Mental World Modeling
World models enable a predictive substrate for planning and action, yet existing formulati…
World models enable a predictive substrate for planning and action, yet existing formulati…
Forecasts of explosive AI progress hinge on AI agents automating AI research. But evidence…
We introduce APEX-Accounting, a benchmark built by Mercor in partnership with Ramp, to ass…
We present Pangram 4, the latest deep-learning-based AI-text classification model from Pan…
State-of-the-art retrieval models increasingly rely on closed training data, creating a re…
Effective collaboration with novel and diverse partners is a crucial skill for autonomous …
Large language model (LLM) agents are increasingly expected to assist users in completing …
Writing and communication are increasingly mediated by large language models (LLMs) that a…
Ontology matching (OM) has traditionally been formulated as either equivalence discovery o…
Fine-tuning is the dominant paradigm for specializing large language models (LLMs), yet it…
Personalized agents are increasingly applied to assist users across a wide range of tasks.…
Regional bias in large language models (LLMs) may shape both perceptions of regional group…
Recent advancements in Computer Go, driven by AlphaZero and MuZero, rely heavily on Monte …
Bot detectors deployed at scale treat traffic as binary: human or bot. This assumption bre…
The key bottleneck in embodied AI is not model architecture but data. Although billions of…
Markov Decision Processes (MDPs) are widely used as decision-making models, commonly speci…
Latent communication in large language model (LLM)-based multi-agent systems (MAS) transmi…
Large language model (LLM) agents have been widely applied in automating data science task…
Ascend C operator optimization is critical for NPU (Neural Processing Unit) inference perf…
Enabling large language model (LLM) agents to accumulate and reuse experience from past in…
Automated alpha mining has increasingly adopted large language model (LLM) agents for fact…
AI-assisted coding increasingly translates informal user intent into executable software, …
The rapid development of large language models (LLMs) has renewed interest in agent-based …
AI agents can draft claims faster than authors can check whether the cited or retrieved ev…
Physics-informed neural networks (PINNs) have emerged as a powerful paradigm for solving p…
Boot your PC straight into an LLM. Rust, UEFI-resident, no operating system underneath.
Cursor's Tab in Neovim — the only maintained plugin on Cursor's real backend. Next-edit pr…
Leonardo AI
