Instruction Tuning with GPT-4
arXiv:2304.03277
Abstract
Prior work has shown that finetuning large language models (LLMs) using machine-generated instruction-following data enables such models to achieve remarkable zero-shot capabilities on new tasks, and no human-written instructions are needed. In this paper, we present the first attempt to use GPT-4 to generate instruction-following data for LLM finetuning. Our early experiments on instruction-tuned LLaMA models show that the 52K English and Chinese instruction-following data generated by GPT-4 leads to superior zero-shot performance on new tasks to the instruction-following data generated by previous state-of-the-art models. We also collect feedback and comparison data from GPT-4 to enable a comprehensive evaluation and reward model training. We make our data generated using GPT-4 as well as our codebase publicly available.
arXiv comment8 pages. Work in progress. Project page: https://instruction-tuning-with-gpt-4.github.io
Cited by in corpus (12)
- A Survey on Multimodal Large Language Models
- ChatGPT: Jack of all trades, master of none
- Learning from models beyond fine-tuning
- A Large Language Model Approach to Educational Survey Feedback Analysis
- Achieving Peak Performance for Large Language Models: A Systematic Review
- Several categories of Large Language Models (LLMs): A Short Survey
- MULTITuDE: Large-Scale Multilingual Machine-Generated Text Detection Benchmark
- Power-LLaVA: Large Language and Vision Assistant for Power Transmission Line Inspection
- Domain-Specific Improvement on Psychotherapy Chatbot Using Assistant
- Fabricator: An Open Source Toolkit for Generating Labeled Training Data with Teacher LLMs
- QCaption: Video Captioning and Q&A through Fusion of Large Multimodal Models
- Backdoor Samples Detection Based on Perturbation Discrepancy Consistency in Pre-trained Language Models