AI - Artificial Intelligence

• MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement 65 upvotes

  Reinforcement learning (RL) is the central training paradigm for advancing large foundation models towards self-improvement

• Multi-Agent Egocentric World Model with Fine-Grained Embodied Interaction 47 upvotes

  Egocentric world models predict first-person observations conditioned on an agent's actions, but most focus on a single agent

• Foundations of Large Language Models 38 upvotes

  This is a book about large language models

• OuroWorld: Bringing Any 3D World Alive as Diverse, Endlessly Looping 3D Cinemagraphs 37 upvotes

  Recent 3D world models generate photorealistic, explorable scenes that remain frozen in time

• Beyond Spatio-Temporal Priors: A Generalizable Approach for Dense Correspondence Matching 34 upvotes

  Dense correspondence matching has historically been bounded by simplifying spatio-temporal priors, such as smooth motion and rigid geometry

• U-Space: Uncovering When and Why Uncertainty Arises in Language Models 33 upvotes

  Large language models are informing decisions with ever-higher stakes

• MC-Sparse: Deconstructing and Closing the Dense-Sparse Attention Gap in Diffusion Transformers 33 upvotes

  Sparse attention is a primary approach to reducing the latency of diffusion transformers in long-sequence generation tasks, such as video and high-resolution 3D asset generation

• TestPrism: Rethinking Test Evaluation Beyond a Single Reference 30 upvotes

  Large language model (LLM) coding agents have advanced test generation across diverse programming tasks

• Memento 3: Model-Based Recursive Self-Improvement through Reflective Rulebooks 28 upvotes

  Learning to act in unfamiliar environments requires agents to infer how the world works and revise that understanding as new evidence arrives

• Post-Training Frontier Text-to-Image Models by Composing Preference and Rubric Rewards 26 upvotes

  Recent text-to-image generation models have achieved remarkable visual quality, but improving them through post-training remains challenging because no single reward signal captures the full range of human preference

• guillaumemeyer/watermarks-remover A privacy-first app that strips AI watermarks from content you own. Python 23.8k stars
• hypit-ai/hypit Clone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-roll, ship 100 variants in one TypeScript 20.8k stars
• yc-software/qm Multiplayer agent harness for work. TypeScript 15.4k stars
• genspark-ai/genoffice Free, open-source AI Office suite: Docs, Sheets, Slides, PDF, Markdown and HTML editors with a built-in AI agent, plus a `genoffice` CLI and TypeScript 9.2k stars
• KKKKhazix/AIHOT 一个自己找热点、自己写日报的网站框架。把信源和精选标准换成你的，它就是你的行业热点站。 TypeScript 7.1k stars
• yi1108/printfilm PRINTFILM: AI short-video marketing and AI short-drama creation platform Python 5.1k stars
• Jakubantalik/thinking-orbs Dotted thought-orb loading indicators for AI & agent UIs, 9 tuned types, two sizes, auto dark/light TypeScript 3.6k stars
• fuxicodex/Fuxi FuXi is a fast, self-contained AI coding agent that lives in your terminal — edit code, run commands, and drive tools, with cost-aware routi Python 3.4k stars
• ShawnPana/phone-harness let your agent control your phone Python 3.2k stars
• shadcn-ui/lint An agent-first linter for Tailwind design systems. Write design system rules that agents can verify. TypeScript 3.2k stars

Running models on hardware you control: local LLM and agent tooling reads land in Learn under the AI topic.