CoIN: A Benchmark of Continual Instruction tuNing for Multimodel Large Language Model

TL;DR

This paper introduces CoIN, a comprehensive benchmark for evaluating Multimodal Large Language Models in continual instruction tuning, revealing current models' challenges with forgetting and proposing a method to improve retention.

Contribution

The paper presents CoIN, a new benchmark for continual instruction tuning of MLLMs, and proposes MoELoRA to mitigate instruction alignment forgetting during sequential learning.

Findings

Current MLLMs suffer from catastrophic forgetting in instruction tuning.

Instruction alignment failure is the main cause of performance decline.

MoELoRA effectively reduces forgetting in instruction alignment.

Abstract

Instruction tuning represents a prevalent strategy employed by Multimodal Large Language Models (MLLMs) to align with human instructions and adapt to new tasks. Nevertheless, MLLMs encounter the challenge of adapting to users' evolving knowledge and demands. Therefore, how to retain existing skills while acquiring new knowledge needs to be investigated. In this paper, we present a comprehensive benchmark, namely Continual Instruction tuNing (CoIN), to assess existing MLLMs in the sequential instruction tuning paradigm. CoIN comprises 10 commonly used datasets spanning 8 task categories, ensuring a diverse range of instructions and tasks. Besides, the trained model is evaluated from two aspects: Instruction Following and General Knowledge, which assess the alignment with human intention and knowledge preserved for reasoning, respectively. Experiments on CoIN demonstrate that current powerful MLLMs still suffer catastrophic forgetting, and the failure in intention alignment assumes the main responsibility, instead of the knowledge forgetting. To this end, we introduce MoELoRA to MLLMs which is effective to retain the previous instruction alignment. Experimental results consistently illustrate the forgetting decreased from this method on CoIN.