发表机构
Zhejiang University; University of Nevada, Reno; Guangdong Institute of Intelligence Science and Technology（浙江大学; 内华达大学雷诺分校; 广东省智能科学与技术研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对现有ICL方法的浅层任务适配问题，提出PromptPath框架，通过提示驱动的路由机制动态重构模型计算，在视觉识别基准上性能优于现有ICL基线且泛化能力强。
AI 中文摘要
上下文学习（ICL）因能让模型仅通过少量“输入-输出”提示示例执行新任务而受到越来越多关注。然而，现有方法存在“浅层任务适配”问题：提示主要作为上下文线索，通过语义表示隐式推断任务意图，而底层计算过程保持不变。这一限制阻碍了特定任务适配，并损害了推理可解释性。我们认为，提示不仅应调控特征表示，还应动态调节模型的计算路径。为此，我们提出PromptPath，这是一种自适应ICL框架，通过提示条件化的动态路径实现计算层面的适配。具体而言，PromptPath引入提示驱动的路由机制，选择性激活并组合轻量级低秩专家，形成适配不同提示的特定任务计算路径。通过将提示信息直接整合到推理过程中，PromptPath动态重构模型计算，以增强任务专业性和可解释性。在3D点云和2D视觉识别基准上开展的大量实验表明，PromptPath始终优于最先进的ICL基线，同时展现出强大的跨域和跨任务泛化能力。
英文摘要
In-context learning (ICL) has attracted increasing attention for enabling models to perform new tasks using only a few ``input--output'' prompt examples. However, existing approaches suffer from \textbf{shallow task adaptation}, where prompts are primarily used as contextual cues to implicitly infer task intent through semantic representations, while the underlying computational process remains unchanged. This limitation restricts task-specific adaptation and compromises inference interpretability. We argue that prompts should not only condition feature representations but also dynamically regulate the model's computation pathways. To this end, we propose \textbf{PromptPath}, an adaptive ICL framework that enables computation-level adaptation through prompt-conditioned dynamic pathways. Specifically, PromptPath introduces a prompt-driven routing mechanism to selectively activate and compose lightweight low-rank experts, forming task-specific computational pathways tailored to different prompts. By integrating prompt information directly into the inference process, PromptPath dynamically reconfigures model computation to enhance task specialization and interpretability. Extensive experiments on 3D point cloud and 2D visual recognition benchmarks demonstrate that PromptPath consistently outperforms state-of-the-art ICL baselines while exhibiting strong cross-domain and cross-task generalization.
