2026年11月3日 技术热点总结

项目简介：Craft Anything: understand how any software works, and craft your own.

技术栈：Rust

核心介绍：Craft Anything English · 日本語 · 简体中文 · 한국어 See how any website, app or game works, then have AI build your own version; an open-source, clean-room engine in pure Rust. Point it at a target, say what you want, and it studies the original with real inspection tools, has a coding agent build a new project, checks the result against the original and refines it. Download Craft Anything · ArtCraft · A

项目数据：⭐ 4,061 Stars，🍴 662 Forks

项目简介：RUN · 润：海外移居手册（欧洲 36 国 · 110 条路线 · 主要移民地区）

技术栈：Python

核心介绍：数字游民 · 技术移民 · 欧盟蓝卡 · 创业自雇 · 投资居留 > 本手册仅供信息参考，不构成移民、法律、税务或投资建议。2025–2026 年大量路线关闭或改规则，网上很多旧攻略已经失效——任何决定前请以各国官方页面为准。详见 免责声明。 为境外雇主或客户远程工作，想换个国家生活 15 条欧洲路线 · 月入 €330 起 拿到当地雇主 offer：工作 → 居留 → 永居 → 入籍 50+ 条路线 · 最快 21 个月永居 毕业后 6–36 个月找工作签证，再转工作签 德国 18 个月 · 芬兰 2 年 · 爱尔兰硕士 2 年 在当地开公司、做自由职业 爱沙尼亚、爱尔兰 STEP、荷兰、丹麦… 靠养老金、租金、存款生活，不在当地工作 葡萄牙 D7 €920/月起 · 西班牙 NLV 用投资换身份，住得少也能保持 希腊 · 塞浦路斯 · 马耳他 · 匈牙利 🇺🇸 H-1B · NIW · EB-5 🇨🇦 Express Entry · 省提名 🇦🇺 澳大利亚 189 · 482 · 858 🇳🇿 SMC · 投资移民 🇯🇵 高度人才 · 经

项目数据：⭐ 1,518 Stars，🍴 88 F…

项目简介：Install Linux next to Windows without a USB stick

技术栈：PowerShell

核心介绍：Install Linux next to Windows without a USB stick. hop is a single Windows app that installs Linux alongside Windows. Pick a system, choose a drive, drag to choose how much space it gets, and click Install. There’s no USB stick to find and nothing to install checksum where the project publishes one. You can also use an ISO you already have. move. Make room clears them in one click and puts your settings back afterwards.

项目数据：⭐ 1,385 Stars，🍴 47 Forks

Perceiving human body motion is critical for multimodal systems that understand and assist people. Yet current evaluations of Multimodal Large Language Models (MLLMs) largely focus on objects, scenes, and coarse actions, leaving their ability to reason about the body itself poorly assessed. This is especially challenging from the egocentric perspective, in which the body is only partially visible and motion must be inferred from occluded and temporally distributed cues. We introduce SOMA, a benchmark for fine-grained body-motion understanding from egocentric video, comprising 1,632 human-ve…

Large Language Models (LLMs) are increasingly used in multi-turn agentic workloads that require sustained interaction with complex environments. Existing reflective prompt optimizers such as GEPA learn from failed rollouts, but agentic failures can also come from prompt weaknesses, harness limitations, stochastic execution, or noisy judges. We argue that such failures may not provide useful learning signal, and may even be harmful; and thus it is essential to curate which tasks are provided to the prompt optimizer for better performance. We introduce GEPA-Curator, a framework that augments …

Federated learning (FL) keeps raw data local, but most deployments still rely on a trusted coordinator for client admission, checkpoint distribution, update collection, aggregation, and rewards. This paper presents AuditFL, a contract-backed clearing layer for auditable FL in open settings. AuditFL keeps training and aggregation off chain, stores model artifacts in an assumed reliable off-chain artifact store, and records only content identifiers and economic state on chain. Clients reserve update tickets with deposits, publish update identifiers, and later claim refunds or rewards; the tas…

Fine-tuning Large Language Models (LLMs) on instruction data can quietly erode their safety, even when no training example looks harmful. Because this safety-degrading data looks benign, standard content filters struggle to catch it. We tackle this problem by developing SimpleLoRIF, a scalable gradient embedding method that identifies fine-tuning examples whose gradients align with unsafe behavior. By simplifying a low-rank attribution method and restricting backpropagation to the final four transformer blocks, SimpleLoRIF can embed datasets on the order of a million examples, boosting grad…

Do research agents have taste, the judgment of which experiments are worth running before their outcomes are known? We give six frontier models seven recent NLP papers. Each model, working as a research agent alone or as a team, proposes minimum viable experiments (MVEs), forecasts whether each will pass a threshold it sets and ranks them. A separate executor then attempts every MVE, including those ranked last, and three model auditors judge the results. Of 415 MVEs, 194 are judged sound, valuable and new, but fewer than half of these (73) are also judged beyond the seed paper and non obvi…

Direct-policy World Action Models (WAM) have demonstrated the benefits of decoupling action generation from explicit future visual prediction for efficient inference in autonomous driving and embodied control. However, they still require tens of denoising steps to achieve competitive action quality, incurring substantial latency and hindering real-time deployment. Simply reducing the number of denoising steps will collapse the nonlinear multi-step denoising trajectory into coarse updates, substantially degrading action quality. To reduce the denoising steps while enabling real-time action g…