Xie Tian

Logic-RL. Reproduce R1 Zero on Logic Puzzle

Spectral-Sphere-Optimizer. Spectral Sphere Optimizer

AI_Game_KingGlory. 王者荣耀AI(在腾讯开悟平台实现) RL algorithm for the game: King Glory (based on kaiwu platform)

VFL_with_LDP. 中科大密码学大作业：在breast cancer数据集上实现了纵向联邦学习与标签差分隐私结合

Efficient_AI. 此项目是我个人对MIT 6.5940 课程作业的答案，学习笔记和心得。

Cyber_Su. 赛博苏神：你的 research 好老师

Bike-REID. 在监控画质下实现对校园自行车的重识别，包含REID模型识别，向量数据库检索，UI展示

Open-SPCT. Community Implementation of paper "Inference-Time Scaling for Generalist Reward Modeling"

spectral-ball-tools. Python

Physics-of-Deepseek-R1. Break down deepseek r1 zero training

DL_intorduction. Python

Wait-For-Attention. Python

speedrun_mup. mup scaling for speedrun

Video-CLIP. Adapting CLIP for Cross Model Video Understanding

Kaggle-Sleep-Detection. 利用腕带传感器加速度数据预测入睡苏醒时间

Data_Privacy. 差分隐私随机梯度下降(DP-SGD)与纵向联邦学习(Vertical Federated Learning )

ControlNet-Lighter. ControlNet with Fewer Parameters, but Comparable Performance.

torchtitan. A PyTorch native platform for training generative AI models

modded-nanogpt. NanoGPT (124M) in 3 minutes

Scaling-Law-from-Scratch. 从头开始，探讨LLM pretrain & post-train阶段的 scaling law与数据配比

Chain-of-Reasoning. Python

InfiniteLogicForge.

LLM-Math-Eval. Python

llm-action. 本项目旨在分享大模型相关技术原理以及实战经验。

TinyLM-from-Scratch. Python

Awesome-Scaling-Law. Papers Collections for Scaling Law

Retrieval-Recs-of-Douban-Data. 豆瓣数据的个性化检索与推荐(2024web作业1)

learn_differential_privacy. Here's some notes taken by a sophmore student,mainly based on Guatum's CS680 and Dwork's wonderful book.I try to

Manifold_Arena. spectral ball or steifel manifold

Watermark-Optimizer. crazy exp

mHC. Python