In:
ACM Computing Surveys, Association for Computing Machinery (ACM), Vol. 55, No. 9 ( 2023-09-30), p. 1-35
Abstract:
This article surveys and organizes research works in a new paradigm in natural language processing, which we dub “prompt-based learning.” Unlike traditional supervised learning, which trains a model to take in an input x and predict an output y as P ( y|x ), prompt-based learning is based on language models that model the probability of text directly. To use these models to perform prediction tasks, the original input x is modified using a template into a textual string prompt x′ that has some unfilled slots, and then the language model is used to probabilistically fill the unfilled information to obtain a final string x̂ , from which the final output y can be derived. This framework is powerful and attractive for a number of reasons: It allows the language model to be pre-trained on massive amounts of raw text, and by defining a new prompting function the model is able to perform few-shot or even zero-shot learning, adapting to new scenarios with few or no labeled data. In this article, we introduce the basics of this promising paradigm, describe a unified set of mathematical notations that can cover a wide variety of existing work, and organize existing work along several dimensions, e.g., the choice of pre-trained language models, prompts, and tuning strategies. To make the field more accessible to interested beginners, we not only make a systematic review of existing works and a highly structured typology of prompt-based concepts but also release other resources, e.g., a website NLPedia–Pretrain including constantly updated survey and paperlist.
Type of Medium:
Online Resource
ISSN:
0360-0300
,
1557-7341
Language:
English
Publisher:
Association for Computing Machinery (ACM)
Publication Date:
2023
detail.hit.zdb_id:
215909-0
detail.hit.zdb_id:
1495309-2
detail.hit.zdb_id:
626472-4
Permalink