Shane

生产级轻量级 LLM 训练-推理-对齐全栈框架

基于强化学习的 LLM Agent 长链路决策优化系统

基于 DeepSeek-R1-Distill-Qwen-1.5B 的金融领域推理增强问答系统

A finance question-answering system that runs a compact AI model on a single consumer GPU, showing its step-by-step reasoning alongside answers and supporting OpenAI-compatible API calls.

支持多 Agent 编排、流式透明与安全沙箱的 AI Agent 框架。

Multi-Agent Stock System

Hyper-Relation Fusion for Solving Multi-depot Vehicle Routing Problems

Data classification prediction based on the Grey Wolf Optimization Algorithm GWO_LSSVM-Adaboost (Mathematical Modeling Competition Code)

Grey Wolf Algorithm Optimization Support Vector Machine Regression Prediction Based on KPCA Kernel Principal Component Analysis (Model Code for Mathematical Modeling Competition)

Multi-objective path optimization system based on ACO and A* algorithm（Business）

Optimizing the multi-feature input classification prediction model of convolutional neural network-support vector machine based on sparrow algorithm (Mathematical Modeling Contest code)

No description.

Python Programming Assignment (Spring 2023)

Crawling Assignment at the University of Hong Kong (Autumn 2024) (Business)

Gobang based on deep reinforcement learning (Autumn 2024)