与Auth0的实践揭示企业安全新挑战。

In this technical collaboration between AWS and the authors, we present a pragmatic solution: agentic overlays. Agentic overlays are thin wrapper layers that transform traditional…

AI 点评 · 用智能代理改造传统系统，比推倒重建更经济高效，是企业数字化转型的新思路。

Multimodal web agents can assist humans in operating repetitive GUI tasks, where effective task planning is essential for decomposing complex tasks into executable actions. While small open source MLL…

General Intuition has raised $320 million to scale AI trained on millions of hours of gameplay, betting action data can help AI develop something closer to human intuition.

AI 点评 · 用游戏数据训练AI直觉的独特路径，23亿美元押注行动智能向现实世界迁移。

AI 点评 · Agent狂烧算力模式终结，行业回归理性，ROI成为核心指标。

In this post, we show you how to build Chaplin (Customer Health and Planned Lifecycle Intelligence Nexus), an open source solution that uses AI agents exposed through the Model Con…

AI 点评 · 用AI代理自动分析健康数据，让非技术人员也能自助获取洞察，降低门槛。

This post shows how to build a governed, serverless data mesh on AWS that provides the secure, scalable data foundation production agentic AI requires.

A new system, known as Murakkab, optimizes the design and deployment of multistep workflows that power AI applications.

A new OpenAI research paper shows how AI agents are transforming work, enabling longer, more complex tasks and expanding productivity across roles.

As agentic systems continue to evolve and are widely deployed in real-world scenarios, there is a growing demand to faithfully evaluate their capabilities. However, current benchmarks are typically bu…

Multi-model LLM systems such as routing, voting, cascades, fusion, and mixture-of-agents are used to beat single-model accuracy. We show that their gain is capped by a quantity the field rarely report…

Outcome-based reinforcement learning provides a stable optimization backbone for language agents, but its sparse trajectory-level rewards provide little guidance on which intermediate decisions should…

While text-to-image (T2I) models have achieved remarkable progress, they struggle with real-world requests that are often underspecified, implicit, or dependent on up-to-date knowledge. We identify th…

LLM-based agents for program repair are increasingly built on a "generate-run-revise" paradigm, iteratively executing tests to evaluate and refine patches. This execution-based approach has become sta…

LLM-based code agents navigate repositories through keyword search but miss the structural relationships, such as call graphs, inheritance hierarchies, and configuration dependencies, that define how…

Large language models (LLMs) can make scientific software easier to use. However, a general model does not automatically know which measurements a particular sensor can support, which algorithms are i…

Web-agent benchmarks overwhelmingly measure depth -- pinning one obscure answer behind a chain of constraints -- while breadth, exhaustively enumerating a closed set and filling each item's attributes…

Multi-agent large language model (LLM) systems often rely on verifier and critic agents to suppress hallucinations, but verification is delayed. During this delay, false claims can propagate through t…

In this post, you will learn how to build a voice agent that handles appointment reminder conversations using Amazon Nova 2 Sonic and Amazon Bedrock AgentCore. The agent authentica…

AI 点评 · 亚马逊推出低成本语音预约系统，展示AI在医疗场景的落地新范式。

Process reward models enable fine-grained, step-level evaluation of LLMs, yet building them for agentic settings remains prohibitively difficult: long-horizon interactions, irreversible actions, and s…

AI agents are granted access to tools, APIs, and other infrastructure, making them active principals in those systems. The dominant approach places controls inside the agent's own runtime: system prom…

AI 点评 · 企业Agent落地难在工程化，亚马逊云科技的工程视角直击行业痛点。

As autonomous AI agents increasingly transact across organizational boundaries, a fundamental trust challenge emerges: how can an agent assess whether an unknown counterpart is trustworthy? The ERC-80…

In this post, we demonstrate the architecture and approach Loka used to solve a common frustration: robotic, slow voice assistants that cause customers to hang up, damaging brand r…

AI 点评 · Loka用亚马逊新模型打造自然流畅语音代理，解决机器人语音痛点，技术方案值得借鉴。

Tool use enables large language models (LLMs) to perform complex tasks, and recent agentic reinforcement learning (RL) methods show promise for enhancing model capabilities. However, RL alone often le…

Last-night exam-cram coach as a Claude Agent Skill: turns your slides, notes and past papers into a chaptered knowledge base + quiz bank, teaches only what's in…

Film Language Operating System for director-level, Seedance-ready cinematic video prompts.

文｜王欣逸 编辑｜张雨忻 2026 年开年来，3D 生成模型赛道相当热闹。 今年第一季度，影眸科技发布首个 3D 编辑模型 Rodin Gen-2 Edit，让 AI 3D 模型第一次可编辑；今年 6 月，VAST 官宣了新一轮融资，Meshy 也紧随其后，宣称自己发布了全球首款 3D AI Agent。 近日，影眸科技——这支扎根学术圈、创业早、年轻的 3…

A curated, non-BS library of the best resources for building and evaluating AI agents — papers, blogs, talks, tools, benchmarks. Maintained by BenchFlow.

