2026年8月9日 技术热点总结

项目简介：Turn documents into AI-ready Markdown with visual understanding

技术栈：Go、document-ai、docx-to-markdown、local-ai、markdown、multimodal、open-source、pdf-to-markdown、vision-language-model

核心介绍：Any document in. AI-ready Markdown out. Turn PDFs, Office files, scans, screenshots, charts, formulas, and diagrams into Markdown your AI can search, quote, and reason over. 简体中文 · English GitHub · Releases · Benchmark doc7 turns PDFs, Office files, scans, screenshots, charts, formulas, and diagrams into Markdown through your own OpenAI-compatible multimodal model. No required OCR stack. No …

项目简介：claude-red is a curated library of offensive security skills designed for the Claude skills system. Each skill is a structured SKILL.md file that primes Claude with expert-level methodology for a specific attack surface — from SQLi to shellcode, EDR evasion to exploit development.

技术栈：JavaScript、claude、claude-code、claude-code-plugin、claude-skills、codex、codex-skill、redteam、redteaming

核心介绍：Maintained by 0xwilliamortiz. claude-red is a curated library of offensive security skills for the Claude Skills system. Each skill is a structured SKILL.md file that primes Claude wi…

技术栈：JavaScript

核心介绍：Upload a video of the two-hand finger-frame gesture — get it back with an video-to-video model (motion, blinks, and all), then composited so the finger frame acts as a window into the animated version. style (full-quality mp4).* 1. Restyle — the uploaded video is sent to Gemini Omni Flash video editing with your chosen style (3D animated movie, anime, claymation, watercolor, or a custom prompt). This is a true video model: the whole

项目数据：⭐ 621 Stars，🍴 72 Forks

技术栈：JavaScript

关键特性：On-Device Inference: Uses LiteRT-LM to run the gemma4-e2b model entirely locally. No internet required after setup.；Voice Interface: Captures microphone audio, processes it, and sends it to the local model.；Optimized UI: Retro-terminal styling custom-built for small hardware screens (like Raspberry Pi displays).；Unified Startup: One script to launch the LLM server, the Python API, and the React frontend.；Python 3.10+。

项目数据：⭐ 574 Stars，🍴 77 Forks

项目简介：灵犀跨境-开源共建版

技术栈：TypeScript

核心介绍：灵犀跨境 Community 是一套面向跨境团队的多租户客户经营 SaaS。它将统一消息工作台、多语言翻译、意图与情绪分析、客户画像、知识库、商机、任务、团队协作和经营分析放在同一套产品中。 本仓库提供可独立运行的 Web、业务 API 和 PostgreSQL 数据层。Telegram 与 WhatsApp 网关实现不在本仓库中；平台仅保留可选的外部适配接口。详细边界见 开源范围。 > Community edition for multilingual customer operations. The repository includes the SaaS application, API and database layer, but excludes Telegram and WhatsApp gateway implementations and all production credentials. U[浏览器] –>|HTTPS / WebSocket| W[Nginx + React] W –>|REST / WS| A[Node.js API]

关键特性：多租户注册、登录、团队、角色权限、Token 撤销与审计日志。；统一消息中心、会话分页、已…

项目简介：A curated collection of machine learning and AI lecture notes from the world’s leading universities. This repository gives you access to the same lecture notes used by students at top institutions such as MIT, helping you learn from the very best educational resources available.

核心介绍：> A curated list of official, free, written course notes on AI/Machine Learning from top universities worldwide — the kind some departments publish instead of assigning a paid textbook. This is a companion project to Awesome Free AI Books, focused on a different (and much rarer) kind of resou…

项目简介：Racing to the bottom of CPU performance

技术栈：C

核心介绍：Instruction latency analysis usually focuses on performance Shame takes the opposite approach: searching for the absolute floor of single-instruction performance. high-latency MMIO region in the PCIe fabric, then starve the fabric while the load is in flight — a fleet of hammer cores pounds a different high-latency MMIO register with tight 4-byte reads, saturating the PCIe root complex and endpoint with non-posted transactions, so CPU 0’s 512-byte fxrstor64 must

