精选归档 · 第 11 页

201220 条 · 共 274

3月12日3月12日周四

星期四 · 2 条
22:10
Ethan Mollick:One Useful Thing(RSS)精选
事物的形态

盘点当前阶段的核心现状与关键特征,基于现有趋势分析接下来可能发生的重要变化与未来走向。


推荐理由:Ethan Mollick 深度剖析 AI 现状与未来走向,洞察犀利

3月9日3月9日周一

星期一 · 1 条
02:00
Andrej Karpathy@karpathy精选
自动研究下一站:异步协作的AI研究社区The next step for autoresearch is that it has to be asynchronously massively collaborative for agents (think: SETI@home style). The goal is not to emulate a single PhD student, it's to emulate a research community of them.Current code synchronously grows a single thread of commits in a particular research direction. But the original repo is more of a seed, from which could sprout commits contributed by agents on all kinds of different research directions or for different compute platforms. Git(Hub) is *almost* but not really suited for this. It has a softly built in assumption of one "master" branch, which temporarily forks off into PRs just to merge back a bit later.I tried to prototype something super lightweight that could have a flavor of this, e.g. just a Discussion, written by my agent as a summary of its overnight run: https://github.com/karpathy/autoresearch/discussions/43 Alternatively, a PR has the benefit of exact commits: https://github.com/karpathy/autoresearch/pull/44 but you'd never want to actually merge it... You'd just want to "adopt" and accumulate branches of commits. But even in this lightweight way, you could ask your agent to first read the Discussions/PRs using GitHub CLI for inspiration, and after its research is done, contribute a little "paper" of findings back.I'm not actually exactly sure what this should look like, but it's a big idea that is more general than just the autoresearch repo specifically. Agents can in principle easily juggle and collaborate on thousands of commits across arbitrary branch structures. Existing abstractions will accumulate stress as intelligence, attention and tenacity cease to be bottlenecks.autoresearch的演进方向应是异步大规模协作,类似SETI@home模式,目标并非模拟单个PhD学生,而是构建多agents研究社区。当前Git/GitHub的主分支机制限制了分布式创新,未来应允许agents在任意分支并行探索不同方向,通过Discussion或PR分享发现而非合并代码。随着智能体算力与注意力瓶颈消失,现有代码协作抽象将面临根本性重构。另有 1 家信源报道X:Andrej Karpathy (@karpathy)
推荐理由:顶级AI科学家提出Agent科研新范式,从模拟个人转向构建分布式智能协作网络

3月3日3月3日周二

星期二 · 1 条
00:00
LlamaIndex:产品、工程与评测精选
AI 评分 60/100
LlamaIndex:不止是 RAG 框架,转向智能体文档处理

LlamaIndex 官方宣布公司使命收窄为构建智能体文档处理基础设施(OCR、抽取与工作流),不再定位为单纯的 RAG 框架。


推荐理由:官方复盘了通用 LLM 框架价值下降的原因,并说明转向文档基础设施的动机和产品布局,读者可借此理解框架类公司的转型逻辑。

3月2日3月2日周一

星期一 · 1 条
08:00
Answer.AI 官方研发博客(RSS)精选
AI 评分 62/100
Answer.AI 分析:OpenAI 与战争部合同能否冻结自主武器相关法律

Answer.AI 的 Jeremy Howard 与 Luke Versweyveld 撰文分析 OpenAI 与战争部合同中“all lawful purposes”条款的含义,指出 OpenAI 方面称该条款指“签约时适用的法律”,但合同语言未包含冻结条款。