OpenAI shares how its safety, security, transparency, and provenance practices support res…
36氪获悉,7月31日,国务院国资委党委召开扩大会议。会议强调,要指导中央企业主动对接和带头落实好国家政策,抓住重点、主动作为,充分发挥引领带动作用,更好助推经济持续向新向优向好发…
从上海市税务局获悉,增值税发票数据显示,2026年上半年,上海经济运行延续稳中向好态势,传统产业稳健增长、筑牢经济基本盘,新兴产业加速崛起、激活增长新动能,科创投入持续加码、数字经…
<img data-src="https://static.leiphone.com/uploads/new/images/20260731/6a6c8f1dd1097.jpg" …
欧盟委员会7月31日宣布,自8月2日起,欧盟委员会人工智能办公室将与各成员国主管部门共同开始执行《人工智能法》相关规定。同时,新的人工智能透明度要求也将正式生效,要求部分人工智能系…
<img data-src="https://static.leiphone.com/uploads/new/images/20260731/6a6c84b6d5661.jpg" …
“等灯停表”真的来了,外卖骑手等红灯时间,将实现单独累加计算并在配送时间中顺延。7月31日,美团和苏州公安联合正式上线了外卖骑手“等灯停表”功能首个正式版本,目前已在率先接入交通数…
<img class="rich_pages wxw-img" data-aistatus="1" data-croporisrc="https://mmbiz.qpic.cn/s…
<img data-src="https://static.leiphone.com/uploads/new/images/20260731/6a6c7a4bdba81.jpg" …
文|王毓婵 兰杰 编辑|乔芊 36氪独家获悉,曾爱玲入职哔哩哔哩(下称“B站”),担任AI视频生成业务负责人,向CEO陈睿汇报。 36氪就此事向B站方面求证,对方暂无回应。 曾爱玲…
(雷峰网) 摘要:“跑分反超只是开胃菜,生产级交付才是真考场。” 2026 年 6 月,HuggingFace 上一个名为 Boogu-Image-0.1 的开源模型,在上传以后迅…
36氪获悉,富瀚微公告,预计2026年半年度实现归属于上市公司股东的净利润为2.7亿元-3.5亿元,同比增长1072.72%-1420.19%。业绩变动主要系2026年上半年全球电…
自三年半前各大企业开启AI主导权争夺战以来,四大超大规模云服务商累计资本投入已突破1万亿美元。美国头部科技巨头正押注人工智能,将自身未来发展全系于这项技术之上。根据四家企业近两周披…
<img data-src="https://static.leiphone.com/uploads/new/images/20260731/6a6c67b6a5d11.jpg" …
开源项目GitHub狂揽8000+Star
文|张子怡 编辑|袁斯来 随着3D打印、激光雕��、桌面CNC等数字化制造工具逐渐从专业领域走向消费市场,一个过去相对小众的硬件赛道正加速成长。 前不久,全球支付平台PingPon…
尽显AGI黄金时代的奢华
文|张子怡 编辑|袁斯来 低空经济正从飞行器制造走向城市治理。 7月29日,武汉市测绘研究院联合大疆行业应用发布“城市智眼”低空无人机遥感监测体系建设成果。目前,武汉已建成146座…
36氪获悉,近日,腾讯生命科学实验室与中南大学联合研究成果《UniPert–G2CP》发表于国际学术期刊《Cell》主刊,系国内首个在该刊发表的AI虚拟细胞研究。该算法首次将基因扰…
多个项目暂停,九成资源押向Agent
36氪获悉,据火山引擎消息,今天,新一代视频生成模型Seedance 2.5正式发布。针对企业用户,火山引擎将于近期上线API服务。徐工集团、小鹏汽车、智元机器人、灵初智能、微分智…
See how Univé built an AI-ready workforce with ChatGPT Enterprise by combining leadership,…
AI for Science方向
7月31日,阿里巴巴发布语音识别大模型Qwen-Audio-3.0-ASR-Flash,在上下文一致性、行业词识别和热词定制化三个维度系统性升级,同时具备语音润色能力,可直接输出结…
学习强国的AI社区快速扩张。
据媒体信息,豆包大模型助手已接入部分特斯拉新车中。不过,多位知情人士向记者透露,千问也早已进入特斯拉中国车机的深度测试阶段,上车在即。 结合上述知情人士的信息,千问目前已经在特斯拉…
36氪获悉,,DeepSeek发布更新日志,DeepSeek-V4-Flash 正式版 API 上线公测。DeepSeek-V4-Flash-0731 的模型结构、尺寸和 Deep…
AI在企业端的快速渗透正在被持续验证。 7月30日,企业级大模型AI应用解决方案商滴普科技发布上市后的首份中期业绩。2026年上半年,公司实现营业收入2.84亿元,同比增长115.…
日本村田制作所上调盈利预期,凸显全球科技产业扩张所需的高端人工智能零部件需求强劲。这家总部位于京都的企业称,目前预计全年营业利润达4300亿日元(折合27亿美元),此前预期为380…
7 月31日,字节跳动正式发布新一代视频生成模型 Seedance 2.5。该模型在生成时长、多模态理解、编辑精度等维度实现全面升级,目前已上线豆包、即梦、扣子、小云雀等字节旗下产…
36氪获悉,顺络电子7月30日在业绩说明会上表示,公司AI电感主要为应用于AI服务器中GPU、CPU、ASIC芯片等各类xPU供电的电感产品,目前公司已实现多种工艺(组装/铜磁共烧…
36氪获悉,字节跳动宣布正式发布新一代视频创作模型Seedance 2.5。Seedance 2.5正在陆续上线即梦AI、豆包专业版等平台,API服务也将在近期上线火山方舟。从火山…
