BD-Merging：证据引导对比学习的偏差感知动态模型合并
BD-Merging: Bias-Aware Dynamic Model Merging with Evidence-Guided Contrastive Learning
对做多任务视觉模型部署的人，这篇给出了一套在测试时分布偏移下仍可靠的无监督模型合并方案：用证据检测头建模统一标签空间上的不确定性，再用邻域证据差异分数（ADS）驱动对比学习训练去偏路由器，按样本动态分配任务/层权重。可直接复用到多任务检测、分割等模型的免数据合并与测试时自适应场景。
原文依据 we present BD-Merging, a bias-aware unsupervised model merging framework that explicitly models uncertainty to achieve adaptive reliability under distribution shift
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（MM）已成为多任务学习（MTL）的一种可扩展范式，它使多个任务特定模型无需重新访问原始训练数据即可被整合。尽管近期取得了进展，MM 在测试时分布偏移下的可靠性仍未得到充分理解。大多数现有 MM 方法通常假设测试数据是干净的，且与训练数据和辅助来源在分布上对齐。然而这一假设在实践中很少成立，往往导致有偏预测与泛化性能下降。为解决该问题，我们提出 BD-Merging，一个偏差感知的无监督模型合并框架，它显式建模不确定性，以在分布偏移下实现自适应可靠性。首先，BD-Merging 引入联合证据检测头，在统一标签空间上学习不确定性，从而捕获 MM 中的跨任务语义依赖。其次，在此证据基础上，我们提出邻接差异分数（ADS），用于量化相邻样本之间的证据对齐程度。第三，在 ADS 引导下，一种差异感知对比学习机制通过对齐一致样本、分离冲突样本来精炼合并后的表征。结合通用无监督学习，该过程训练出一个去偏路由器，它按样本自适应地分配任务特定或层特定的权重，从而有效缓解分布偏移的不利影响。跨多种任务的广泛实验表明，与最优 MM 基线相比，BD-Merging 取得了更优的有效性与鲁棒性。
英文原文摘要 1641 字符
Model Merging (MM) has emerged as a scalable paradigm for multi-task learning (MTL), enabling multiple task-specific models to be integrated without revisiting the original training data. Despite recent progress, the reliability of MM under test-time distribution shift remains insufficiently understood. Most existing MM methods typically assume that test data are clean and distributionally aligned with both the training and auxiliary sources. However, this assumption rarely holds in practice, often resulting in biased predictions with degraded generalization. To address this issue, we present BD-Merging, a bias-aware unsupervised model merging framework that explicitly models uncertainty to achieve adaptive reliability under distribution shift. First, BD-Merging introduces a joint evidential head that learns uncertainty over a unified label space, capturing cross-task semantic dependencies in MM. Second, building upon this evidential foundation, we propose an Adjacency Discrepancy Score (ADS) that quantifies evidential alignment among neighboring samples. Third, guided by ADS, a discrepancy-aware contrastive learning mechanism refines the merged representation by aligning consistent samples and separating conflicting ones. Combined with general unsupervised learning, this process trains a debiased router that adaptively allocates task-specific or layer-specific weights on a per-sample basis, effectively mitigating the adverse effects of distribution shift. Extensive experiments across diverse tasks demonstrate that BD-Merging achieves superior effectiveness and robustness compared to state-of-the-art MM baselines.