MCITlib: Multimodal Continual Instruction Tuning Library and Benchmark

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Guo, Haiyang, Zhu, Fei, Zhao, Hongbo, Zeng, Fanhu, Liu, Wenzhuo, Ma, Shijie, Wang, Da-Han, Zhang, Xu-Yao
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908739492118528
author Guo, Haiyang
Zhu, Fei
Zhao, Hongbo
Zeng, Fanhu
Liu, Wenzhuo
Ma, Shijie
Wang, Da-Han
Zhang, Xu-Yao
author_facet Guo, Haiyang
Zhu, Fei
Zhao, Hongbo
Zeng, Fanhu
Liu, Wenzhuo
Ma, Shijie
Wang, Da-Han
Zhang, Xu-Yao
contents Continual learning enables AI systems to acquire new knowledge while retaining previously learned information. While traditional unimodal methods have made progress, the rise of Multimodal Large Language Models (MLLMs) brings new challenges in Multimodal Continual Learning (MCL), where models are expected to address both catastrophic forgetting and cross-modal coordination. To advance research in this area, we present MCITlib, a comprehensive library for Multimodal Continual Instruction Tuning. MCITlib currently implements 8 representative algorithms and conducts evaluations on 3 benchmarks under 2 backbone models. The library will be continuously updated to support future developments in MCL. The codebase is released at https://github.com/Ghy0501/MCITlib.
format Preprint
id arxiv_https___arxiv_org_abs_2508_07307
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MCITlib: Multimodal Continual Instruction Tuning Library and Benchmark
Guo, Haiyang
Zhu, Fei
Zhao, Hongbo
Zeng, Fanhu
Liu, Wenzhuo
Ma, Shijie
Wang, Da-Han
Zhang, Xu-Yao
Computer Vision and Pattern Recognition
Artificial Intelligence
Continual learning enables AI systems to acquire new knowledge while retaining previously learned information. While traditional unimodal methods have made progress, the rise of Multimodal Large Language Models (MLLMs) brings new challenges in Multimodal Continual Learning (MCL), where models are expected to address both catastrophic forgetting and cross-modal coordination. To advance research in this area, we present MCITlib, a comprehensive library for Multimodal Continual Instruction Tuning. MCITlib currently implements 8 representative algorithms and conducts evaluations on 3 benchmarks under 2 backbone models. The library will be continuously updated to support future developments in MCL. The codebase is released at https://github.com/Ghy0501/MCITlib.
title MCITlib: Multimodal Continual Instruction Tuning Library and Benchmark
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2508.07307