My learning notes/codes for ML SYS.

veRL: Volcano Engine Reinforcement Learning for LLM

Open Thoughts: Fully Open Data Curation for Thinking Models

Clean, minimal, accessible reproduction of DeepSeek R1-Zero

Finetune Llama 3.3, DeepSeek-R1 & Reasoning LLMs 2x faster with 70% less memory

Lightweight tool to identify Data Contamination in LLMs evaluation

Large Reasoning Models

Training and Benchmarking LLMs for Code Preference.

Pytorch implementation of Tree Preference Optimization (TPO)

Clustering results demo

[SafeGenAi @ NeurIPS 2024] Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates

Implementation of the MATRIX framework (ICML 2024)

Official code for the paper: WALL-E: World Alignment by NeuroSymbolic Learning improves World Model-based LLM Agents

Arena-Hard-Auto: An automatic LLM benchmark.

Command-line YAML, XML, TOML processor - jq wrapper for YAML/XML/TOML documents

Secrets of RLHF in Large Language Models Part I: PPO

PaL: Program-Aided Language Models (ICML 2023)

SGLang is a fast serving framework for large language models and vision language models.

DataComp for Language Models

Official repo for the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas"

Scrape from Twitter using Nitter instances

A library with extensible implementations of DPO, KTO, PPO, ORPO, and other human-aware loss functions (HALOs).

It is said that, Ilya Sutskever gave John Carmack this reading list of ~ 30 research papers on deep learning.

Together Mixture-Of-Agents (MoA) – 65.1% on AlpacaEval with OSS models