CPM: A Large-scale Generative Chinese Pre-trained Language Model

TL;DR

This paper introduces CPM, a large-scale Chinese pre-trained language model with 2.6 billion parameters trained on 100GB of data, demonstrating strong performance on various Chinese NLP tasks, especially in few-shot and zero-shot settings.

Contribution

The paper presents CPM, the largest Chinese pre-trained language model to date, with generative pre-training on extensive Chinese data, enabling improved NLP task performance.

Findings

CPM achieves state-of-the-art results on multiple Chinese NLP tasks.

Effective in few-shot and zero-shot learning scenarios.

Provides publicly available code and model parameters.

Abstract

Pre-trained Language Models (PLMs) have proven to be beneficial for various downstream NLP tasks. Recently, GPT-3, with 175 billion parameters and 570GB training data, drew a lot of attention due to the capacity of few-shot (even zero-shot) learning. However, applying GPT-3 to address Chinese NLP tasks is still challenging, as the training corpus of GPT-3 is primarily English, and the parameters are not publicly available. In this technical report, we release the Chinese Pre-trained Language Model (CPM) with generative pre-training on large-scale Chinese training data. To the best of our knowledge, CPM, with 2.6 billion parameters and 100GB Chinese training data, is the largest Chinese pre-trained language model, which could facilitate several downstream Chinese NLP tasks, such as conversation, essay generation, cloze test, and language understanding. Extensive experiments demonstrate that CPM achieves strong performance on many NLP tasks in the settings of few-shot (even zero-shot) learning. The code and parameters are available at https://github.com/TsinghuaAI/CPM-Generate.