通过子空间感知的模型合并跨越领域
Bridging Domains through Subspace-Aware Model Merging
对视觉研究者而言，这是一种无需额外训练即可把多个领域微调模型（如不同成像域、不同风格数据上微调的 ViT）合并为单一模型的手段：通过 SVD 提取各任务矩阵的主奇异向量并构造共享正交基，再剪掉投影后的非对角分量来消除子空间冲突，可直接用于多域视觉模型的部署与跨域泛化提升。
原文依据 SCORE finds a shared orthogonal basis by computing the principal components of the concatenated leading singular vectors of all models.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论模型合并将多个任务特定模型整合为单一统一模型。近期研究在提升分布内或多任务场景下的合并性能方面取得了进展，但模型合并中的领域泛化问题仍鲜有探索。我们研究了合并在不同领域上微调的模型会如何影响对未见领域的泛化。通过使用奇异值分解分析任务矩阵中的参数竞争，我们表明，与传统的多任务设定相比，合并在不同分布偏移下训练的模型会在其子空间之间引发更强的冲突。为缓解该问题，我们提出 SCORE（子空间冲突消解合并），一种旨在缓解此类奇异子空间冲突的方法。SCORE 通过计算所有模型前导奇异向量拼接后的主成分来寻找共享正交基，随后将每个任务矩阵投影到该共享基上，并剪除非对角分量以去除冲突的奇异方向。在多种架构与模型规模下，SCORE 在领域泛化设定中平均持续优于现有模型合并方法，证明了其有效性与可扩展性。
英文原文摘要 1265 字符
Model merging integrates multiple task-specific models into a single consolidated one. Recent research has made progress in improving merging performance for in-distribution or multi-task scenarios, but domain generalization in model merging remains underexplored. We investigate how merging models fine-tuned on distinct domains affects generalization to unseen domains. Through an analysis of parameter competition in the task matrix using singular value decomposition, we show that merging models trained under different distribution shifts induces stronger conflicts between their subspaces compared to traditional multi-task settings. To mitigate this issue, we propose SCORE (Subspace COnflict-Resolving mErging), a method designed to alleviate such singular subspace conflicts. SCORE finds a shared orthogonal basis by computing the principal components of the concatenated leading singular vectors of all models. It then projects each task matrix into the shared basis, pruning off-diagonal components to remove conflicting singular directions. SCORE consistently outperforms, on average, existing model merging approaches in domain generalization settings across a variety of architectures and model scales, demonstrating its effectiveness and scalability.