ACE-Merging：基于自适应协方差估计的无数据模型合并
ACE-Merging: Data-Free Model Merging with Adaptive Covariance Estimation
对视觉研究者而言，这是一种无需原始训练数据即可把多个视觉专家（如不同下游任务的微调模型）合并为单一模型的闭式方法，利用微调更新估计输入协方差来抑制任务间干扰，可直接用于多任务模型部署与权重融合。
原文依据 the input covariance of each task can be estimated directly from its fine-tuning update
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旨在将多个任务特定专家组合成单一模型，但任务间干扰往往导致严重性能退化，尤其当这些专家在异构目标上训练时。现有的无数据方法虽然实用，但大多依赖参数空间启发式规则，未显式建模任务统计特性。本文表明，在局部线性近似下，每个任务的输入协方差可直接由其微调更新估计得到，从而在无数据设定下为参数变化与数据几何之间提供有原则的桥梁。基于这一洞见，我们提出 ACE-Merging，一个具有自适应协方差归一化、集体结构先验和谱精炼的闭式模型合并框架。在视觉与语言基准上，ACE-Merging 在无数据基线中取得强劲的整体性能。特别地，它在 GPT-2 上的平均分数较先前方法提升超过 4 分，同时具备良好的可扩展性和有竞争力的效率。代码见 https://github.com/unravel-xu/ACE-Merging/tree/main。
英文原文摘要 1141 字符
Model merging aims to combine multiple task-specific experts into a single model, but inter-task interference often causes severe degradation, especially when the experts are trained on heterogeneous objectives. Existing data-free methods are practical, yet largely rely on parameter-space heuristics without explicitly modeling task statistics. In this paper, we show that under a local linear approximation, the input covariance of each task can be estimated directly from its fine-tuning update, providing a principled bridge between parameter changes and data geometry in the data-free setting. Based on this insight, we propose ACE-Merging, a closed-form model merging framework with adaptive covariance normalization, a collective structural prior, and spectral refinement. Across both vision and language benchmarks, ACE-Merging achieves strong overall performance among data-free baselines. In particular, it improves the average score on GPT-2 by more than 4 points over prior methods, while also delivering strong scalability and competitive efficiency. Our code is available at https://github.com/unravel-xu/ACE-Merging/tree/main.