MMG-Vid: Maximizing Marginal Gains at Segment-level and Token-level for Efficient Video LLMs
cs.CV
Submitted: 2025-08-28
Updated: 2026-10-11
Comments: 10 pages, 3 figures. Accepted by AAAI 2026. Code: https://github.com/ylnxxts/MMG-Vid
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- LongVILA: Scaling Long-Context Visual Language Models for Long Videos
- Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis
- FrameFusion: Combining Similarity and Importance for Video Token Reduction on Large Vision Language Models
- Filter, Correlate, Compress: Training-Free Token Reduction for MLLM Acceleration
- PruneVid: Visual Token Pruning for Efficient Video Large Language Models
- LLaVA-OneVision: Easy Visual Task Transfer
- Video Compression Commander: Plug-and-Play Inference Acceleration for Video Large Language Models
- Global Compression Commander: Plug-and-Play Inference Acceleration for High-Resolution Large Vision-Language Models
- Shifting AI Efficiency From Model-Centric to Data-Centric Compression
- LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models
- HoliTom: Holistic Token Merging for Fast Video Large Language Models
- FastVID: Dynamic Density Pruning for Fast Video Large Language Models
- LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMs
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction
- VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning
- VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding
- LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models
- Long Context Transfer from Language to Vision
- Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models