Improving Multimodal Large Language Models Using Continual Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Srivastava, Shikhar, Harun, Md Yousuf, Shrestha, Robik, Kanan, Christopher
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912534860136448
author Srivastava, Shikhar
Harun, Md Yousuf
Shrestha, Robik
Kanan, Christopher
author_facet Srivastava, Shikhar
Harun, Md Yousuf
Shrestha, Robik
Kanan, Christopher
contents Generative large language models (LLMs) exhibit impressive capabilities, which can be further augmented by integrating a pre-trained vision model into the original LLM to create a multimodal LLM (MLLM). However, this integration often significantly decreases performance on natural language understanding and generation tasks, compared to the original LLM. This study investigates this issue using the LLaVA MLLM, treating the integration as a continual learning problem. We evaluate five continual learning methods to mitigate forgetting and identify a technique that enhances visual understanding while minimizing linguistic performance loss. Our approach reduces linguistic performance degradation by up to 15% over the LLaVA recipe, while maintaining high multimodal accuracy. We also demonstrate the robustness of our method through continual learning on a sequence of vision-language tasks, effectively preserving linguistic skills while acquiring new multimodal capabilities. Project webpage: https://shikhar-srivastava.github.io/cl-for-improving-mllms
format Preprint
id arxiv_https___arxiv_org_abs_2410_19925
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Improving Multimodal Large Language Models Using Continual Learning
Srivastava, Shikhar
Harun, Md Yousuf
Shrestha, Robik
Kanan, Christopher
Computation and Language
Computer Vision and Pattern Recognition
Machine Learning
Generative large language models (LLMs) exhibit impressive capabilities, which can be further augmented by integrating a pre-trained vision model into the original LLM to create a multimodal LLM (MLLM). However, this integration often significantly decreases performance on natural language understanding and generation tasks, compared to the original LLM. This study investigates this issue using the LLaVA MLLM, treating the integration as a continual learning problem. We evaluate five continual learning methods to mitigate forgetting and identify a technique that enhances visual understanding while minimizing linguistic performance loss. Our approach reduces linguistic performance degradation by up to 15% over the LLaVA recipe, while maintaining high multimodal accuracy. We also demonstrate the robustness of our method through continual learning on a sequence of vision-language tasks, effectively preserving linguistic skills while acquiring new multimodal capabilities. Project webpage: https://shikhar-srivastava.github.io/cl-for-improving-mllms
title Improving Multimodal Large Language Models Using Continual Learning
topic Computation and Language
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2410.19925