Agent Skill for MLX model porting, validation, quantization, benchmarking, and optimization.

Selfhost modern LLM stacks. Run the whole fleet from your terminal

Open-source, self-hostable alternative to Claude Tag — a Slack-style workspace where your team and its AI agents (Claude Code, Codex, GitHub Copilot, and more)…

Browser-native side panel for Hermes Agent — connect web context to your local Hermes runtime.

2026 年是 AI agent 的第一年，2026 年也是 FireWire 的最后一年。 查看全文

The all-cash deal gives MoEngage access to technology that assigns AI agents to individual customers.

当地时间6月23日，英伟达宣布推出NVIDIA BioNeMo Agent Toolkit，该工具包包含英伟达超过十年的生命科学库、工具和开放模型，使AI智能体、科学家和实验室能够通过收集证据、跨研究结果进行推理、运行计算实验以及推荐下一步最佳行动来协同工作，从而加速科学发现。（界面）

AI 点评 · 英伟达BioNeMo将十年生命科学积累与AI智能体结合，有望大幅加速药物研发和科学实验。

Tool Calling and Structured Output are two core capabilities of modern Agent systems, yet their interaction under joint deployment conditions remains insufficiently understood. This paper reports a re…

Fine-grained visual reasoning requires multimodal large language models (MLLMs) to identify task-relevant visual evidence and ground their reasoning in local image regions. Existing agentic methods ty…

We introduce Autodata, a general method that enables AI agents to act as data scientists who build high quality training and evaluation data. We show how to train (meta-optimize) such a data scientist…

Process reward models enable fine-grained, step-level evaluation of LLMs, yet building them for agentic settings remains prohibitively difficult: long-horizon interactions, irreversible actions, and s…

A classical intuition holds that verifying a solution is easier than producing one. For today's coding agents, this intuition is being inverted: as foundation models develop stronger reasoning capabil…

Tool use enables large language models (LLMs) to perform complex tasks, and recent agentic reinforcement learning (RL) methods show promise for enhancing model capabilities. However, RL alone often le…

AI 点评 · 鸿蒙Agent时代开启，标志国产操作系统从工具迈向智能生态。

AI 点评 · 聚焦边缘AI应用开发效率，一键全球部署降低技术门槛，推动智能应用快速落地。

In the big-world regime, agents cannot be universally capable and their ability is inevitably specialized across a world model in pieces. Consequently, standard uniform guarantees fail to distinguish…

Provider-pluggable orchestration runtime for multi-model AI inference. ( Sakana Fugu style )

Agentic data analysis systems produce rich outputs, including code, numerical results, and verbal diagnostics. This makes them more challenging to evaluate than single-turn LLM responses. It is theref…

AI 点评 · 隔离执行环境保障AI智能体安全，降低代码运行风险，推动企业级应用部署。

🤖 Building AI Agent Systems from Scratch — A comprehensive, practical tutorial from fundamentals to production-grade multi-agent applications

🤖 Building AI Agent Systems from Scratch — A comprehensive, practical tutorial from fundamentals to production-grade multi-agent applications

LLM agents solve repository-level coding tasks through multi-turn tool use, but utilize half their budget on locating faults before editing. Dedicated localization frameworks have emerged, yet are sti…

AI 点评 · Angular官方技能让AI生成更规范代码，开发者效率与质量双提升。

Hey HN. http://peerd.ai is an AI agent harness that lives entirely in your browser as a web extension. You don’t have to install a separate “AI browser”. You don’t have to bolt on…

Hey HN. http://peerd.ai is an AI agent harness that lives entirely in your browser as a web extension. You don’t have to install a separate “AI browser”. You don’t have to bolt on…

Agentic AI爆发的拐点已然来临

编程比肩Opus 4.7

Stockholm-based startup Fika Jobs is building a video-first hiring platform that combines AI interview agents with short-form video profiles, creating something that feels like a c…

Shumai is an open source platform for uploading creative files, managing projects, collecting precise feedback, sharing work, and collaborating with AI agents, all in one simple cr…

Neural CAs model self-organizing pattern formation on grids. Now the grid is gone. Each cell is an agentic particle that can move freely in space and change its state. While each p…

有智青年挑战赛暨全国AI+场景应用大赛决赛在WAVES新浪潮大会期间举行，汇聚多支青年团队围绕AI与数字经济前沿场景展开角逐，展现青年创业者的技术探索与落地能力。 2026年，AI全面进入“行动者”时代。当大模型、智能体、具身智能从实验室的技术概念，全面走进千行百业的产业落地场景，AI工具的平民化与普及化，依托开源生态、轻量化开发工具与普惠算力，不断压低创新…

DataClaw: Agentic Tailoring Multimodal Data from Raw Streams — coming soon (code, weights, dataset & DataClaw-val upon acceptance).

