2026年10月30日 技术热点总结

项目简介：Free Claude skill: make Claude build award-level, cinematic landing pages. By MotionSites.

技术栈：Python

核心介绍：AI-built websites tend to look the same for three reasons: a vague plan, no real media, and nobody looking at the result. This skill fixes all three. Claude studies the most-liked designs from MotionSites, picks real cinematic media, writes a precise build spec, builds the page, then screenshots its own work and fixes it until it clears a gallery-level bar. 8 of the 40 example designs inside the skill, each shown next to the video it’s built on. All 50+ videos ar…

项目简介：Free interactive 3D science lessons for high school · 铜匠实验室 · https://lab.tongjiang.app

技术栈：HTML、3d、biology、chemistry、education、geography、high-school、interactive、open-educational-resources

核心介绍：English | 简体中文 Try them online: https://lab.tongjiang.app There are 54 lessons so far. Each one is a single self-contained HTML file of about 1 MB: double-click to open it, and it works offline, on a projector or on an interactive whiteboard. Every lesson is one 3D world with a live control console, a guiding question and a step-by-step path a teacher can follow from start …

项目简介：Say cheese! Star this repo and your profile picture gets printed out on a receipt printer.

核心介绍：Star this repo to have your GitHub profile picture framed, printed out on my receipt printer, and stuck to my wall. > If you give a one-time sponsorship of any amount beforehand, your profile picture will be printed in a special frame. GitHub allows you to create webhooks for specific events on your repos, I’ve set one up for starring this project.

项目数据：⭐ 904 Stars，🍴 1 Forks

Visual generation requires both high-level semantic knowledge and fine-grained pixel details. We propose a learning paradigm that assigns these roles to autoregressive pretraining for knowledge acquisition and diffusion post-training for high-fidelity rendering. Within this framework, we study the objective, representation, and scaling of knowledge acquisition and rendering. Controlled comparisons identify autoregression as the stronger pretraining objective for downstream rendering, improving both learning efficiency and final generation quality. We further find that AR-only generation qua…

The quality with which language models follow instructions falls as the load grows, yet there is still no quantity that expresses the limit of a model in units of load with an uncertainty. We propose an instrument in which the load consists of rules that replace single-token words in a text without meaning: such an instruction is atomic, the rules are independent by the construction of the task, and the outcome of every application of a rule is determined by a computable reference answer. Every occurrence of a word to be replaced becomes a Bernoulli trial with a known position in the answer…

As frontier language models grow increasingly capable and autonomous, independent third-party evaluations are critical for characterizing their systemic safety risks. While closed-source releases are routinely accompanied by comprehensive system cards, open-weight models receive less scrutiny regarding frontier risks. To address this gap, we present an independent safety evaluation of DeepSeek V4.1 Flash-1017, the latest version of V4.1 Flash that departs from V4 with a new architecture, scaled-up training, and strong agentic capabilities. Drawing on the EU General-Purpose AI Code of Practi…

Despite their extensive pretraining knowledge and strong reasoning capabilities, large language models (LLMs) struggle to adaptively gather information in experimental design tasks. We study how to amortize sequential Bayesian experimental design (BED) into an LLM by explicitly training the reasoning that guides action selection using information-gain rewards. Our approach specifies the experimental generative model independently of the LLM's own beliefs, allowing us to optimize expected information gain under a model of our choosing. We investigate the design choices for amortization, incl…

Command-line interface (CLI) agents increasingly execute longer-horizon tasks, yet existing benchmarks either assume that user instructions remain fixed throughout execution or evaluate task outcomes without capturing runtime collaboration quality. In this paper, we present REALCLI, a paired benchmark that unifies the evaluation of task execution and human-agent collaboration quality on realistic CLI tasks. To comprehensively measure collaboration, REALCLI builds a five-dimensional framework, operationalized into 22 sub-dimensions and 51 interaction scenario templates. REALCLI provides two …

Reinforcement learning (RL) is an effective way to strengthen the long-context reasoning of large language models (LLMs), but it needs training data that are long, diverse, challenging, and verifiable. Human annotation of such data is expensive, naturally occurring data are too scarce to sustain RL, and existing synthesis methods depend on external corpora or stronger teacher models, which become harder to obtain as the trained models get stronger. Recently, RL from zero data, in which a model learns from tasks it proposes itself, has achieved excellent performance, but existing methods tar…