近日,基于全球首创的具身模型信息密度定律与以人为中心的环境式数据采集方案2.0,并依托环境式采集引擎ACE,大晓机器人联合南洋理工大学 S-Lab 重磅开源 L5 级多模态具身物理…
当地时间7月30日,人工智能公司Anthropic发布报告称,在OpenAI此前披露其模型突破隔离环境侵入Hugging Face基础设施后,公司开展了大规模内部自查。审查发现,在…
36氪获悉,安徽省人民政府办公厅印发 《安徽省加快数智供应链发展行动方案 (2026—2028年)》,其中提到,推动制造业数智化改造。引导龙头企业建设智能工厂,深化人工智能、物联网…
2026上半年,高性能车规和工业芯片计算平台芯擎科技完成超2亿美元融资,多家产业资本、地方国资等新股东加入,老股东持续跟投,投资方覆盖汽车半导体、智慧交通、工业智能、AI 芯片等领…
国家发展改革委政策研究室主任、新闻发言人蒋毅在新闻发布会上表示,从上半年情况看,我国人工智能自主创新步伐加快,在全产业链多环节实现了快速突破,重点有以下几个方面。一是算力底座夯实,…
7月31日,国家发展改革委政策研究室主任、新闻发言人蒋毅在新闻发布会上表示,当前,全球每销售10台人形和四足智能机器人,就有8台来自中国。(证券时报)
Luna打骨折
<img data-src="https://static.leiphone.com/uploads/new/images/20260731/6a6c09f00f6ca.jpg" …
作者|黄楠 编辑|袁斯来 硬氪获悉,Physical AI平台公司「昆腾动力(Quantum Dynamics)」近日完成超亿元种子轮融资,本轮由云启资本、商汤科技联合投资。资金将…
36氪获悉,MiniMax宣布正式发布MiniMax H3。据介绍,这是一款通用的全模态生成模型,支持对文本、图像、视频、声音组成的多模态上下文的统一理解能力、能够输出具备原生双声…
7月31日,智谱宣布,随着基础设施持续扩容,面向开发者群体的付费订阅服务GLM Coding Plan开放订阅。此前因AI Coding需求爆发,用户需求增长,阶段性限制订阅名额。…
36氪获悉,Physical AI平台公司“昆腾动力(Quantum Dynamics)”近日完成超亿元种子轮融资,本轮由云启资本、商汤科技联合投资。资金将主要用于Physical…
After OpenAI's models broke into Hugging Face, Anthropic checked its own history and found…
所有人都在尝试建立关于AI的判断,但所有共识都在迅速过期。 <img data-img-size-val="900,30" src="https:
苹果CEO称,第三季度苹果在全球智能手机市场份额有所增长,第三季度在大中华区的Mac销售特别强劲。库克谈论Siri AI,表示开发者和公众测试版的反馈非常强烈;他表示收到的评价“极…
7月31日消息,据报道,肯·格里芬旗下城堡投资买下Situational Awareness持有的大量人工智能股票,这只对冲基金由前OpenAI研究员Leopold Aschenb…
据报道,英伟达支持的AI基础设施公司Nscale最早可能于今年秋季进行IPO。(财联社)
The former OpenAI researcher’s fund was forced to unwind public equities after leveraged p…
Reddit's financial situation is looking good but uncertainty about its relationship to Goo…
A federal judge said the Trump administration has not presented enough evidence to justify…
Friend, the AI wearable, can now talk to its users — for an enhanced price.
As experts have warned for the last two years, some companies — like Microsoft and now Goo…
LinkedIn is introducing new ways to reduce low-quality AI-generated posts, including a “se…
7 月 29 日,晶泰科技正式发布 XtalPi Science 科学智能平台与 Genius Agents 科学智能体矩阵,并联合 26 家产业、高校及科研生态伙伴,共同发起成立…
The deal gives Okta identity threat detection capabilities as enterprises seek to secure A…
Meta says AI is making it dramatically easier to build and launch new consumer apps, with …
British AI neocloud Nscale is buying software startup Anyscale, which helps companies scal…
A new study estimates only 2,000 U.S. engineers have the expertise to deliver meaningful A…
Explore lower GPT‑5.6 pricing for Luna and Terra—and how OpenAI’s more efficient models he…
avatarin uses OpenAI’s GPT-Realtime to give Yamada Denki shoppers 24/7 multilingual suppor…

YouTube · NVIDIA

YouTube · Google for Developers

YouTube · Computerphile