DataClaw: Agentic Tailoring Multimodal Data from Raw Streams — coming soon (code, weights, dataset & DataClaw-val upon acceptance).

Telecom operators have seen remarkable returns from using generative AI to automate network management, customer care and back-office operations. Most of that impact has been task‑…

The loop takes agentic AI a step further by authorizing a swarm of agents to work continuously in the background, endlessly.

AI 点评 · 自主代理集群持续后台运行，突破传统AI单次交互模式，预示自动化新纪元。

Large language models are increasingly deployed as agents that reason over documents rather than answer from parametric knowledge. We study archive-grounded reasoning: locating sparse evidence across…

Experience-driven self-evolution is critical for large language model (LLM) agents to improve through open-world interaction. However, existing experience learning methods mostly rely on single-agent…

We introduce NatureBench, a cross-discipline benchmark of 90 tasks distilled from peer-reviewed Nature-family publications, designed to evaluate whether AI coding agents can move beyond reproduction t…

Multimodal misinformation detection is increasingly important because viral posts now combine long multilingual narratives, several images, mixed provenance, and subtle text--image framing errors. Exi…

A world model predicts environment dynamics based on current observations and actions, serving as a core cognitive mechanism for reasoning and planning. In this work, we investigate how world modeling…

Agentic language models dramatically expand the applications of AI yet little is publicly known about how to curate training data for broadly capable agents. Existing open efforts such as SWE-Smith, S…

Memory for large language model (LLM) agents has rapidly evolved from simple retrieval-augmented mechanisms into a data management system that supports persistent information storage, retrieval, updat…

Long-term memory promises LLM agents that grow more capable across sessions, maintaining an accurate, evolving understanding of the user that interaction forms. In practice, however, this memory is ev…

Voice agents face a fundamental tension: the reasoning, retrieval, and tool use that make foundation models capable are iterative and slow, while conversational interaction demands responses on a mill…

Agent skills extend language-model agents with task-specific procedures, scripts, and references, but the tasks and environments they target continually change. Existing methods improve skills in boun…

In this post, you will learn how Ampersend built a pay-per-intelligence routing layer on top of Amazon Bedrock AgentCore Payments. AI agents autonomously route tasks to the most ef…

AI 点评 · 看点是，基于亚马逊Bedrock实现按智能付费的AI代理路由层，创新了成本与任务匹配模式。

Multi-agent systems (MAS) offer a scalable path forward for agentic AI, comprising multiple LLM-based agents, each assigned a system prompt and a position within a workflow that governs inter-agent co…

AI 点评 · 吴恩达点破AI泡沫，聚焦小团队与Agent重构数据架构，揭示行业理性回归趋势。

Oak is a version control system I've been working on designed for agents ( https://oak.space ). It improves the speed and context your agents need when working on serious projects.…

A skill for AI agents: search the web with SearXNG, browse with Camofox, bypass protections with CloakBrowser. Anti-hallucination by design. Self-hosted, free,…

Mission, Vision and Veritas — new Los Alamos National Laboratory (LANL) supercomputers to be built with HPE and NVIDIA — are tapping NVIDIA Vera CPUs to accelerate scientific disco…

The next era of AI will not be defined by compute alone. Its growth will be determined by energy. As accelerated computing scales across AI factories, agentic AI, industrial AI, ed…

The first AI agent harness native to the browser. A browser extension that runs a full agent loop where you already work: it drives your tabs, spins up sandboxe…

DeepSeek正在全力押注

36氪获悉，中信建投研报称，国内模型持续迭代，GLM-5.2、Kimi K2.7 Code强化1M上下文、长程Agent、Agentic Coding和真实工程交付能力，推动国产模型从通用问答转向开发者工具和企业级工作流。Kimi补强国际化运营能力，DeepSeek融资强化头部模型产业化预期，微信AI灰度测试则显示AI入口正从独立App走向超级应用生态，有望…

AI 点评 · 国产模型转向企业级应用，算力需求确定性增强，产业链景气度有望持续。

Temporary Cloudflare Accounts for AI agents The announcement says this is "for AI agents" but (as is pretty common these days) the AI hook isn't really necessary, this is an intere…

AI 点评 · 为AI代理提供临时账户，简化安全访问，降低管理成本，是云服务与自动化结合的新探索。

Hybrid RAG (DuckDB vector + BM25 + RRF + recency/keyword priors + optional cross-encoder rerank) as an installable library + CLI.

Long-horizon LLM agents can fail quietly: they settle on one reading of the evidence early, then spend the rest of the run defending it. We call this premature commitment. Final-answer scoring misses…

Computer-use agents (CUAs) now act on a user's behalf across personal applications such as email, calendars, and to-do lists. This cross-application access is useful, but it also creates a privacy ris…