Continual Learning Using Only Large Language Model Prompting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qiu, Jiabao, Ke, Zixuan, Liu, Bing
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908168671461376
author Qiu, Jiabao
Ke, Zixuan
Liu, Bing
author_facet Qiu, Jiabao
Ke, Zixuan
Liu, Bing
contents We introduce CLOB, a novel continual learning (CL) paradigm wherein a large language model (LLM) is regarded as a black box. Learning is done incrementally via only verbal prompting. CLOB does not fine-tune any part of the LLM or add any trainable parameters to it. It is particularly suitable for LLMs that are accessible via APIs. We also propose a new CL technique, called CIS, based on incremental summarization that also overcomes the LLM's input length limit. Experiments show CIS outperforms baselines by a very large margin.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15479
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Continual Learning Using Only Large Language Model Prompting
Qiu, Jiabao
Ke, Zixuan
Liu, Bing
Computation and Language
Artificial Intelligence
We introduce CLOB, a novel continual learning (CL) paradigm wherein a large language model (LLM) is regarded as a black box. Learning is done incrementally via only verbal prompting. CLOB does not fine-tune any part of the LLM or add any trainable parameters to it. It is particularly suitable for LLMs that are accessible via APIs. We also propose a new CL technique, called CIS, based on incremental summarization that also overcomes the LLM's input length limit. Experiments show CIS outperforms baselines by a very large margin.
title Continual Learning Using Only Large Language Model Prompting
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2412.15479