2026年11月9日 技术热点总结

项目简介：Films made with code. Short films, long YouTube videos, ads, explainers and music videos: an AI coding agent writes the script, builds every frame in code, designs the sound and hands you an MP4.

技术栈：Python

核心介绍：Kodla yazılmış filmler. Kısa filmler, uzun YouTube videoları, reklamlar, açıklayıcılar ve müzik videoları. Fikri ve zevki sen getirirsin; bir yapay zekâ kodlama ajanı senaryoyu yazar, her kareyi kodla kurar, sesi tasarlar ve sana bitmiş bir MP4 verir. English · 简体中文 · 日本語 · Kurulum · Hızlı başlangıç · SSS cinecode ile yapılmış filmlerden 30 saniye. Her kare ba…

技术栈：C

核心介绍：version 1.1.1) into a native x86-64 application for Windows, with a Vulkan renderer (docs/architecture.md). > No game files are included. You need your own legally obtained copy of the game, dumped from your own console. > Instructions for obtaining those files will not be provided. > The game is built from your files on your machine, so the result contains the game. It is for your own use. Do not share it.

关键特性：Native code, not emulation；60 fps by default, or a limit of 30 to 240 fps, the display’s refresh rate or unlimited；Native widescreen and ultrawide (the …

项目简介：《搞机工具箱》Web移植版

技术栈：HTML

项目数据：⭐ 1,212 Stars，🍴 104 Forks

项目简介：A small, fast container runtime for macOS, Windows and Linux

技术栈：Rust

核心介绍：A tiny, fast container runtime for macOS, Windows and Linux. Runs your Docker images and works with the docker CLI, Compose and buildx you already use. Download · Features · Build from source · Releases Kiri is a lightweight replacement for Docker Desktop and OrbStack, written in Rust. It runs containers in one small Linux VM (or straight on the kernel on Linux) and speaks the Docker Engine API, so your tools just work. The name comes from the Igbo word *ntakịrị*, “small”.

项目数据：⭐ 1,190 Star…

项目简介：Emulate macOS on iOS and iPadOS

技术栈：C

核心介绍：Henge macOS on your iPhone and iPad. Install How it works Website Releases Download macOS, install it on the device, and use it.

项目数据：⭐ 1,117 Stars，🍴 84 Forks

Large language model (LLM) agents typically append raw tool outputs to their context. Over multi-step interactions, long and noisy observations incur substantial context overhead and dilute the agent's attention to task-relevant information. Existing token pruning and summarization methods remain constrained by discrete text representations, limiting the achievable compression ratios.
In this work, we challenge the prevailing assumption that agents must consume tool observations as explicit text tokens. To this end, we propose Coca-Cola, a continuous latent context compression model for L…

Offline retrieval-augmented generation (RAG) on edge devices is updated through signed knowledge packs that carry text and precomputed embedding vectors. A signature proves who published a pack, not that its vectors encode its text. We study this text-vector integrity gap. A verifier that recomputes vectors and accepts them within a cosine tolerance concedes a predictable share of queries to an attacker who controls the vectors: the hijack rate equals the distribution of angular retrieval margins at the tolerance (the margin law). Margins are small in five corpora and two embedding models, …

A deployable text-to-image diffusion model must hold several capabilities at once, such as instruction following, text rendering and human preference, and must sample in a few steps. Capabilities and few-step sampling are built in separate stages: a full-step model first consolidates capabilities through multi-teacher on-policy distillation, then serves as a fixed teacher for step distillation. Every full-step model we evaluate scores lower once compressed, and the second stage cannot begin until the first finishes. We introduce StepX-OPD, which merges the two stages: one student acquires e…

LLM-guided evolutionary search uses AI agents to improve programs for mathematical and computational problems by iteratively refining solutions and combining high-scoring methods. However, the process is often slow because the LLM generates a long textual chain of thought for each round and must repeat this generation many times to maintain population diversity. We present SwarmEvolve, a framework in which hundreds of low-cost agents per round work in swarms, reason in latent space without decoding to text, and share information by passing KV caches. Each swarm studies the problem from a di…

Large language models (LLMs) have shown increasing promise in research-level problem solving, yet a comprehensive collection of such problems is still missing, and evaluating them on research-level problems remains challenging. We introduce OpenFrontiers, a cross-disciplinary resource that combines a structured corpus of open research problems with a retrospective benchmark of recently resolved problems in mathematics. The corpus records problem statements, context, progress, and other information for 12,151 problems. The benchmark contains 200 problems, 110 resolved in 2026 and 90 still op…

Mathematical modeling turns an open-ended real-world problem into assumptions, a model, a solution, and a defended recommendation, and its answers matter only if they are both close to a credible reference and physically feasible; existing benchmarks either score solver fragments or grade reports with an LLM judge, leaving those numbers unchecked. We introduce TB-MathModel (the Terminal-style Mathematical Modeling Bench), the first benchmark to make complete contest-style modeling problems agentic, verifiable, and terminal at once, over 18 real CUMCM and MCM problems (2023–2025). Each task…