Large Language Models Meet NLP: A Survey

Libo Qin,Qiguang Chen,Xiachong Feng,Yang Wu,Yongheng Zhang,Yinghui Li,Min Li,Wanxiang Che,Philip S. Yu
2024-05-21
Abstract:While large language models (LLMs) like ChatGPT have shown impressive capabilities in Natural Language Processing (NLP) tasks, a systematic investigation of their potential in this field remains largely unexplored. This study aims to address this gap by exploring the following questions: (1) How are LLMs currently applied to NLP tasks in the literature? (2) Have traditional NLP tasks already been solved with LLMs? (3) What is the future of the LLMs for NLP? To answer these questions, we take the first step to provide a comprehensive overview of LLMs in NLP. Specifically, we first introduce a unified taxonomy including (1) parameter-frozen application and (2) parameter-tuning application to offer a unified perspective for understanding the current progress of LLMs in NLP. Furthermore, we summarize the new frontiers and the associated challenges, aiming to inspire further groundbreaking advancements. We hope this work offers valuable insights into the {potential and limitations} of LLMs in NLP, while also serving as a practical guide for building effective LLMs in NLP.
Computation and Language,Artificial Intelligence
What problem does this paper attempt to address?
The paper attempts to address the following issues: 1. **The current application of large language models (LLMs) in natural language processing (NLP) tasks**: The paper explores the existing applications of LLMs in the field of NLP, particularly how these models are used for various NLP tasks. 2. **Whether traditional NLP tasks have already been solved by LLMs**: The paper evaluates the effectiveness of LLMs in solving traditional NLP tasks, discussing whether these models have been able to replace or surpass traditional NLP methods. 3. **The future development direction of LLMs in the field of NLP**: The paper looks ahead to the future trends of LLMs in the field of NLP, proposing new research frontiers and challenges, aiming to provide direction for future innovations. To answer these questions, the paper undertakes the following work: - **Introducing a unified classification system**: The paper proposes a classification system that includes "parameter freezing applications" and "parameter tuning applications" to provide a unified perspective on the progress of LLMs in the field of NLP. - **Summarizing new research frontiers and their challenges**: The paper summarizes the latest research frontiers of LLMs in the field of NLP and points out the related challenges, aiming to inspire more breakthrough progress. - **Providing rich resources**: The paper creates a resource collection that includes open-source implementations, relevant corpora, and research papers to support the work of researchers. Through this work, the paper hopes to provide valuable insights to researchers and promote the further development of LLMs in the field of NLP.