推荐理由:文章援引美国合同法判例逐条分析 OpenAI 与国防部合同的用语,说明为何签约时的法律未必能冻结未来的立法变化。

2月26日2月26日周四

星期四 · 1 条
00:00
Anthropic:Newsroom(网页)精选
Anthropic CEO就国防部谈判发表声明

Anthropic CEO Dario Amodei声明,尽管Claude已广泛用于美军情报分析、网络作战等任务,且公司曾主动切断数亿美元收入阻止中国关联企业使用,但拒绝两项用途:大规模国内监控和完全自主武器。Amodei认为前者威胁民主价值,后者技术不可靠且缺乏监督。国防部威胁将其标记为"供应链风险"并强制移除安全措施。Anthropic坚持原则,但表示如被移除将确保平稳过渡,希望继续服务国防。


推荐理由:Anthropic CEO声明宁可退出军方合作,也不开放自主武器与大规模监控权限

2月25日2月25日周三

星期三 · 1 条
00:06
Nathan Lambert:Interconnects(RSS)精选
蒸馏对中国 LLM 到底有多重要?

针对 Anthropic 关于"蒸馏攻击"的最新论述,分析模型蒸馏技术对中国大语言模型的实际影响。探讨通过蒸馏 GPT、Claude 等模型来训练中国 LLM 的效果与争议,评估该方法在提升模型性能与降低训练成本方面的作用,以及可能引发的知识产权与安全问题。


推荐理由:技术权威视角拆解'蒸馏攻击',厘清中国大模型能力来源争议

2月18日2月18日周三

星期三 · 1 条
01:27
Nathan Lambert:Interconnects(RSS)精选
开源模型的永无止境的追赶

开源模型与闭源巨头(如 GPT、Claude)之间的能力差距持续存在,形成永无止境的追赶态势。文章探讨了知识蒸馏技术对缩小差距的作用,分析了开源与闭源模型在创新时间尺度上的差异,以及开源模型如何通过专业化模型在特定领域寻找获胜路径。同时指出当前开源生态在基础研究和资源投入上的缺失环节,并评估了这种追赶模式的可持续性。


推荐理由:开源与闭源鸿沟难越,但蒸馏与专业化或是破局关键

2月11日2月11日周三

星期三 · 1 条
14:11
Steve Yegge:Medium(RSS)精选
AI 评分 75/100
AI 吸血鬼

AI(如 Claude Code)确实能带来 10 倍生产力提升,但创造的价值大部分被公司捕获,员工可能过度劳累却收获甚微。微软内部已自发大量采用 Claude Code。这种效率加速迫使工作节奏不断加快,导致许多早期采用者(包括作者自己)出现严重的“午睡攻击”和日常疲劳。公司作为资本机器难以放缓脚步,形成一种让从业者无论是否使用 AI 都被持续“抽血”的困境。


推荐理由:Steve Yegge 提出的 'AI 吸血鬼' 概念精准捕捉了 AI 生产力带来的过劳危机,并给出了对抗资本压榨的实用公式,每个在 AI 浪潮中奔命的开发者都该停下看看。

2月6日2月6日周五

星期五 · 1 条
00:45
Dwarkesh Patel:Podcast & Blog(RSS)精选
埃隆·马斯克--「36个月内,部署AI最便宜的地方将是太空」

埃隆·马斯克预测,36个月后太空将成为部署人工智能成本最低的地点。他指出,长期专注于软件领域的从业者即将面临硬件层面的严峻挑战。这一判断暗示,随着AI算力需求爆发式增长,地面数据中心的能源消耗与散热限制将推高计算成本,而太空环境凭借丰富的太阳能和天然散热优势,可能在未来三年内成为AI基础设施部署的更经济选择。


推荐理由:Musk预判三年内太空将成为AI算力最优解,软件工程师需直面硬件物理约束

2月5日2月5日周四

星期四 · 2 条
02:00
Nathan Lambert:Interconnects(RSS)精选
Nvidia 为何构建开源模型:对话 Bryan Catanzaro

Interconnects 第17期访谈中,Nvidia 副总裁 Bryan Catanzaro 系统回顾了 Nemotron 开源模型项目的技术演进与战略定位。访谈涵盖该系列模型从研发初期到当前版本的迭代历程,剖析了英伟达在开源 AI 领域的布局逻辑,并披露了 Nemotron 在合成数据生成与模型训练效率方面的最新进展及未来规划。


