基于功能双锚点的模型融合方法

摘要

模型融合是一种高效的后训练策略，用于整合共享基础模型多个微调检查点的知识。现有方法在参数空间中进行操作，通过组合任务向量来缓解冲突，但仍受限于参数不一致性。我们提出功能双锚点（FDA）框架，该方法转而对输入表示空间进行建模。FDA是合成的输入样本，其诱导的梯度与任务向量对齐，能够捕捉相对于预训练模型的任务特定功能偏移。这一视角搭建了联合多任务训练与事后融合的桥梁，兼具鲁棒性与灵活性。我们进一步引入理论驱动的初始化方案，并证明FDA与参数空间模型融合具有互补性。综合实验验证了FDA在模型融合中的有效性。

English

Model merging is an efficient post-training strategy for integrating knowledge from multiple finetuned checkpoints of a shared foundation model. Existing methods operate in the parameter space, combining task vectors to mitigate conflicts, but remain constrained by parameter inconsistencies. We propose Functional Dual Anchors (FDAs), a framework that instead models the input-representation space. FDAs are synthetic inputs whose induced gradients align with task vectors, capturing task-specific functional shifts relative to the pretrained model. This perspective bridges joint multi-task training and post-hoc merging, offering both robustness and flexibility. We further introduce a principled initialization scheme and show that FDAs are complementary to parameter-space model merging. Comprehensive experiments demonstrate the effectiveness of FDAs in model merging.

基于功能双锚点的模型融合方法

Model Merging with Functional Dual Anchors

摘要

Support