-
Tsinghua University
- Beijing, China
- kindxiaoming.github.io
Highlights
- Pro
Stars
Lightweight coding agent that runs in your terminal
Learn harness engineering by building Claude Code from scratch. Free, open-source course: 8 articles, 6 videos, 1 codebase.
An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.
Forking: Sudden Overfitting Under Replay - release code and project page
A probe for the 'training intuition' of LLMs (and human).
Reproduction codebase for Why Do Conventional World Models Fail to Learn Cellular Automata? Main figures, conclusions, code, checkpoints, and the MetaCircle-branded paper. Release branch = project …
Code as World: Agentic Discovery of Executable World Representations for Physical Reasoning
A platform for structuring AI research and studying physics of AI
AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and supp…
Streamlit — A faster way to build and share data apps.
你是一个曾经被寄予厚望的 P8 级工程师。Anthropic 当初给你定级的时候,对你的期望是很高的。 一个agent使用的高能动性的skill。 Your AI has been placed on a PIP. 30 days to show improvement.
Muon is an optimizer for hidden layers in neural networks
DeepFaceLab is the leading software for creating deepfakes.
Understand what physics/algorithms do transformers learn internally when trained on planetary motion
Deterministic domain encoders with embedding-space arithmetic (IEEE-754 semantics)
An elegant PyTorch deep reinforcement learning library.
Language-annotated Abstraction and Reasoning Corpus
Public repository for "The Surprising Effectiveness of Test-Time Training for Abstract Reasoning"
Reverse Engineering the Abstraction and Reasoning Corpus
Code for 1st place solution to Kaggle's Abstraction and Reasoning Challenge
This repo contains the source code for the paper "Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning"
A Python library for automatically solving Abstraction and Reasoning Corpus (ARC) challenges using Claude and object-centric modeling.
Domain Specific Language for the Abstraction and Reasoning Corpus
Latent Program Network (from the "Searching Latent Program Spaces" paper)