C^2Path: Class-Conditional Pathway Decoupling for Vision-Language Incremental Object Detection

arXiv:2608.21937v1 Announce Type: new Abstract: Incremental Object Detection (IOD) aims to enable detectors to continuously learn novel categories while preserving previously acquired knowledge. However, existing methods suffer from two forms of extbf{class knowledge coupling}: class boundary erosion induced by shared parameter updates and class representation entanglement arising from mixed feature encoding. We argue that effective incremental learning requires class-specific computational pathways that enable isolated parameter updates and separated class-wise injection. To this end, we propose extbf{C^2Path}, a class-conditional pathway decoupling framework for vision-language incremental object detection that leverages token-level class cues to establish dedicated and updatable computational pathways for different categories. Specifically, C^2Path introduces a category expert library and a class-conditional decoupling module. The expert library consists of learnable low-rank computational nodes that capture category-specific knowledge, while the decoupling module generates class-aware routing signals to dynamically compose extit{ClassLoRA} adapters from these experts, thereby forming class-specific computational pathways for isolated updates and separated injection across categories. Extensive experiments on COCO 2017 under multiple incremental learning settings demonstrate that C^2Path consistently outperforms state-of-the-art methods, providing an effective and scalable solution for continual category expansion in vision-language detectors.

Source: arXiv cs.CV | 2026-08-25