VisionTrim: Unified Vision Token Compression for Training-Free MLLM Acceleration
Abstract
VisionTrim is a training-free framework that accelerates multimodal large language models by selecting dominant visual tokens and merging them with text-guided complementation, improving efficiency without performance loss.
Multimodal large language models (MLLMs) suffer from high computational costs due to excessive visual tokens, particularly in high-resolution and video-based scenarios. Existing token reduction methods typically focus on isolated pipeline components and often neglect textual alignment, leading to performance degradation. In this paper, we propose VisionTrim, a unified framework for training-free MLLM acceleration, integrating two effective plug-and-play modules: 1) the Dominant Vision Token Selection (DVTS) module, which preserves essential visual tokens via a global-local view, and 2) the Text-Guided Vision Complement (TGVC) module, which facilitates context-aware token merging guided by textual cues. Extensive experiments across diverse image and video multimodal benchmarks demonstrate the performance superiority of our VisionTrim, advancing practical MLLM deployment in real-world applications. The code is available at: https://github.com/hanxunyu/VisionTrim.
Community
An efficient vision token compression framework with two modules, Dominant Vision Token Selection (DVTS) and Text-Guided Vision Complement (TGVC).
arXiv explained breakdown of this paper 👉 https://arxivexplained.com/papers/visiontrim-unified-vision-token-compression-for-training-free-mllm-acceleration
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- ViTCoP: Accelerating Large Vision-Language Models via Visual and Textual Semantic Collaborative Pruning (2026)
- Towards Lossless Ultimate Vision Token Compression for VLMs (2025)
- Delta-LLaVA: Base-then-Specialize Alignment for Token-Efficient Vision-Language Models (2025)
- TrimTokenator-LC: Towards Adaptive Visual Token Pruning for Large Multimodal Models with Long Contexts (2025)
- FlashVLM: Text-Guided Visual Token Selection for Large Multimodal Models (2025)
- Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Prior (2025)
- HIPPO: Accelerating Video Large Language Models Inference via Holistic-aware Parallel Speculative Decoding (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper