My personal website

Preparing for ML Interviews.

ToolOrchestra is an end-to-end RL training framework for orchestrating tools and agentic workflows.

ArcherCodeR is an open-source initiative enhancing code reasoning in large language models through scalable, rule-governed reinforcement learning.

The Entropy Mechanism of Reinforcement Learning for Large Language Model Reasoning.

Hybrid Latent Reasoning via Reinforcement Learning

Official implementation of the paper "Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space"

Model merging is a highly efficient approach for long-to-short reasoning.

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

Unified Efficient Fine-Tuning of 100+ LLMs (ACL 2024)

Material for cuda-mode lectures

Code for Adam-mini: Use Fewer Learning Rates To Gain More https://arxiv.org/abs/2406.16793

The official implementation of MARS: Unleashing the Power of Variance Reduction for Training Large Models

GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection

Fira: Can We Achieve Full-rank Training of LLMs Under Low-rank Constraint?

Leveraging LLMs to design and optimize nanophotonics

The implementation of AICircuit: A Multi-Level Dataset and Benchmark for AI-Driven Analog Integrated Circuit Design

Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients.

Code repo for the paper "SpinQuant LLM quantization with learned rotations"

This repository contains code for the MicroAdam paper.