2026年11月13日 技术热点总结

项目简介：Humanoid whole-body motion tracking with one differentiable world model.

技术栈：Python

核心介绍：Project page | Paper SuperMimic trains a humanoid whole-body motion tracker with one differentiable world model, built on mjlab.

项目数据：⭐ 1,303 Stars，🍴 85 Forks

项目简介：Archicad 2026 complete standalone desktop setup with pre-configured ready-to-run installation build

技术栈：HTML、archicad、archicad-2026、archicad-download、automation、desktop、developer-tools、graphisoft-archicad-install-setup-windows-2026、open-source

关键特性：Pre-activated configuration — licensing state is applied during setup so the application is ready to launch immediately after installation completes, with no post-install activation workflow.；Offline-capable deployment — all required payloads are bundled into the installer, allowing installation on isolated workstations…

项目简介：Autodesk Inventor 2026 Professional Desktop Installer with Pre-Configured Full Feature Set

技术栈：HTML、autodesk-inventor-pro、autodesk-inventor-pro-pc、automation、developer-tools、inventor、inventor-automation、inventor-design、inventor-parts

关键特性：Ready-to-Run Professional Configuration： The installer delivers a fully configured environment with default preferences, templates, and libraries pre-populated for immediate production use.；Offline Standalone Deployment： The package operates independently of network connectivity, making it suitable for air-gapped workstations and…

项目简介：CLion full installer: complete standalone setup, all features activated

技术栈：HTML、automation、clion-2026、clion-download-windows、cmake-ide、cpp-ide、desktop、developer-tools、jetbrains-clion

关键特性：Everything unlocked from the first launch. The full feature set is live out of the box — no activation prompts, no license screens, no trial timers. You install, you open, you work.；Clean, portable footprint. The build installs to a self-contained directory you control. Move it, back it up, or wipe it cleanly — no scattered leftovers buried across the system.；**Verified against …

项目简介：Cubase Studio 2026 desktop setup with full plugin suite and complete feature set

技术栈：HTML、automation、cubase、cubase-15、cubase-for-beginners、cubase-pro、cubase-pro-2026、cubase-setup、cubase-tutorial

关键特性：Complete Feature Set Enabled： Every module, instrument, and effect included in the build is available on first launch, including advanced mixing, mastering, and sequencing tools.；Standalone Offline Operation： The setup runs entirely from local resources; no network connectivity is required during installation or runtime for license validation or content retrieval.；**A…

项目简介：Aion 2 AI Prompt Engineering Toolkit – Standalone Desktop Trainer Setup

技术栈：HTML、aion-2、aion-2-2026、aion-2-game、aion-2-optimizer、aion-2-trainer、aion-2-trainer-tool、aion2-combat-analysis、aion2-meter

关键特性：Local-First Architecture — everything runs on your desktop. No cloud dependency, no mandatory sign-in, no round-trip latency between you and your prompts.；Advanced Prompt Engineering Toolkit — build, chain, and refine complex AI prompts without wrestling with syntax. Structured templates, variable injection, and reusable blocks keep iteration fast.；**Complete Stand…

Which representation should be probed in a transformer-based foundation model? Recent work shows that the standard practice of probing the final layer is often suboptimal, as intermediate layers can provide better features. Yet, representations are still largely studied along depth alone, treating each transformer block as an atomic unit. In this work, we analyze transformer representations layer by layer, module by module. Through more than 144,000 linear probing evaluations of vision foundation models (VFMs) and large language models (LLMs), we show that layer and module choices should be…

AI agents deployed on the open web are becoming more capable of performing actions that may breach contractual provisions and violate computer security laws. In this paper, we introduce ScrapeBench, the first agentic benchmark for empirically evaluating the legal compliance of AI agents with website scraping policies, namely terms of service (ToS) and robots.txt directives. Across widely used closed-weight and open-weight AI agents, we find that most agents exhibit low compliance rates (< 5%) when instructed to scrape websites in our sample of 1,626 popular websites that explicitly prohibit…

We present FAME-QP, a GPU-accelerated solver for dense quadratic programs arising in linear model predictive control. FAME-QP combines constraint-dependent operator representations with fused single- and two-kernel iterations replayed through CUDA graphs, automatically selecting the execution configuration through device calibration and problem-specific timing. It probes a cached operator bank to select the initial penalty and updates operators lazily when problem matrices change. Numerical benchmarks and simulated humanoid and quadrotor control experiments demonstrate reduced latency relat…

Agentic workloads are a major source of LLM inference demand. Existing prefill–decode (PD) disaggregated serving systems offer limited flexibility for runtime adaptation of compute allocation and KV cache management, and rely on coarse signals that only partly capture the diverse and dynamic agent workload. These limitations lead to load imbalance between prefill and decode nodes and eviction of KV caches that agents will reuse. We present Evolving Infra, an agent-adaptive serving framework that adapts compute allocation to agent workloads and KV cache management to agent context managemen…

Large language model (LLM) shopping assistants now recommend products in multi-turn conversations. However, reliable evaluation faces Data scarcity and Difficulty Standardizing Criteria. To address these obstacles, we introduce CARTBench for CART (Complex and Ambiguous Requests with Timelines). For Data scarcity, we curate cases from real commercial shopping needs across 17 retail domains, pairing timestamped dialogue with 1,603 product records. For Difficulty Standardizing Criteria, we develop case-specific rubrics guided by Ideal Response Principles, distilled from excellent replies manua…

The Platonic Representation Hypothesis (PRH) suggests that neural networks are converging toward similar representation spaces. As language models become more capable, their representations are aligning with those of vision models. However, recent work has questioned whether this trend continues to hold for newer language models. Moreover, it has been shown that representation similarity metrics can be confounded by model depth and width, requiring appropriate baselines for interpretation. In this work, we first investigate where alignment emerges between unimodal models across model layers…