2026年10月31日 技术热点总结

项目简介：dsh-go — DeepSeek Harness 装进口袋：手机本机运行，也能遥控你的电脑。

技术栈：Dart、ai-agent、android、android-app、coding-agent、deepseek、deepseek-harness、dsh、dsh-plugin

核心介绍：简体中文 · English DeepSeek Harness 的 Android 客户端。手机上直接跑，也能遥控你电脑上的 DSH。 Deepy 和鲸娘，陪你干活 干活时敲代码，做完来个空翻，等你确认会举手。打开「App 外显示」，它们还会跟你到桌面。

项目数据：⭐ 2,448 Stars，🍴 102 Forks

项目简介：Static recompilation of The Legend of Zelda: Tears of the Kingdom to native Windows x86-64, with a Vulkan renderer. Game files not included.

技术栈：C

核心介绍：> No game files are included. You need your own legally obtained copy of *Tears of the Kingdom*, dumped from your own console: the base game and the 1.2.1 or 1.4.3 update. 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. Early development. The game is playable on Windows; visual bug…

项目简介：🛡️ BLA (Blind LLM Agents) — Adversarial defense shield that protects binaries, Electron apps, JS bundles, Python, and WebAssembly against AI reverse-engineering agents like REA. Poison -> Exhaust -> Hallucinate.

技术栈：TypeScript、adversarial-ai、claude-code、cursor、electron-security、llm-security、mcp、mcp-server、obfuscation

核心介绍：Quick start · Empirical results · Format coverage · Benchmark · FAQ > BLA is a defensive tool for protecting your own software. See DISCLAIMER.md. > Get BLA from qybaihe/bla or npm bla-agents.

项目数据：⭐ 1,393 Stars，🍴 189 Forks

项目简介：Mika — 你手机上的私人 AI 助理，替你操作 App、上网、把事办成。An on-device AI agent for Android that operates apps and the web to get things done for you.

技术栈：Kotlin、accessibility、ai-agent、android、automation、jetpack-compose、kotlin、llm、mobile-agent

核心介绍：简体中文 | English 它在一块看不见的后台屏上干活，你照样刷你的手机。 An on-device AI agent for Android that operates apps and the web to get things done for you — on a hidden background screen, so you can keep using your phone. 下载 APK · 安装 · 快速开始 · 能做什么 · 常见问题 · 参与贡献

项目数据：⭐ 1,239 Stars，🍴 203 Forks

项目简介：A Claude skill that turns vibecoded, AI-looking UI into a designed interface

技术栈：JavaScript

核心介绍：Created by Damir Sokolovsky (@heydamir), who started the Unslopping series on X.

项目数据：⭐ 1,208 Stars，🍴 56 Forks

Quarterly earnings forecasting requires an agent to connect financial evidence to the income-statement line items that produce a company's reported earnings. We introduce EPSBench, a benchmark for studying this ability across 149 firm-quarters from 86 US companies. Agents read a fixed collection of filings, earnings-call transcripts, and news, then estimate company-specific income-statement drivers with cited evidence. A shared accounting calculator converts these estimates into GAAP diluted earnings per share (EPS), providing a common basis for comparing forecast assumptions. We evaluate s…

Self-distillation offers a way to improve reasoning models when a suitable external teacher is unavailable. However, common approaches rely on privileged information to construct a stronger self-teacher, limiting their use when only unlabeled prompts are available. In this paper, we introduce Weakening-Induced On-Policy Self-Distillation (WISD), which instead constructs a weaker counterpart of the original model and uses the contrast between the two as a training signal without requiring privileged information. We explore two ways to construct the weaker model: low-bit quantization and magn…

Autoregressive (AR) image generators are powerful in principle but inflexible in practice: training with a fixed raster-scan order hardcodes a single factorization of the image distribution into the model, leaving it unable to inpaint, outpaint, or edit at inference time. Existing solutions, including the modern discrete-diffusion family, train from scratch on a mixture of token orderings, paying roughly 3x the compute of standard raster-scan training; discrete diffusion further sacrifices compatibility with the AR inference ecosystem (KV caching, vLLM) by construction. We argue that this c…

Language agents increasingly act through external tools and interactive environments, making the interface between the model and its executable capabilities part of the decision process. Model Context Protocol (MCP) and command-line interfaces (CLI) can expose the same semantic operations through markedly different model-facing representations. We study whether these representational differences change agent behavior when executable capabilities are matched. We introduce Matched Interface Comparison and Analysis (MICA), which renders MCP and CLI from a shared semantic action contract and ro…

Developers download quantized large language models (LLMs) from public registries to run them locally. None of the registries we examined documented functional testing of their files. We tested all 305 eligible code-capable GGUF tags (the llama.cpp format) across 15 model lines in the official Ollama library, plus 22 popular Hugging Face community files. A 15-task code screen identified suspects, which we retested on all 164 HumanEval+ tasks. Seven official tags (2.3%) solved at most six of them. All were low-bit quantizations of models with at most four billion parameters. We rebuilt five …

Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning algorithms. We show that this algorithmic separation is not fundamental. By deriving suitable RL objectives as maximum-likelihood fitting to model-generated target distributions, we place all of these stages under a single likelihood-based training algorithm, unified training. This perspective turns the foundation-model training lifecycle into a problem of choosing which data to train on, in what proportions, and w…