2026年10月29日 技术热点总结

项目简介：Free, open-source, offline checks for traces of known mercenary spyware. Works on Windows, macOS and Linux, Android phones and iPhone backups.

技术栈：Python

核心介绍：A free, open-source check for traces of known mercenary spyware (Pegasus, Predator, Graphite, Candiru and others) on computers, Android phones and iPhone backups. temporary database snapshots and the reports you ask for. upload feature. Python network calls are blocked and redirected paths are refused. This is not an operating-system network sandbox; see PRIVACY.md. what could not be checked. A clean result is never…

技术栈：C

核心介绍：English · 中文 · Українська HakiOS runs the open-source XNU kernel (xnu-8796.121.2) with my own boot loader and my own drivers on an Android phone with a Qualcomm SoC, and boots the stock iOS 16.5 user space on top of it: launchd, backboardd, SpringBoard, the apps. The same thing runs in QEMU (virt, aarch64) on a Nothing is flashed. The loader goes into RAM through fastboot boot, the system lives in RAM for

项目数据：⭐ 1,900 Stars，🍴 326 Forks

项目简介：中文版 AI 姿势参考图库 · 1439 组 / 2878 张图片 / 完整汉化 · 零依赖静态画廊

技术栈：HTML、ai-art、chinese、cloudflare-pages、gallery、i18n、pose、reference-images、static-site

核心介绍：> 288 组 / 1439 张 AI 人体姿势参考图，日文数据完整汉化，附可直接部署的响应式静态画廊。 手机端预览 原站是一套面向 AI 绘画 / 3D / 摄影参考的姿势图库，元数据全部为日文，且不提供批量素材下载。本仓库做了三件事： 1. 全量镜像 —— 1439 条记录对应的 1439 张缩略图（JPG 300×400）与 1439 张原图（PNG 900×1200）全部落地，共 2878 个文件 / 659.5 MB。 2. 完整汉化 —— 日文字段逐条译为中文，覆盖 4 个体系、26 个分类、288 个姿势名、5 个视角，零遗漏。 3. 可用的前端 —— 零依赖响应式画廊，完整实现原站的分类筛选、多角度切换、分组浏览、灯箱导航与原图下载，并针对手机重新设计布局。 页面内置了完整的 SEO 与社交分享配置：

关键特性：4 个体系页签，显示各体系的图片数与姿势组数；姿势名、编号、ID、分类关键词搜索，支持 / 快速…

项目简介：AI digital-human talking-head videos from one photo: Renoise MCP generation + 5 editing styles + 48 creator-grade packaging components. Agent skill for Claude Code / Codex / OpenClaw / WorkBuddy …

技术栈：TypeScript

核心介绍：Script, lip-synced AI presenter, voice clean-up, captions, and pro-YouTuber-style packaging — fully automated by your AI agent. > This 27-second livestream ad has no real presenter. Face, voice, studio, captions and the price animation are all AI. > The only input was one sentence: *”Use this digital human to make a 30-second livestream ad for the WorkBuddy …

项目简介：Evidence-first genealogy skill for AI agents (Codex, Claude Code): church books, civil, military, deportation and repression records across the former USSR. Replies in your language.

技术栈：Python、agent-skills、archives、claude-code、codex、family-history、familysearch、genealogy、ussr

核心介绍：Coverage includes all 15 former Soviet republics: Russia, Ukraine, Belarus, Moldova, Estonia, Latvia, Lithuania, Georgia, Armenia, Azerbaijan, Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan. Archives are chosen by the event’s historical jurisdiction and current record custodian;…

项目简介：EL2 hypervisor and toolchain that boots macOS on the ASUS A16 (Snapdragon X2 Elite), with an Adreno Metal/IOGPU stack

技术栈：Python

核心介绍：Snapintosh boots Apple’s macOS 27 (build 26A428) on an ASUS Zenbook A16 laptop with a Qualcomm Snapdragon X2 Elite (X2E-96-100). It is an EL2 monitor (hypervisor) and the toolchain around it. The monitor runs Apple’s stock kernelcache (the J700/T8140 kernel) unmodified by default. It supplies the platform contracts that the kernel expects from Apple silicon: the SPTM and TXM monitor calls, translation of the kernel’s 16 KiB page tables, and…

项目简介：A public release of the Steal My Bitcoin challenge

技术栈：Python

核心介绍：sponsored with ❤️ by > One live PHP callback against the Steal My Bitcoin challenge, start to finish, in a single standard-library Python file. This README explains stealmybitcoin_poc.py, a single-file Python reproduction of the request sequence used in my Steal My Bitcoin CTF run. It uses only Python’s standard library. The script builds the requests itself, checks each leak before moving on, and records the HTTP bytes from a new run.

