Reinforcement Learning via Self-Distillation (SDPO)

This is the official github repor for our paper submitting to ACL 2026 about the data analysis on CIRCUIT_EDU_HW sampled in Gatech over the course ECE 2040 throughout the Spring 2025 Term.

Vercel's official collection of agent skills

MCP Toolbox for Databases is an open source MCP server for databases.

Implementation of my RAG system that won all categories in Enterprise RAG Challenge 2

Get your documents ready for gen AI

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.

Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)

AI suite powered by state-of-the-art models and providing advanced AI/AGI functions. Includes AI personas, AGI functions, world-class Beam multi-model chats, text-to-image, voice, response streamin…

Solve Visual Understanding with Reinforced VLMs

Understanding R1-Zero-Like Training: A Critical Perspective

Advanced AI Explainability for computer vision. Support for CNNs, Vision Transformers, Classification, Object detection, Segmentation, Image similarity and more.

[NeurIPS 2024] SimPO: Simple Preference Optimization with a Reference-Free Reward

Set up a modern web app by running one command.

Data Mining Class Project

RewardBench: the first evaluation tool for reward models.

Recipes to train reward model for RLHF.

Training Sparse Autoencoders on Language Models

Code for the paper "A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders"

SOTA Open Source TTS

Official repository for ICML 2024 paper "On Prompt-Driven Safeguarding for Large Language Models"

Official data repository for ICML 2024 paper "On Prompt-Driven Safeguarding for Large Language Models"

Sparse Autoencoder for Mechanistic Interpretability

Using sparse coding to find distributed representations used by neural networks.

This is the code corresponding to the paper "Resolve Domain Conflicts for Generalizable Remote Physiological Measurement." accepted in ACM MM 2023.

This is the official Gtihub repo for our paper: "BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models".

Code of NAACL 2024 paper "Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections".

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.

Official Implementation of ICLR 2022 paper, ``Adversarial Unlearning of Backdoors via Implicit Hypergradient''