Exploring Forgetting in Large Language Model Pre-Training

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liao, Chonghua, Xie, Ruobing, Sun, Xingwu, Sun, Haowen, Kang, Zhanhui
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914984086208512
author Liao, Chonghua
Xie, Ruobing
Sun, Xingwu
Sun, Haowen
Kang, Zhanhui
author_facet Liao, Chonghua
Xie, Ruobing
Sun, Xingwu
Sun, Haowen
Kang, Zhanhui
contents Catastrophic forgetting remains a formidable obstacle to building an omniscient model in large language models (LLMs). Despite the pioneering research on task-level forgetting in LLM fine-tuning, there is scant focus on forgetting during pre-training. We systematically explored the existence and measurement of forgetting in pre-training, questioning traditional metrics such as perplexity (PPL) and introducing new metrics to better detect entity memory retention. Based on our revised assessment of forgetting metrics, we explored low-cost, straightforward methods to mitigate forgetting during the pre-training phase. Further, we carefully analyzed the learning curves, offering insights into the dynamics of forgetting. Extensive evaluations and analyses on forgetting of pre-training could facilitate future research on LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2410_17018
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Exploring Forgetting in Large Language Model Pre-Training
Liao, Chonghua
Xie, Ruobing
Sun, Xingwu
Sun, Haowen
Kang, Zhanhui
Computation and Language
Catastrophic forgetting remains a formidable obstacle to building an omniscient model in large language models (LLMs). Despite the pioneering research on task-level forgetting in LLM fine-tuning, there is scant focus on forgetting during pre-training. We systematically explored the existence and measurement of forgetting in pre-training, questioning traditional metrics such as perplexity (PPL) and introducing new metrics to better detect entity memory retention. Based on our revised assessment of forgetting metrics, we explored low-cost, straightforward methods to mitigate forgetting during the pre-training phase. Further, we carefully analyzed the learning curves, offering insights into the dynamics of forgetting. Extensive evaluations and analyses on forgetting of pre-training could facilitate future research on LLMs.
title Exploring Forgetting in Large Language Model Pre-Training
topic Computation and Language
url https://arxiv.org/abs/2410.17018