Jiawei Liu
ganler
phd'ing @Illinois
Languages
Top Repositories
Reproducing R1 for Code with Reliable Rewards
🚀A header only thread pool library implemented with modern C++. Fast and easy to use.
Some CS notes during Jiawei's undergrad.
General system research material (not limited to paper) reading notes.
Course project of CG(CS100433, 2019 fall, Tongji Univ.): A real-time billiard simulator using both ray tracing and graphics pipeline.
Repositories
73No description provided.
Reproducing R1 for Code with Reliable Rewards
A benchmark for LLMs on complicated tasks in the terminal
No description provided.
🚀A header only thread pool library implemented with modern C++. Fast and easy to use.
verl: Volcano Engine Reinforcement Learning for LLMs
Some CS notes during Jiawei's undergrad.
Course project of CG(CS100433, 2019 fall, Tongji Univ.): A real-time billiard simulator using both ray tracing and graphics pipeline.
Artifact for ESEC/FSE'23 paper "NeuRI: Diversifying DNN Generation via Inductive Rule Inference"
Go ahead and axolotl questions
Enhanced implementation of Kint in LLVM-14, in C++17.
General system research material (not limited to paper) reading notes.
“连续六年成为全世界最受喜爱的语言，无GC也无需手动内存管理、极高的性能和安全性、过程/OO/函数式编程、优秀的包管理、JS未来基石" — 工作之余的第二语言来试试Rust吧。<<Rust语言圣经>>拥有全面且深入的讲解、生动贴切的示例、德芙般丝滑的内容，甚至还有JS程序员关注的WASM和Deno等专题。这可能是目前最用心的Rust中文学习教程/书籍
https://nnsmith-asplos.rtfd.io Artifact of "NNSmith: Generating Diverse and Valid Test Cases for Deep Learning Compilers" ASPLOS'23
A fast and scalable general purpose sandbox code execution engine.
Ganler's vLLM utilities.
SGLang is a fast serving framework for large language models and vision language models.
This repo is a pytorch implementation of 2 stream convolutional neural networks, which can be used in detached single channel frames series.
A framework for the evaluation of autoregressive code generation language models.
Inference code for CodeLlama models
CrossCodeEval: A Diverse and Multilingual Benchmark for Cross-File Code Completion
A telegram bot that sends you a message when the GPU is in use
Python package designed for single-cell calling cards data
Rust for C++ programmers
Collect simple coverage information in memory.
No description provided.
FuzzBench - Fuzzer benchmarking as a service.
Image Filter: project for 2020 Google ML winter camp.
OSS-Fuzz - continuous fuzzing for open source software.
The Lepton AI SDK to make building AI applications simple.