2026年10月26日 技术热点总结

项目简介：GPU tensor unlock source for RTX 4090 and experimental RTX 4070 Ti SUPER; firmware generated locally

技术栈：C

核心介绍：New to this project? Start with the copy-paste Quick Start. This Linux kernel module enables the SM speed-select override before the stock NVIDIA driver starts. RTX 4090 (10de:2684, AD102) support has demonstrated approximately 2× FP16 and BF16 tensor throughput with FP32 accumulation: The override survived PCIe Function Level Reset (FLR), stock NVIDIA 615.78.08 initialization and CUDA benchmarks. INT8, dp4a and scalar FP32 stayed near their

项目数据：⭐ 2,57…

项目简介：11 Claude skills for day trading: scan thousands of charts, track breaking news, time every entry and exit, and replay every outcome before you risk a dollar.

技术栈：Python、backtesting、claude、claude-skills、day-trading、stock-market

核心介绍：11 Claude skills for day trading: scan thousands of charts, track breaking news, time every entry and exit, and replay every outcome before you risk a dollar. Free, MIT, no signup, no API key. The tools are plain Python that read free public data: Yahoo Finance charts, Nasdaq Trader’s symbol lists, SEC EDGAR One scans thousands of charts for s…

项目简介：Import .sql files into TypeScript with full type inference and zero codegen.

技术栈：Rust

核心介绍：Import .sql files into your TypeScript code with full type inference Read the docs → If you have a file like this: select u.id, u.name, count(p.id)::int as post_count left join posts p on p.user_id = u.id where u.id = :id::int You can import it directly from a TypeScript file like this: import postgres from “postgres”; import getAuthor from “./get_author.sql”; const db = postgres(process.env.DATABASE_URL); const author = await getAuthor.run(db, { id: 1 });

项目数据：⭐ 1,263 Sta…

项目简介：Proof-of-Concept (PoC) exploit for php2shell (CVE-2026-109618 / CVE-2026-109619), demonstrating remote code execution in vulnerable PHP environments. For security research and educational purposes only.

技术栈：Shell

核心介绍：> Critical Advisory (CVSS v4.0: 9.2 – Critical)： On October 20, details of two pre-auth Remote Code Execution (RCE) flaws in the PHP core leaked online ahead of coordinated disclosure. Exploitation allows arbitrary command execution straight from the web server. php2shell refers to two critical vulnerabilities (CVE-2026-109618 & CVE-2026-109619) affectin…

项目简介：Ely Office: offline-first desktop apps in Rust and GPUI

技术栈：Rust

核心介绍：Offline-first desktop apps in Rust and GPUI. Files stay in your folders; no account, no cloud service. Download from office.elygpui.com or the latest release: Windows, macOS (Apple silicon and Intel) and Linux (AppImage, deb, rpm, Arch, Nix; Debian 11 and Ubuntu 20.04 or newer). ElyMail signs in to Microsoft with OAuth, and to Gmail and iCloud with an app password. scripts/doctor.sh # toolchain versions scripts/fetch-references.sh # upstream sources at pinned commits

项目数据：⭐ 99…

Robots must act under physical and social constraints while adapting to changes in their environment and human needs. Existing embodied-agent systems often focus on specified goals or abstract execution into symbolic actions, limiting evaluation in an evolving world. We introduce TARSphere, a robot-oriented full-stack framework for physically and socially grounded long-horizon embodied agents. TARSphere-Infra runs robot policies in a physics simulator with embodied NPCs and constrained dialogue. An event-driven finite-state machine (FSM) uses completed actions and delivered information to u…

Split language models keep input text on a client but send intermediate representations to a server. Attacks can reconstruct text from these messages. Comparing privacy evaluations is difficult when attackers see different messages, know different model weights or use different datasets and recovery measures. We present LLM-Attacker, a framework that records these choices and lets researchers check, compare and extend an evaluation. Compatible attacks reuse recorded messages, and their predictions are saved for independent scoring. The scoring rule requires a linked undefended reference bef…

Recursive self-improvement (RSI) relies on evaluation feedback to assess progress and guide further research, yet repeatedly running complex benchmarks is costly and slows iteration. Human experts reduce this cost by selecting benchmark subsets or designing compact suites. We ask whether AI agents can automate this design process and introduce EvalResearchBench (ERB), a benchmark for autonomous evaluation research. Given target materials, development references, candidate APIs, and fixed time and API budgets, an agent called the researcher selects or synthesizes tasks, implements graders, a…

Diffusion language models can write several tokens in one forward pass, and they decide during generation how many passes to spend. Two outputs of the same token length can therefore vary a lot in cost. To understand the underlying factors of this cost variation, we analyze two diffusion models across 10 tasks. We find that the number of forward passes per token depends heavily on the task. Text copied or lightly edited from the input is cheap, while open-ended writing is costly. In DiffusionGemma, generating a token for story writing requires more than three times as many forward passes as…

Diffusion Language Models (DLMs) introduce unique quantization challenges due to their iterative denoising process. In this work, we reveal that quantization sensitivity is both taskand time-dependent, with different downstream tasks exhibiting distinct precision requirements across denoising steps. We further show that these patterns can be predicted without groundtruth labels or task-specific evaluation metrics. Building on these findings, we introduce T 2Quant, a task and time-aware mixedprecision quantization framework that uses a label-free sensitivity signal to optimally allocate acti…

Effective test-time personalization requires LLM agents to adapt to query-dependent user preferences from lightweight interaction. We study persona selection with binary comparisons between candidate responses as the online signal. We present ID-TAP, a scalable, IDS-inspired neural dueling-bandit framework for large discrete persona spaces. ID-TAP combines a query-conditioned neural utility model with a lightweight Bradley-Terry explorer, and transfers comparison evidence across related queries through similarity-weighted warm starts. The neural model shares statistical strength across quer…