Pre-Trained Models: Past, Present and Future
Preprint
Abstract
Large-scale pre-trained models (PTMs) such as BERT and GPT have recently achieved great success and become a milestone in the field of artificial intelligence (AI). Owing to sophisticated pre-training objectives and huge model parameters, large-scale PTMs can effectively capture knowledge from massive labeled and unlabeled data. By storing knowledge into huge parameters and fine-tuning on specific tasks, the rich knowledge implicitly encoded in huge parameters can benefit a variety of downstream tasks, which has been extensively demonstrated via experimental verification and empirical analysis. It is now the consensus of the AI community to adopt PTMs as backbone for downstream tasks rather than learning models from scratch. In this paper, we take a deep look into the history of pre-training, especially its special relation with transfer learning and self-supervised learning, to reveal the crucial position of PTMs in the AI development spectrum. Further, we comprehensively review the latest breakthroughs of PTMs. These breakthroughs are driven by the surge of computational power and the increasing availability of data, towards four important directions: designing effective architectures, utilizing rich contexts, improving computational efficiency, and conducting interpretation and theoretical analysis. Finally, we discuss a series of open problems and research directions of PTMs, and hope our view can inspire and advance the future study of PTMs.Keywords
All Related Versions
This publication has 46 references indexed in Scilit:
- PTR: Prompt Tuning with Rules for Text ClassificationAI Open, 2022
- XGPT: Cross-modal Generative Pre-Training for Image Captioning2021
- Towards a universal continuous knowledge baseAI Open, 2021
- CPM: A large-scale generative Chinese Pre-trained language modelAI Open, 2021
- A Large-Scale Chinese Short-Text Conversation Dataset2020
- Pre-trained models for natural language processing: A surveyScience China Technological Sciences, 2020
- Reconciling modern machine-learning practice and the classical bias–variance trade-offProceedings of the National Academy of Sciences, 2019
- Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image AnnotationsInternational Journal of Computer Vision, 2017
- Time Constraints and Resource Sharing in Adults' Working Memory Spans.Journal of Experimental Psychology: General, 2004
- Improving predictive inference under covariate shift by weighting the log-likelihood functionJournal of Statistical Planning and Inference, 2000