一个超网络可生成多种架构模型，无需重新设计训练。
Universal Hypernetworks for Arbitrary Models
- 用参数与任务描述符生成权重，解耦生成器与目标模型结构。
- 在视觉、文本等多任务上表现接近直接训练结果。
- 支持跨模型通用与递归生成，适合快速原型开发。
传统超网络通常针对特定基础模型结构设计，更换架构需重新设计并从头训练。本文提出通用超网络（UHN），一种固定架构的生成器，能根据确定性参数、架构和任务描述符预测权重。该基于描述符的范式将生成器架构与目标网络参数化解耦，使单一生成器可实例化不同架构和任务家族中的异构模型。实证表明：(1) 一个固定UHN在视觉、图、文本和公式回归基准上表现与直接训练相当；(2) 同一UHN支持同族模型的多模型泛化及跨异构模型的多任务学习；(3) UHN可稳定递归生成，最多支持三重中间生成的UHN，最终产出基础模型。代码已开源。
原文摘要 · Abstract (English)
Conventional hypernetworks are typically engineered around a specific base-model parameterization, so changing the target architecture often entails redesigning the hypernetwork and retraining it from scratch. We introduce the \emph{Universal Hypernetwork} (UHN), a fixed-architecture generator that predicts weights from deterministic parameter, architecture, and task descriptors. This descriptor-based formulation decouples the generator architecture from target-network parameterization, so one generator can instantiate heterogeneous models across the tested architecture and task families. Our empirical claims are threefold: (1) one fixed UHN remains competitive with direct training across vision, graph, text, and formula-regression benchmarks; (2) the same UHN supports both multi-model generalization within a family and multi-task learning across heterogeneous models; and (3) UHN enables stable recursive generation with up to three intermediate generated UHNs before the final base model. Our code is available at https://github.com/Xuanfeng-Zhou/UHN.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。