项目数据：⭐ 542 Stars，🍴 5 Forks

技术栈：Vue

核心介绍：面向工业控制系统的智能异常检测、攻击溯源与联动响应平台。 InduSecAgent 将工业流程图、设备与点位说明、正常运行时序数据等多模态信息统一建模为工业时空图（Industrial Spatio-Temporal Graph，ISTG），利用图神经网络学习传感器、执行器和工艺环节之间的空间依赖与时间演化规律，为工业现场提供可解释、可追踪的安全监测能力。 > 当前仓库包含可视化演示界面、模型训练与推理代码、实时数据采集服务及示例数据。接入真实 PLC 前，请在隔离的测试网络中完成配置校验和风险评估。 A[工业图像与流程文档] –> D[工业依赖图构建] B[PLC 点位与设备描述] –> D C[正常运行时序数据] –> E[数据预处理] D –> F[工业时空图] F –> G[模型训练或预训练模型加载] H[PLC 实时数据] –> I[实时异常检测] I –> J[异常区间与组件定位] J –> K[传播路径溯源] K –> L[告警展示与联动响应] 浏览器访问终端中显示的本地地址，通常为 http://localhost:5173。

关键特性：多模态信息融合：支持工业流程图、工艺说明、PLC 点位描述和正常运行数据等输入。；工业依赖图构建：建立传感器、执行器、控制器与工艺过程之间…

Selecting a complete 3D object from a reconstructed scene with minimal user effort is essential for practical scene editing and embodied interaction. Existing 3DGS-based methods either retrain the Gaussian representation to embed per-object labels, or build dense multi-view SAM observations, both requiring heavy computation and dense viewpoint coverage that is rarely available in practice. We present GaussianSelector, a training-free framework for interactive 3D object selection from sparse views and sparse scribble guidance. Operating directly on native Gaussian primitives, we coarsen dens…

Robot learning is splitting into two bets: policies that bake competence into frozen weights (vision-language-action, or VLA, models), and agents that write and refine their own executable skills as code. This survey organises the field around that axis of weights versus skills. Its central analytical contribution is a deep-dive that arranges code-as-policy methods by their degree of self-improvement, from zero-shot program synthesis, through closed-loop self-repair and persistent skill memory, to the sparsely populated cell in which execution feedback, skill memory, and evolutionary search…

World models have attracted significant attention for their ability to capture and predict the structure and dynamics of the physical world. In this emerging landscape, Joint Embedding Predictive Architectures (JEPA) offer a particularly compelling direction.
We study a largely unexplored regime: populous, crowded, and chaotic Global South urban environments, which we call DENSEWORLD. Unlike the lower-density, lane-structured settings that dominate existing evaluations, these scenes exhibit soft spatial boundaries, extreme agent heterogeneity, persistent occlusion, and rapid social negoti…

Latent diffusion modeling (LDM), a prominent paradigm, utilizes tokenizers to map input signal to compressed representation. This dependency positions tokenizer as an integral part of generation process itself, since it affects learning speed, quality of synthesized samples and lay foundation for later applications. This report presents series of KVAE tokenizers for audio, image and video, all designed for subsequent text-conditioned generation: KVAE-Audio, a continuous full-band 48 kHz tokenizer with a 50 Hz latent of 64 channels; KVAE-3D — two causal video tokenizers for 4x16x16 and 4x8x…

Computer-use agents pay full frontier inference to re-derive routines their user has already performed, because an agent's memory today records what the user said, not what the user did. We compile passively captured screen activity into agent memory with a deterministic, zero-model pipeline: it segments a local capture stream into typed activity frames, bounded episodes carrying application, site, timing, input volume, and evidence pointers back to the raw rows, with no model in the loop, so the output is byte-identical, cacheable, and mechanically auditable. On one professional's single-u…

Data agents enable natural-language analytics over organizational workspaces, where relevant evidence may be scattered across databases, structured files, long documents, and multimedia. Existing benchmarks largely isolate structured querying, retrieval, or open-ended analysis, leaving heterogeneous evidence discovery, complete tabular outputs, and deterministic evaluation insufficiently unified. We introduce DataSpace, a benchmark in which data agents produce verifiable tabular results from task-local heterogeneous workspaces. It contains 410 cross-language tasks and 7,439 artifacts totali…