Guardado en:
Detalles Bibliográficos
Autores principales: Ye, Junjie, Yang, Yuming, Zhang, Qi, Gui, Tao, Huang, Xuanjing, Wang, Peng, Shi, Zhongchao, Fan, Jianping
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2409.15825
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929682786549760
author Ye, Junjie
Yang, Yuming
Zhang, Qi
Gui, Tao
Huang, Xuanjing
Wang, Peng
Shi, Zhongchao
Fan, Jianping
author_facet Ye, Junjie
Yang, Yuming
Zhang, Qi
Gui, Tao
Huang, Xuanjing
Wang, Peng
Shi, Zhongchao
Fan, Jianping
contents Large language models (LLMs) encode extensive world knowledge through pre-training on massive datasets, which can then be fine-tuned for the question-answering (QA) task. However, effective strategies for fine-tuning LLMs for the QA task remain largely unexplored. To address this gap, we categorize supervised fine-tuning (SFT) data based on the extent of knowledge memorized by the pretrained LLMs and conduct a series of empirical analyses. Our experiments, involving four LLMs from three different model families, focus on three key factors: the amount of data required for SFT, the impact of different SFT datasets on model performance, and how data requirements vary across LLMs. The results show that as few as 60 data points during the SFT stage can activate the knowledge encoded during pre-training, enabling LLMs to perform the QA task. Additionally, SFT with data of varying memory levels has a significant impact on LLM performance, with the optimal dataset differing based on the specific model being fine-tuned. Future research will delve deeper into the mechanisms underlying these phenomena.
format Preprint
id arxiv_https___arxiv_org_abs_2409_15825
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle 60 Data Points are Sufficient to Fine-Tune LLMs for Question-Answering
Ye, Junjie
Yang, Yuming
Zhang, Qi
Gui, Tao
Huang, Xuanjing
Wang, Peng
Shi, Zhongchao
Fan, Jianping
Computation and Language
Artificial Intelligence
Large language models (LLMs) encode extensive world knowledge through pre-training on massive datasets, which can then be fine-tuned for the question-answering (QA) task. However, effective strategies for fine-tuning LLMs for the QA task remain largely unexplored. To address this gap, we categorize supervised fine-tuning (SFT) data based on the extent of knowledge memorized by the pretrained LLMs and conduct a series of empirical analyses. Our experiments, involving four LLMs from three different model families, focus on three key factors: the amount of data required for SFT, the impact of different SFT datasets on model performance, and how data requirements vary across LLMs. The results show that as few as 60 data points during the SFT stage can activate the knowledge encoded during pre-training, enabling LLMs to perform the QA task. Additionally, SFT with data of varying memory levels has a significant impact on LLM performance, with the optimal dataset differing based on the specific model being fine-tuned. Future research will delve deeper into the mechanisms underlying these phenomena.
title 60 Data Points are Sufficient to Fine-Tune LLMs for Question-Answering
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2409.15825