项目数据：⭐ 983 Stars，🍴 145 Forks

项目简介：High fidelity scanner for php2shell (CVE-2026-109618 and CVE-2026-109619)

技术栈：Python

核心介绍：A command-line tool for detecting CVE-2026-109618 and CVE-2026-109619 in PHP applications. The scanner sends a minimal multipart control request followed by up to three crafted multipart probes. The probes use PHP array keys nested 62 levels deep and increase from 1 MiB to 20 MiB, matching the detection logic used by Assetnote’s internal scanner.

项目数据：⭐ 958 Stars，🍴 140 Forks

This paper introduces Quaternion Approximate Networks (QUAN), a novel deep learning framework that leverages quaternion algebra for rotation equivariant image classification and object detection. Unlike conventional quaternion neural networks attempting to operate entirely in the quaternion domain, QUAN approximates quaternion convolution through Hamilton product decomposition using real-valued operations. This approach preserves geometric properties while enabling efficient implementation with custom CUDA kernels. We introduce Independent Quaternion Batch Normalization (IQBN) for training …

Forecasting an unresolved event requires finding relevant evidence and deciding how strongly it supports each outcome. Yet outcome-based training of Large Language Model (LLM) forecasters often begins with preassembled evidence at a single prediction date, leaving evidence acquisition outside the learning loop. We introduce AlphaBet, an agentic reinforcement learning (RL) framework that trains a single LLM policy to search, reason, and report probabilities across multiple pre-resolution dates. Each dated forecast is scored by a proper scoring rule against the eventual outcome, allowing one …

Phenomena responsible for emerging technologies originate from novel molecular states at buried interfaces. Yet quantitatively resolving these features is out of reach in most cases. Resonant scattering has demonstrated sensitivity to these local states but is limited due to a lack of accurate interpretive models. We leverage a machine learning (ML)-optimized optical model to analyze polarized resonant soft X-ray scattering (RSoXS) from molecular thin films. This new analysis establishes the direct and quantitative interpretation of molecular orientation, chemical identity and packing state…

Bengali, spoken by over 230 million people, remains severely under-resourced in spoken language research. Existing open Bengali ASR corpora are confined to short, read or crowd-sourced speech and lack speaker diarization and long-form audio.
We present Lipi-Ghor, a large-scale weakly-supervised Bengali speech corpus comprising 885 hours of audio sourced from 998 YouTube videos across 596 unique channels, spanning 17 diverse domains including audiobooks, drama, religious sermons, news, and regional dialects. Each recording is annotated in SSTT format (Speaker, Speech, Transcription, Timest…

Learning velocity fields directly in pixel space is difficult and slow. We show that coarse second-order image statistics can ease this task: frozen Wiener estimates improve velocity learning, with gains extending beyond the linear predictions they explicitly provide. Moreover, we observe similar but complementary gains from frozen bases and pixel decoding. These observations motivate Wiener-informed Decoding (WiD), which uses statistical knowledge both for direct prediction and to organize decoder computation without modifying native blocks. WiD combines a trainable Wiener prediction path …

Outcome RL raises pass@1 but not pass@k at large k (Yue et al., 2025). But every pass@1 is scored under a token budget. Every recipe also trains under one. On physics and math problems at fixed token budgets, we compare the pass@1 of 20 released RL models (1.5B to 32B) with that of the thinking models they started from. We ask whether each pass@1 gain is changed behaviour or new knowledge. We divide each change in pass@1 into three parts. Finishing measures answering before the budget runs out. Reliability measures being right once finished. Reach measures problems RL solves that its base a…