推荐理由:NVIDIA副总裁亲述开源模型战略,揭示芯片巨头如何通过开放生态锁定行业标准
00:00
Anthropic:Engineering(事故复盘 + 工程实践 · 网页)精选
AI 评分 78/100
用并行Claude智能体团队从零构建C编译器

研究人员采用“智能体团队”方法,让多个Claude实例在无人工干预下并行协作开发代码。为进行压力测试,团队指派16个智能体从零编写一个能编译Linux内核的Rust版C编译器。项目消耗近2000次会话和约2万美元,最终产出10万行代码的编译器,可成功在x86、ARM和RISC-V架构上构建Linux 6.9内核。研究重点在于设计支持长时间自主运行的智能体团队框架,包括如何编写测试以保持智能体不偏离方向,以及如何通过基于文本文件的锁机制协调多智能体并行任务分配。


推荐理由:Anthropic 研究员用 16 个 Claude 并行写了个能编译 Linux 内核的 C 编译器,2000 次会话花了两万刀。真正值钱的不是编译器本身,而是他总结的 agent 团队协作方法论,做多 agent 系统的人该逐段拆。

2月4日2月4日周三

星期三 · 1 条
02:31
Jim Fan@DrJimFan精选
AI 评分 72/100
从"下一个词预测"到"世界建模":AI预训练的第二范式http://x.com/i/article/2018744045779238912The Second Pre-training ParadigmNext word prediction was the first pre-training paradigm. Now we are living through the second paradigm shift: world modeling, or “next physical state prediction”. Very few understand how far-reaching this shift is, because unfortunately, the most hyped use case of world models right now is AI video slop (and coming up, game slop). I bet with full confidence that 2026 will mark the first year that Large World Models lay real foundations for robotics, and for multimodal AI more broadly.In this context, I define world modeling as predicting the next plausible world state (or a longer duration of states) conditioned on an action. Video generative models are one instantiation of it, where “next states” is a sequence of RGB frames (mostly 8-10 seconds, up to a few minutes) and “action” is a textual description of what to do. Training involves modeling the future changes in billions of hours of video pixels. At the core, video WMs are learnable physics simulators and rendering engines. They capture the counterfactuals, a fancier word for reasoning about how the future would have unfolded differently given an alternative action. WMs fundamentally put vision first.VLMs, in contrast, are fundamentally language-first. From the earliest prototypes (e.g. LLaVA, Liu et al. 2023), the story has mostly been the same: vision enters at the encoder, then gets routed into a language backbone. Over time, encoders improve, architectures get cleaner, vision tries to grow more “native” (as in omni models). Yet it remains a second-class citizen, dwarfed by the muscles the field has spent years building for LLMs. This path is convenient. We know LLMs scale. Our architectural instincts, data recipe design, and benchmark guidance (VQAs) are all highly optimized for language.For physical AI, 2025 was dominated by VLAs: graft a robot motor action decoder on top of a pre-trained VLM checkpoint. It’s really “LVAs”: language > vision > action, in decreasing order of citizenship. Again, this path is convenient, because we are fluent in VLM recipes. Yet most parameters in VLMs are allocated to knowledge (e.g. “this blob of pixels is a Coca Cola brand”), not to physics (“if you tip the coke bottle, it spreads into a brown puddle, stains the white tablecloth, and ruins the electric motor”). VLAs are quite good in knowledge retrieval by design, but head-heavy in the wrong places. The multi-stage grafting design also runs counter to my taste for simplicity and elegance.Biologically, vision dominates our cortical computation. Roughly a third of our cortex is devoted to processing pixels over occipital, temporal, and parietal regions. In contrast, language relies on a relatively compact area. Vision is by far the highest-bandwidth channel linking our brain, our motors, and the physical world. It closes the “sensorimotor loop” — the most important loop to solve for robotics, and requires zero language in the middle.Nature gives us an existential proof of a highly dexterous physical intelligence with minimal language capability. The ape.I’ve seen apes drive golf carts and change brake pads with screwdrivers like human mechanics. Their language understanding is no more than BERT or GPT-1, yet their physical skills are far beyond anything our SOTA robots can do. Apes may not have good LMs, but they surely have a robust mental picture of "what if"s: how the physical world works and reacts to their intervention.The era of world modeling is here. It is bitter lesson-pilled. As Jitendra likes to remind us, the scaling addicts, “Supervision is the opium of the AI researcher.” The whole of YouTube and the rise of smart glasses will capture raw visual streams of our world at a scale far beyond all the texts we ever train on.We shall see a new type of pretraining: next world states could include more than RGBs - 3D spatial motions, proprioception, and tactile sensing are just getting started.We shall see a new type of reasoning: chain of thought in visual space rather than language space. You can solve a physical puzzle by simulating geometry and contact, imagining how pieces move and collide, without ever translating into strings. Language is a bottleneck, a scaffold, not a foundation.We shall face a new Pandora’s box of open questions: even with perfect future simulation, how should motor actions be decoded? Is pixel reconstruction really the best objective, or shall we go into alternative latent spaces? How much robot data do we need, and is scaling teleoperation still the answer? And after all these exercises, are we finally inching towards the GPT-3 moment for robotics?Ilya is right after all. AGI has not converged. We are back to the age of research, and nothing is more thrilling than challenging first principles.作者指出,AI预训练正经历从"下一个词预测"到"世界建模"的根本性范式转变。世界模型的核心是预测给定行动后的下一个物理状态序列,本质上是可学习的物理模拟器,并将视觉置于首位。相比之下,当前主流的视觉语言模型本质是语言优先,视觉是次要输入。生物智能中视觉处理占据皮层计算的主导地位,是连接大脑、动作与物理世界的高带宽通道。作者以猿类为例,证明强大的物理智能可独立于高级语言存在。他预测,2026年大型世界模型将为机器人技术和多模态AI奠定真正基础,而YouTube等平台的海量视觉数据将远超文本规模,推动这一新范式发展。
推荐理由:Jim Fan 把世界模型定义为第二次预训练范式转移,核心论点是视觉优先而非语言优先,这个框架对做机器人和多模态的人是真正的路线判断,不是又一篇水文。

