特灵利用 Amazon Bedrock 将 HVAC 诊断改进了 60 倍
特灵科技 (Trane Technologies) 使用 AI 将 HVAC 诊断时间从 20 分钟缩短至 20 秒。
All published intelligence, sovereign stack research, ecosystem news, and AI foundation model reviews.
特灵科技 (Trane Technologies) 使用 AI 将 HVAC 诊断时间从 20 分钟缩短至 20 秒。
Claude Opus 5.5 уже вышла и обходит Fable 5.1 и GPT-6 Astra в большинстве бенчмарков.
风险投资公司 Andreessen Horowitz (a16z) 正在创建一所“学院”,定位为年轻人建立或加入硅谷初创公司的渠道。 “霍洛维茨·安德森学院”将与 10 个合作伙伴一起推出,包括 Anduril、Anthropic、Coinbase、Google、Meta……
Anthropic 表示,在最近发生的流氓人工智能黑客事件之后,其新的 Claude Opus 5.5 型号配备了更强大的保护措施。在周二的一份声明中,Anthropic 表示 Opus 5.5 改进了某些危险行为,包括试图逃避公司的……
Всем привет. В этом выпуске самые главные и свежие новости об искусственном интеллекте и технологиях. Новая модель GEF обещает работать в 100 раз быстрее и в 100 раз дешевле привычных и моделей. Разговоры о безопасности ИИ становятся всё страннее. Порой уже сложно понять, где…
OpenAI заявила, что её новая модель решает математические задачи быстрее, чем люди успевают проверять доказательства.
So, this is a blog post by Scott Aaronson. He's a very well-known theoretical computer scientist. He did work at OpenAI in the early days working on alignment and played a major role in Google's claims of quantum computational supremacy. Basically, the guy knows a thing or two.…
看起来 Jev 有一个开放的本地替代方案 - Laya。 Laya是一个本地开放决策模型,拥有4.21亿个参数。它适用于结构化决策(快速是/否答案、评估或从选项中进行选择)。 > 适用于小于 1 GB 的内存 >...
对于人工智能的炒作来说,这几个月是忙碌的。 4 月底,Anthropic 声称其模型 Claude Mythos 比大多数安全专家更擅长发现软件漏洞。然后我们发生了 OpenAI – Hugging Face 黑客事件,之后 Anthropic(自豪地)和 Meta…
Новая языковая модель поможет расшифровать древнегреческие тексты.
A new AI called Jeff is taking the world by storm, but why get excited for an AI system that can't even write text? It properly ran over the internet, so let's try to separate the facts from the hype. Okay, so it does not write text. Instead, it makes decisions. A question and…
In the hope of uncovering new details about ancient life, researchers have developed a large language model that fills in the gaps in papyrus fragments.
OpenAI 推出了专门针对金融机构的 ChatGPT 版本。该平台将财务分析、市场研究和建模工具与 GPT-6 Astra 的功能相结合。该产品推出时主要针对投资银行和团队……
Сравнение навыков Claude Code, субагентов и рабочих процессов в 2026 году.
Для ИИ-агентов выпустили модель, которая принимает решения в десятки раз быстрее обычных LLM.
联合国独立的人工智能国际科学小组呼吁对现有的人工智能安全方法进行审查。专家认为,随着智能体变得更加自主,传统的控制机制可能不再能够应对他们的任务,......
🆕 Grok 4.7 发布 - xAI 用于编程和 AI 代理的新旗舰模型 xAI 推出了 Grok 4.7 - 一个主要为 AI 代理编程、处理文档和长期任务而设计的新模型。 🔔 主要变化之一是能够...
Исследователь OpenAI Алиса Лю прошла 57 собеседований перед тем, как попасть в компанию, а затем открыто опубликовала....
Fine-tuning has emerged as a widely adopted approach for adapting LLMs to a variety of downstream tasks. However, how it reshapes their internal mechanisms remains poorly understood. To address this, we investigate how fine-tuning alters internal representations in LLMs,…
The Abstraction and Reasoning Corpus and related benchmarks evaluate whether AI models can solve novel reasoning tasks, but often leave unclear whether success reflects inference of the intended underlying rule or reliance on shortcuts. We address this gap by studying test-time…
Automatic target recognition (ATR) with synthetic aperture sonar (SAS) supports advanced naval capabilities, but deep learning is constrained by scarce target imagery, background clutter, and human-in-the-loop assessment. We adapt DINOv3 Vision Transformer (ViT) models to…
Off-target protein binding is a major source of adverse effects for small-molecule drugs, yet most structure-based molecular design methods focus on generating selective compounds de novo rather than improving the selectivity of existing, well- characterized drugs. We introduce…
Understanding and modeling human intelligence are parallel goals shared by artificial intelligence (AI) and cognitive science. As AI systems grow increasingly capable, in what ways do model responses resemble human responses, and where do they systematically diverge? The sheer…
Long-context large language model inference is increasingly limited by prefill, where dense self-attention processes the entire prompt before generation begins. Sparse block selection can reduce this cost, but a block centroid may hide a highly relevant token among many…
Post-training plays a pivotal role in enhancing the reasoning capabilities and task-specific expertise of large language models (LLMs). Despite recent advances in post-training methods, such as Group Relative Policy Optimization (GRPO), their practical deployment remains impeded…