实测Claude Code，你会重新认识AI的编程能力边界
Article URL: https://davidpreichert.substack.com/p/if-you-havent-recently-used-claude Comments URL: https://news.ycombinator.com/item?id=49157496 Poin…
大模型变强后，哪些Agent“必备技能”正在被开发者主动淘汰？来看看一线实战清理清单。
As LLMs get stronger, I'm pruning my agent skill library. I've removed multi-step workflows (like superpowers) and web wrappers (like Agent Browse) th…
Claude Code之父揭秘：模型能力已远超产品边界，是时候解开Harness的缰绳了。
Claude code之父：大模型是有机生物，做好AI产品疏胜于堵
润建股份详解如何从Token生产到价值闭环，降低企业AI应用成本，推动落地。
7月17日，2026世界人工智能大会在上海开幕。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��在大会首日同步开启现场对话。润建股份董事、副总裁丁永在WAIC现场接受36氪「氪话未来」特邀专访，围绕Token生产、应用落地、价值闭环及出海布局等话题，分享了润建股份打造Tok…
国产模型实力追赶，金山办公已将内部开发全面切换，行业趋势迎来转折点。
IT之家 7 月 15 日消息，在今日举行的金山办公 2026 AI 生产力大会上，金山办公首席执行官章庆元就当前人工智能大模型行业的发展趋势发表了自己的看法。他指出，大模型的稀缺性正在降低，各模型之间的能力差距正逐步缩小，没有谁能形成绝对壁垒。 章庆元称：“过去三年，行业一度担心 OpenAI 会…
大模型能否应对"HardChoices"？这篇Konvens 2026论文给出了前瞻性测评。
arXiv:2607.11471v1 Announce Type: new Abstract: A lot of research attention has been devoted to checking whether large language models (LLMs) are poli…
前沿模型与开源模型差距实测：哪些任务只有Opus/GPT能搞定？
ive been seeing a recurring claim that open (weight) models 6 months behind the frontier are good enough for the majority of ‘work’. if you've had a c…
一篇剖析大模型排序能力与偏差的前沿论文，揭示三元组与分诊机制背后的伦理陷阱，AI从业者必读。
arXiv:2606.30412v1 Announce Type: cross Abstract: From housing allocation for households experiencing homelessness to triage in emergency departments,…
实证研究揭示脚本与格式对LLM数值推理能力的显著影响，为优化大模型数学表现提供新视角。
arXiv:2601.15251v2 Announce Type: replace Abstract: Large language models (LLMs) have achieved impressive proficiency in basic arithmetic, rivaling hu…
探讨为何更聪明的模型也难以成为真正的AI同事，揭示当前AI的能力边界与协作本质。
Article URL: https://usize.github.io/blog/2026/april/why-no-ai-coworkers.html Comments URL: https://news.ycombinator.com/item?id=48650166 Points: 1 # …
用一条简单汽车清单提问，揭示当前AI模型常识推理的短板。
Claude: please give me a list of famous cars named after fish, like the Plymouth Barracuda for example. Comments URL: https://news.ycombinator.com/ite…
从Anthropic的一次演示切入，透视AI主动性、创造力与安全边界的现状，施奈尔带你冷静审视技术浪潮中的暗流。
On June 9th, Anthropic released its Fable generative AI model. Three days later, the US government classified it as a dangerous munition, and used its…
即便警告危险，Anthropic仍推出能发现与利用漏洞的AI模型，双刃剑效应凸显。
The US government crackdown on Anthropic’s Claude Fable 5 and Mythos 5 hides a glaring truth: AI models with advanced hacking capabilities will soon b…
提出基于低秩因子的LLM评估新范式，突破传统基准分数局限，揭示模型真实能力。
arXiv:2507.20208v2 Announce Type: replace Abstract: Current evaluations of large language models (LLMs) rely heavily on a growing collection of benchm…
对照实验揭示GAIA基准中脚手架如何混淆模型真实能力，为agent评估提供关键修正。
arXiv:2606.08529v1 Announce Type: cross Abstract: Published agent capability scores conflate what a model can do with what its scaffold lets it do, an…
Anthropic新模型Claude Fable 5在Vercel AI Gateway上线，内置安全分类器抵御高风险用途。
Claude Fable 5 from Anthropic is now available on AI Gateway . A Mythos-class model, Fable 5 is a notable step up over prior Claude models on long-run…
探索LLM如何将工具推理内化，摆脱对外部文档的依赖，提升工具掌握效率的新框架。
arXiv:2604.10788v2 Announce Type: replace-cross Abstract: Tool-Integrated Reasoning (TIR) has emerged as a promising direction by extending Large Lang…
Claude Opus 4.7在106个真实办公任务中完全通过率仅3.8%，这场残酷评测撕碎了AI全自动办公的幻想。
UniPat AI 发布 SaaS-Bench 评测，Claude 等主流大模型在真实办公任务中完全通过率最高仅 3.8%，AI 全自动办公远未落地。
轻量级大模型在法律AI中展现潜力，这篇论文系统探索了小于2B参数模型在法院观点生成任务上的表现。
arXiv:2605.16770v1 Announce Type: new Abstract: Criminal Court View Generation (CVG) is a critical task in Legal Artificial Intelligence (Legal AI), i…