Saved in:
Bibliographic Details
Main Authors: Li, Jiahao, Lu, Yang, Zhang, Yachao, Wang, Fangyong, Xie, Yuan, Qu, Yanyun
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.01275
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Open-vocabulary semantic segmentation (OVSS) conducts pixel-level classification via text-driven alignment, where the domain discrepancy between base category training and open-vocabulary inference poses challenges in discriminative modeling of latent unseen category. To address this challenge, existing vision-language model (VLM)-based approaches demonstrate commendable performance through pre-trained multi-modal representations. However, the fundamental mechanisms of latent semantic comprehension remain underexplored, making the bottleneck for OVSS. In this work, we initiate a probing experiment to explore distribution patterns and dynamics of latent semantics in VLMs under inductive learning paradigms. Building on these insights, we propose X-Agent, an innovative OVSS framework employing latent semantic-aware ``agent'' to orchestrate cross-modal attention mechanisms, simultaneously optimizing latent semantic dynamic and amplifying its perceptibility. Extensive benchmark evaluations demonstrate that X-Agent achieves state-of-the-art performance while effectively enhancing the latent semantic saliency.