What type of study is this?

This is a Experimental Study study.

October 20, 2025Open Access

Pyramid Token Pruning for High-Resolution Large Vision-Language Models via Region, Token, and Instruction-Guided Importance

Key Points

Pyramid Token Pruning effectively reduces inference latency, enabling faster processing of high-resolution images.
Experimental results indicate that PTP lowers computational costs by preserving only relevant visual tokens while maintaining model performance.
The method integrates visual saliency and instruction-guided relevance to enhance token selection from image regions.
PTP was tested across 13 diverse benchmarks, demonstrating its versatility and effectiveness in various scenarios.

Abstract

Large Vision-Language Models (LVLMs) have recently demonstrated strong multimodal understanding, yet their fine-grained visual perception is often constrained by low input resolutions. A common remedy is to partition high-resolution images into multiple sub-images for separate encoding, but this approach drastically inflates the number of visual tokens and introduces prohibitive inference overhead. To overcome this challenge, we propose Pyramid Token Pruning (PTP), a training-free strategy that hierarchically integrates bottom-up visual saliency at both region and token levels with top-down instruction-guided relevance. Inspired by human visual cognition, PTP selectively preserves more tokens from salient regions while further emphasizing those most relevant to task instructions. Extensive experiments on 13 diverse benchmarks show that PTP substantially reduces computational cost, memory usage, and inference latency, with negligible performance degradation.

Pyramid Token Pruning for High-Resolution Large Vision-Language Models via Region, Token, and Instruction-Guided Importance

Key Points

Abstract

Cite This Study

Also Consider

Also Consider