# MMG-Vid: Maximizing Marginal Gains at Segment-level and Token-level for Efficient Video LLMs

arXiv: https://arxiv.org/abs/2508.21044 · HTML: https://mlsanity.com/paper/2508.21044/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning in multimodal models, one of 12 directions in Efficient LLM inference and serving.
- First posted on arXiv 28 Aug 2025.

## Placed in

- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Built on by (most cited first, 14 of 14)

- [Accelerating Streaming Video Large Language Models via Hierarchical Token Compression](https://mlsanity.com/paper/2512.00891/) (2025)
- [Variation-aware Vision Token Dropping for Faster Large Vision-Language Models](https://mlsanity.com/paper/2509.01552/) (2025)
- [GIFT: Global Irreplaceability Frame Targeting for Efficient Video Understanding](https://mlsanity.com/paper/2603.25072/) (2026)
- [InfoMerge: Information-aware Token Compression for Efficient Video Large Language Models](https://mlsanity.com/paper/2606.02161/) (2026)
- [OmniDocLayout: Towards Diverse Document Layout Generation via Coarse-to-Fine LLM Learning](https://mlsanity.com/paper/2510.26213/) (2025)
- [V-CAST: Video Curvature-Aware Spatio-Temporal Pruning for Efficient Video Large Language Models](https://mlsanity.com/paper/2603.27650/) (2026)
- [KiToke: Kernel-based Interval-aware Token Compression for Video Large Language Models](https://mlsanity.com/paper/2604.03414/) (2026)
- [EchoPrune: Interpreting Redundancy as Temporal Echoes for Efficient VideoLLMs](https://mlsanity.com/paper/2605.10050/) (2026)
- [ShaRP: SHAllow-LayeR Pruning for Efficient Video Large Language Models](https://mlsanity.com/paper/2512.05385/) (2025)
- [DINO-VO: Learning Where to Focus for Enhanced State Estimation](https://mlsanity.com/paper/2604.04055/) (2026)
- [Sink-Token-Aware Pruning for Fine-Grained Video Understanding in Efficient Video LLMs](https://mlsanity.com/paper/2604.20937/) (2026)
- [CARVE: Cross-Slice Anisotropic Reallocation of Visual Evidence for Efficient 3D Medical Volume Understanding](https://mlsanity.com/paper/2608.04515/) (2026)
- [Why Is Video Still So Expensive? A Survey of Inference-Efficiency Mechanisms in Video and Audiovisual LLMs](https://mlsanity.com/paper/2609.10355/) (2026)
- [GleanVID: Complementary Token Selection for Efficient Video Large Language Models](https://mlsanity.com/paper/2609.37042/) (2026)

## Abstract (arXiv)

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.