CREMA: Generalizable and Efficient Video-Language Reasoning via Multimodal Modular Fusion

TL;DR

CREMA introduces a modular, efficient framework for video-language reasoning that integrates multiple modalities without extensive retraining, achieving comparable or better results with significantly fewer trainable parameters.

Contribution

The paper presents CREMA, a novel multimodal fusion framework that efficiently incorporates diverse modalities into video reasoning models with minimal parameter updates.

Findings

Achieves state-of-the-art performance on 7 video-language reasoning tasks.

Reduces over 90% of trainable parameters compared to strong baselines.

Effectively integrates multiple modalities like audio, thermal, and touch without extra annotation.

Abstract

Despite impressive advancements in recent multimodal reasoning approaches, they are still limited in flexibility and efficiency, as these models typically process only a few fixed modality inputs and require updates to numerous parameters. This paper tackles these critical challenges and proposes CREMA, a generalizable, highly efficient, and modular modality-fusion framework that can incorporate any new modality to enhance video reasoning. We first augment multiple informative modalities (such as optical flow, 3D point cloud, audio, thermal heatmap, and touch map) from given videos without extra human annotation by leveraging sensors or existing pre-trained models. Next, we introduce a query transformer with multiple parameter-efficient modules associated with each accessible modality. It projects diverse modality features to the LLM token embedding space, allowing the model to integrate different data types for response generation. Furthermore, we propose a novel progressive multimodal fusion design supported by a lightweight fusion module and modality-sequential training strategy. It helps compress information across various assisting modalities, maintaining computational efficiency in the LLM while improving performance. We validate our method on 7 video-language reasoning tasks assisted by diverse modalities, including conventional VideoQA and Video-Audio/3D/Touch/Thermal QA, and achieve better/equivalent performance against strong multimodal LLMs, including OneLLM, BLIP-2, and SeViLA while reducing over 90% trainable parameters. We provide extensive analyses of CREMA, including the impact of each modality on reasoning domains, the design of the fusion module, and example visualizations.