1月28日1月28日周三

星期三 · 1 条
00:55
Ethan Mollick:One Useful Thing(RSS)精选
管理是 AI 的超能力

智能体(agents)时代,管理能力将成为人类 thriving 的核心超能力。在 AI 主导的未来,懂得如何管理比单纯的技术能力更能决定成败。


推荐理由:沃顿教授 Ethan Mollick 深度解析 Agent 时代的管理变革与机遇

1月22日1月22日周四

星期四 · 2 条
01:05
Nathan Lambert:Interconnects(RSS)精选
精通 Agents

AI Agents 的能力正逼近关键临界点,其性能飞跃已超出传统工作模式的承载范围。这要求从业者必须重新界定工作范畴、重构项目管理流程并革新任务执行策略。从需求规划到交付标准,现有方法论面临全面调整,组织与个人亟需掌握与智能体协作的新范式,以适应这一技术变革带来的深层影响。


推荐理由:Agent工具迫使开发者重构工作流,资深研究者分享进阶路径
00:00
METR:Notes(网页)精选
AI 评分 62/100
METR 时间视界论文作者澄清时间视界的局限

METR 时间视界论文主要作者发文澄清该指标的常见误读,称过去 9 个月 AI 时间视界增长约 6 倍。


推荐理由:作者作为论文主要作者逐条澄清方法局限,读者可据此更准确地使用和理解时间视界这类指标。