Applications
MMG-Vid: Maximizing Marginal Gains at Segment-level and Token-level for Efficient Video LLMs
arXiv:2508.21044v2 Announce Type: replace Abstract: Video Large Language Models (VLLMs) excel in video understanding, but their excessive visual tokens pose a significant computational challenge for r
arXiv:2508.21044v2 Announce Type: replace Abstract: Video Large Language Models (VLLMs) excel in video understanding, but their excessive visual tokens pose a significant computational challenge for real-world applications. Current methods aim to enhance inference efficiency by visual token pruning. However, they do not consider the dynamic characteristics and temporal dependencies of video frames, as they perceive video understanding as a multi-frame task. To address these challenges, we propose MMG-Vid, a novel training-free visual token pruning framework that removes redundancy by Maximizing Marginal Gains at both segment-level and token-level. Specifically, we first divide the video into segments based on frame similarity, and then dynamically allocate the token budget for each segment to maximize the marginal gain of each segment. Subsequently, we propose a temporal-guided DPC algorithm that jointly models inter-frame uniqueness and intra-frame diversity, thereby maximizing the marginal gain of each token. By combining both stages, MMG-Vid can maximize the utilization of the limited token budget, significantly improving efficiency while maintaining strong performance. Extensive experiments demonstrate that MMG-Vid can maintain over 99.5% of the original performance, while effectively reducing 75% visual tokens and accelerating the prefilling stage by 3.9x on LLaVA-OneVision-7B. Code is available at https://github.com/ylnxxts/MMG-Vid.
Related
- InfoMerge: Information-aware Token Compression for Efficient Video Large Language Models
- ERASE: Eliminating Redundant Visual Tokens via Adaptive Two-Stage Token Pruning
- Token Reduction via Local and Global Contexts Optimization for Efficient Video Large Language Models
- VisionSelector: End-to-End Learnable Visual Token Compression for Efficient Multimodal LLMs
- Attention-Free and Lightweight Token Reduction for Efficient Vision-Language Models
Source: arXiv cs.CV | 2026-10-14