Macaw-LLM: Multi-Modal Language Modeling with Image, Audio, Video, and Text Integration

TL;DR

Macaw-LLM is a multi-modal language model that integrates visual, audio, video, and text data, enabling LLMs to handle diverse modalities for complex real-world tasks.

Contribution

The paper introduces Macaw-LLM, a novel multi-modal LLM with a new alignment module and a large-scale multi-modal instruction dataset, advancing multi-modal capabilities in LLMs.

Findings

Successfully integrates multi-modal data with textual information.

Constructed a large-scale multi-modal instruction dataset.

Open-sourced data, code, and models for future research.

Abstract

Although instruction-tuned large language models (LLMs) have exhibited remarkable capabilities across various NLP tasks, their effectiveness on other data modalities beyond text has not been fully studied. In this work, we propose Macaw-LLM, a novel multi-modal LLM that seamlessly integrates visual, audio, and textual information. Macaw-LLM consists of three main components: a modality module for encoding multi-modal data, a cognitive module for harnessing pretrained LLMs, and an alignment module for harmonizing diverse representations. Our novel alignment module seamlessly bridges multi-modal features to textual features, simplifying the adaptation process from the modality modules to the cognitive module. In addition, we construct a large-scale multi-modal instruction dataset in terms of multi-turn dialogue, including 69K image instances and 50K video instances. We have made our data, code and model publicly available, which we hope can pave the way for future research in multi-modal LLMs and expand the capabilities of LLMs to handle diverse data modalities and address complex real-world scenarios.