RE-Adapt: Reverse Engineered Adaptation of Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fleshman, William, Van Durme, Benjamin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916258822225920
author Fleshman, William
Van Durme, Benjamin
author_facet Fleshman, William
Van Durme, Benjamin
contents We introduce RE-Adapt, an approach to fine-tuning large language models on new domains without degrading any pre-existing instruction-tuning. We reverse engineer an adapter which isolates what an instruction-tuned model has learned beyond its corresponding pretrained base model. Importantly, this requires no additional data or training. We can then fine-tune the base model on a new domain and readapt it to instruction following with the reverse engineered adapter. RE-Adapt and our low-rank variant LoRE-Adapt both outperform other methods of fine-tuning, across multiple popular LLMs and datasets, even when the models are used in conjunction with retrieval-augmented generation.
format Preprint
id arxiv_https___arxiv_org_abs_2405_15007
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RE-Adapt: Reverse Engineered Adaptation of Large Language Models
Fleshman, William
Van Durme, Benjamin
Computation and Language
Artificial Intelligence
Machine Learning
We introduce RE-Adapt, an approach to fine-tuning large language models on new domains without degrading any pre-existing instruction-tuning. We reverse engineer an adapter which isolates what an instruction-tuned model has learned beyond its corresponding pretrained base model. Importantly, this requires no additional data or training. We can then fine-tune the base model on a new domain and readapt it to instruction following with the reverse engineered adapter. RE-Adapt and our low-rank variant LoRE-Adapt both outperform other methods of fine-tuning, across multiple popular LLMs and datasets, even when the models are used in conjunction with retrieval-augmented generation.
title RE-Adapt: Reverse Engineered Adaptation of Large Language Models
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2405.15007