2026年9月7日 技术热点总结

项目简介：Open-source 3D anatomy explorer: 2,234 selectable BodyParts3D meshes, system layers, search, and exploded views.

技术栈：TypeScript

核心介绍：An interactive 3D anatomy explorer built with React, Three.js, and shadcn/ui. Take the BodyParts3D adult male reference apart into 2,234 individually selectable meshes, explore 15 anatomical systems, and search 3,432 named concepts. Requires Node.js 22.13 or newer. No API keys or accounts are needed. Open http://localhost:3016. To build the static site, run npm run build; the output is in dist/. node scripts/validate-atlas.mjs no…

项目简介：Local-first WeChat intelligence system with a read-only CLI, Codex skills, searchable chat history, daily briefings, follow-ups and opportunity tracking.

技术栈：Python

核心介绍：微信个人情报库：把本地微信聊天变成可检索、可核查、可行动的个人情报，包括联系人历史、群聊主题、待回复、承诺、商机、复联线索，以及任意指定时间范围的情报报告。 这不是 Prompt 大礼包，而是一个独立的微信旗舰项目。仓库同时提供只读数据入口和情报工作流，并配有可执行入口、边界、测试和全虚构样例。 当前首发版本为 v0.9.2-preview.2。微信相关代码已经具备公开测试条件，但不是“安装后自动读取所有人的完整微信历史”：完整数据库模式需要本人授权的本地数据库和访问材料。Reader 核心不获取密钥、不重签名、不注入、不 Hook 微信；可选的实验性接入助手有独立授权和副作用边界，见下文。 微信能力在代码中分成四层，方便独立测试和维护；对用户仍是一套产品、一次安装： Rion 的通用 Skill 合集 rionwu-skills 只负责收录、发现和链接本项目，不复制微信读取器源码或 Git 历史。

*…

项目简介：Minimal graphics API. Built on top of latest Vulkan extensions. As close as possibly to my “No Graphics API” blog post and the SIGGRAPH talk.

技术栈：C++

核心介绍：NoGraphicsAPI is an experimental Vulkan 1.4 implementation of the ideas in Sebastian Aaltonen’s *No Graphics API*. It explores how much of a conventional graphics API disappears when shaders use 64-bit GPU pointers, texture and sampler descriptors live in application-owned GPU memory, and synchronization describes hazards instead of The Vulkan backend is implemented and exercised by three example applications. Metal sup…

项目简介：Free open-source project designed for turning youtube-viedos into viral short videos. Highlight detection, subtitles, translation, voiceover, all in one for your content.

技术栈：Python、ai、ai-video、python、short-video-maker、video-generation

核心介绍：A free open-source project designed to turn youtube-videos into viral short videos. Highlight detection, subtitles, translation, voiceover, all in one for your content: no pre-clip credits or any watermarks. Designed for creators who want an alternative to short-video SaaS tools like OpusClip or Vidyo.ai for free.

关键特性：API: Fr…

项目简介：A macOS app that pins usage limits from Claude Code, Cursor, Codex, and Antigravity to a screen edge.

技术栈：Swift

核心介绍：A macOS app that pins a small black notch to a screen edge, showing how much of each coding assistant’s usage limit you have burned — and whether it is still working, done, or waiting on you. Hover a ring for its limit windows and when they reset. Claude’s ring shows the same current session window Claude Code’s own /usage leads with, so the Codenotch never signs in anywhere. Every reading is borrowed from a credential

项目数据：⭐ 687 Stars，🍴 91 Forks

Robot learning increasingly depends on broad and diverse demonstrations, yet collecting robot data remains expensive and poorly suited to covering the long tail of real-world tasks. To address this bottleneck, we introduce RoboTok, an internet-scale data engine that, given a query human manipulation video, retrieves manipulation-relevant human demonstrations from web videos for training dexterous robot policies. Specifically, we learn a latent motion space from 3D hand trajectories expressed in estimated actor-centered reference frames. This representation enables manipulation behaviors to …

Speech brain-computer interfaces (speech BCIs) translate neural activity into language, offering a path towards restoring speech for people with paralysis and, more broadly, enabling new forms of natural human-computer interaction. Despite this promise, the field lacks a common measure of progress because systems use different datasets, recording methods, types of speech, and vocabularies, so their reported scores are rarely comparable. Underlying this measurement problem are two unresolved questions: (i) what distribution of words should a speech BCI enable a user to communicate, and (ii) …

Visual fluency in generated video does not imply physical reliability, and a scalar quality score alone is incapable of indicating the obligation a clip violates or the moment it fails. We present VeriPhy, an auditable physical-verification system in which a text-only planner compiles the prompt into typed physical obligations and a statically validated execution plan before any frame is observed. During execution, observations gate and scope only declared calls to frozen low-level experts (e.g., segmentation and tracking, counting, eleven typed physical measurements over the resulting trac…

Reinforcement learning with verifiable rewards (RLVR) substantially improves single-sample accuracy (pass@1) but causes the policy's solution space to contract, diminishing the returns of test-time scaling. In this work, we investigate where inside a reasoning trajectory this breadth is lost: does the policy fail to access a valid solution family, or does it fail to execute computation once initiated? To disentangle access from execution, we analyze the Countdown task, whose solution space can be exhaustively enumerated into discrete entrance families defined by the first operand and operat…

Reinforcement Learning from Verifiable Rewards works well when a task has a programmatic checker, but most long-horizon agent domains have none. We work in the outcome-blind setting, where ground-truth success signals are not available. Multi-criteria rubrics are a popular way to supply such a reward; they are scored once per trajectory, but a single scalar is a poor signal across tens of steps. We propose DRACO: Distributing Rubric-based Advantage for Credit Optimization. It generates rubrics dynamically during training to track the policy's evolving capability, scores those rubrics once p…

For scientific progress, we need benchmarks that test the limits of state-of-the-art models, and evaluation methods that inform us about failure cases. As models get stronger, standard benchmarks for machine translation are approaching saturation. Further, automatic translation metrics are unreliable, vulnerable to reward-hacking, and provide unactionable assessments. Even gold human evaluation is not problem-free, because it often lacks reproducibility, objectivity, and scalability. Overall, this prevents us from tracking objective progress in the field and identifying pathways for improve…