Han Zhou's Projects
An Open-source Framework for Data-centric, Self-evolving Autonomous Language Agents
🤖 AgentVerse 🪐 is designed to facilitate the deployment of multiple LLM-based agents in various applications, which primarily provides two frameworks: task-solving and simulation
We unified the interfaces of instruction-tuning data (e.g., CoT data), multiple LLMs and parameter-efficient methods (e.g., lora, p-tuning) together for easy use. Meanwhile, we created a new branch to build a Tabular LLM.(我们分别统一了丰富的IFT数据(如CoT数据,目前仍不断扩充)、多种训练效率方法(如lora,p-tuning)以及多种LLMs,三个层面上的接口,打造方便研究人员上手的LLM-IFT研究平台。同时tabular_llm分支构建了面向表格智能任务的LLM。
Instruct-tune LLaMA on consumer hardware
Collection of Tools and Papers related to Adapters (aka Parameter-Efficient Transfer Learning/ Fine-Tuning)
Awesome-LLM-Prompt-Optimization: a curated list of advanced prompt optimization and tuning methods in Large Language Models
Awesome-LLM-Robustness: a curated list of Uncertainty, Reliability and Robustness in Large Language Models
A collection of parameter-efficient transfer learning papers focusing on computer vision and multimodal domains.
ICML'2022: Black-Box Tuning for Language-Model-as-a-Service & EMNLP'2022: BBTv2: Towards a Gradient-Free Future with Large Language Models
Few-shot Learning of GPT-3
An original implementation of "Noisy Channel Language Model Prompting for Few-Shot Text Classification"
Official implementation for the paper "DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models"
Official implementation of "DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning" in ICML'24
DSPy: The framework for programming—not prompting—foundation models
Function Vectors in Large Language Models [ICLR 2024]
Code for our paper: "GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Models"
Code for In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering
Official repository for ICML 2024 paper "On Prompt-Driven Safeguarding for Large Language Models"
Data and code for the preprint "In-Context Learning with Long-Context Models: An In-Depth Exploration"
MAD: The first work to explore Multi-Agent Debate with Large Language Models :D
Source codes of my 4YP in University of Oxford
The official repository for Multi3WOZ: A Multilingual, Multi-Domain, Multi-Parallel Dataset for Training and Evaluating Culturally Adapted Task-Oriented Dialog Systems (Hu et al., to appear; TACL)