AI exposure by U.S. occupations and work tasks and the effect on wages
作者：Chiara Chanoi，华盛顿公平增长中心 Chris Bangert-Drowns，华盛顿公平增长中心 摘要：这项对劳动力市场人工智能暴露程度的分析建立在皮尤研究中心和人工智能公司 Anthropic 之前的工作以及 Equitable Growth 自己的工作质量系列的基础上，确认了性别、种族、教育程度以及美国职业和工作任务对人工智能暴露程度的差异。 对工资的影响首先出现在公平增长上。
Albanese visits Trump as US democracy circles a golden drain
安东尼·阿尔巴内塞 (Anthony Albanese) 与唐纳德·特朗普 (Donald Trump) 的会面相对平淡无奇，但当谈到总统对 AUKUS 和关键矿产的“承诺”时，问题就出在细节上。 阿尔巴内塞 (Anthony Albanese) 在美国民主圈子出现黄金流失之际拜访特朗普后，首先出现在澳大利亚研究所 (The Australia Institute) 上。
NOVEMBER 2025 ACTIVE-DUTY CORPORAL AND SERGEANT PROMOTION AUTHORITY
R 222006Z 10 月 25 日MARADMIN 497/25MSGID/GENADMIN/CMC 华盛顿 DC MRA MM//SUBJ/2025 年 11 月
FY25 DC M&RA PERSONNEL AND PROGRAMS AWARDS SOLICITATION
R 221703Z 10 月 25 日 MARADMIN 496/25MSGID/GENADMIN/CMC MRA DAG SECT 华盛顿 DC//SUBJ/FY25 DC M&RA
FY26 FOREIGN AREA STAFF NON-COMMISSIONED OFFICER (FAS) PROGRAM STUDY TRACK SOLICITATION MESSAGE
R 211456Z OCT 25MARADMIN 493/25MSGID/GENADMIN/CMC PPO PL WASHINGTON DC//SUBJ/FY26 国外地区工作人员
FY26 COMMANDANT'S PROFESSIONAL INTERMEDIATE-LEVEL EDUCATION BOARD (CPIB) SELECTION RESULTS
R 211256Z 10 月 25 日 MARADMIN 492/25MSGID/GENADMIN/CMC 华盛顿 DC MRA MM//SUBJ/FY26 指挥官
SOLICITATION FOR THE 2026 MARINE CORPS INSTALLATIONS AND LOGISTICS AWARDS
R 221437 OCT 25MARADMIN 495/25MSGID/GENADMIN/CMC L 华盛顿特区// 2026 年主题/征集
SOLICITATION FOR THE 2026 MARINE CORPS AVIATION AWARDS
R 211617Z OCT 25MARADMIN 494/24MSGID/GENADMIN/CMC 华盛顿 DC AVN ASM//SUBJ/征集
R 201511Z 10 月 25 日MARADMIN 491/25MSGID/GENADMIN/CMC 华盛顿 DC//SUBJ/FY26 海军陆战队降落伞
Metagenomi generates millions of novel enzymes cost-effectively using AWS Inferentia
In this post, we detail how Metagenomi partnered with AWS to implement the Progen2 protein language model on AWS Inferentia, achieving up to 56% cost reduction for high-throughput enzyme generation workflows. The implementation enabled cost-effective generation of millions of novel enzyme variants u
When Transformers Sing: Adapting SpectralKD for Text-Based Knowledge Distillation
探索 Transformers 的频率指纹以指导更智能的知识蒸馏《当 Transformers Sing：采用 SpectralKD 进行基于文本的知识蒸馏》一文首先出现在 Towards Data Science 上。
Soulgen AI Clone Maker: My Unfiltered Thoughts
以数字方式“克隆”某人的想法一直感觉像是科幻小说中的承诺之一，远远超前于我们的时代。 Then I stumbled across Soulgen’s AI Clone Maker, and honestly, it feels like they’ve managed to drag that fantasy into something you can actually play with today. Creating AI clones here isn’t just about spitting out a lookalike—it’s about building a person
China’s ShengShu Unveils Vidu Q2 — The Bold New Contender Taking Aim at OpenAI’s Sora
生成视频竞赛变得更加激烈。中国人工智能初创公司圣数科技推出了其最新产品 Vidu Q2，该模型旨在挑战 OpenAI 的电影巨头 Sora。 The platform can generate full-motion clips from text prompts and up to seven reference images, giving creators the ability to blend faces, objects, and scenes into one continuous narrative.该模型的首次亮相是通过一份独家报告宣布的，这标志着中国深入推进生成视频领域的决
Tried Xmode AI Image generator for 1 Month: My Experience
探索 Xmode.ai 就像偶然发现了那个隐藏的地下派对——未经审查、可定制且毫无歉意的 NSFW。以下是其图像生成方面的内幕，专为任何渴望突破创意界限的人量身定制。 Xmode.ai 的成果 Xmode.ai 不会搞砸。它专为想要零过滤、最大程度的控制以及确保其创作保持私密性的创作者而打造。 If you’ve ever daydreamed about generating adult content that looks hyper-real, exactly how you imagined—this walk-through might just convince you to di
Type It, Tap It, Boom — Snapchat’s New Imagine Lens Turns Words into Wild Visuals
Snapchat 刚刚推出了一些非常疯狂的东西——一个让你的想象力完全占据屏幕的工具。 The new Imagine Lens turns simple text prompts into full-blown AI-generated images, and it’s now available to users in the U.S. If you’ve ever wanted to see yourself as a cyberpunk samurai or floating through space in a coffee cup, this feature’s got your
Brazil Turns WhatsApp Into a Bank Teller as Generative AI Transforms Everyday Finance
在巴西，与银行聊天有了全新的含义。 A wave of financial institutions and fintechs are rolling out generative AI systems that let users transfer money, pay bills, and check balances directly through WhatsApp, using conversational prompts instead of traditional app navigation. It’s the first large-scale integration of
Contract AI’s Reliability Problem: When AI Gets It Wrong
作者：Pedram Abrari，Pramata 首席技术官。这是我们由三部分组成的系列文章中的第二篇文章，探讨为什么将合约加载到大型语言模型中并不是一种有效的方法......
