Paper · arXiv 2504.10068
Mavors: Multi-granularity Video Representation for Multimodal Large Language Model
Yang Shi, Jiaheng Liu, Yushuo Guan, Zhenhua Wu, Yuanxing Zhang, Zihao Wang, Weihong Lin, Jingyun Hua, Zekun Wang, Xinlong Chen, Bohan Zeng, Wentao Zhang, Fuzheng Zhang, Wenjing Yang, Di Zhang
Mavors, a novel framework for long-context video understanding, uses a multi-granularity video representation to preserve spatial fidelity and temporal continuity through 3D convolutions, Vision Transformers, and transformer-based dependency modeling.
Abstract
Long-context video understanding in multimodal large language models (MLLMs) faces a critical challenge: balancing computational efficiency with the retention of fine-grained spatio-temporal patterns. Existing approaches (e.g., sparse sampling, dense sampling with low resolution, and token compression) suffer from significant information loss in temporal dynamics, spatial details, or subtle interactions, particularly in videos with complex motion or varying resolutions. To address this, we propose Mavors, a novel framework that introduces Multi-granularity video representation for holistic long-video modeling. Specifically, Mavors directly encodes raw video content into latent representations through two core components: 1) an Intra-chunk Vision Encoder (IVE) that preserves high-resolution spatial features via 3D convolutions and Vision Transformers, and 2) an Inter-chunk Feature Aggregator (IFA) that establishes temporal coherence across chunks using transformer-based dependency modeling with chunk-level rotary position encodings. Moreover, the framework unifies image and video understanding by treating images as single-frame videos via sub-image decomposition. Experiments across diverse benchmarks demonstrate Mavors' superiority in maintaining both spatial fidelity and temporal continuity, significantly outperforming existing methods in tasks requiring fine-grained spatio-temporal reasoning.