[论文] HyperCT: Low-Rank Hypernet for Unified Chest CT Analysis

## 论文概要 **研究领域**: CV **作者**: Fengbei Liu, Sunwoo Kwak, Hao Phung 等 **发布时间**: 2026-04-03 **arXiv**: [2604.03224](https://arxiv.org/abs/2604.03224) ## 中文摘要 非增强胸部CT为常规肺部和机会性肺外筛查提供了丰富机会。虽然多任务学习可以统一这些不同任务，但标准的硬参数共享方法往往在建模不同病理方面表现不佳。我们提出HyperCT，一个通过超网络动态适配Vision Transformer主干的框架。为确保计算效率，我们集成低秩适配，使模型能够回归任务特定的低秩权重更新而非完整参数。 ## 原文摘要 Non-contrast chest CTs offer a rich opportunity for both conventional pulmonary and opportunistic extra-pulmonary screening. While Multi-Task Learning (MTL) can unify these diverse tasks, standard hard-parameter sharing approaches are often suboptimal for modeling distinct pathologies. We propose HyperCT, a framework that dynamically adapts a Vision Transformer backbone via a Hypernetwork. To ensure computational efficiency, we integrate Low-Rank Adaptation (LoRA), allowing the model to regress task-specific low-rank weight updates rather than full parameters. Validated on a large-scale dataset of radiological and cardiological tasks, \method{} outperforms various strong baselines, offering a unified, parameter-efficient solution for holistic patient assessment. Our code is available at ht... --- *自动采集于 2026-04-06* #论文 #arXiv #CV #小凯