SEATrack：简单、高效且自适应的多模态跟踪器
SEATrack: Simple, Efficient, and Adaptive Multimodal Tracker
AMG-LoRA 将 LoRA 与自适应互引导结合，用于对齐双流跟踪器中的跨模态注意力图；HMoE 用分层混合专家实现高效的全局关系建模，这两者都是可复用的 PEFT 与融合设计，可直接迁移到其他多模态视觉任务。
原文依据 Parameter-efficient fine-tuning (PEFT) in multimodal tracking reveals a concerning trend where recent performance gains are often achieved at the cost of inflated parameter budgets
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论多模态跟踪中的参数高效微调（PEFT）呈现出一种令人担忧的趋势：近期的性能提升往往以膨胀的参数预算为代价，这从根本上侵蚀了 PEFT 的效率承诺。在本工作中，我们提出 SEATrack，一种简单、高效且自适应的双流多模态跟踪器，从两个互补的视角解决这一性能与效率的困境。我们首先优先考虑匹配响应的跨模态对齐，这是一个尚未被充分探索但至关重要的因素，我们认为它对于打破这一权衡至关重要。具体而言，我们观察到现有双流方法中模态特定的偏差会产生相互冲突的匹配注意力图，从而阻碍有效的联合表征学习。为缓解这一问题，我们提出 AMG-LoRA，它将用于域自适应的低秩适配（LoRA）与自适应互引导（AMG）无缝集成，以动态地细化和对齐跨模态的注意力图。随后，我们偏离传统的局部融合方法，引入分层混合专家（HMoE），以实现高效的全局关系建模，在跨模态融合中有效平衡表达能力和计算效率。凭借这些创新，SEATrack 在 RGB-T、RGB-D 和 RGB-E 跟踪任务上，在平衡性能与效率方面相较于最优方法取得了显著进展。代码已公开。
英文原文摘要 1443 字符
Parameter-efficient fine-tuning (PEFT) in multimodal tracking reveals a concerning trend where recent performance gains are often achieved at the cost of inflated parameter budgets, which fundamentally erodes PEFT's efficiency promise. In this work, we introduce SEATrack, a Simple, Efficient, and Adaptive two-stream multimodal tracker that tackles this performance-efficiency dilemma from two complementary perspectives. We first prioritize cross-modal alignment of matching responses, an underexplored yet pivotal factor that we argue is essential for breaking the trade-off. Specifically, we observe that modality-specific biases in existing two-stream methods generate conflicting matching attention maps, thereby hindering effective joint representation learning. To mitigate this, we propose AMG-LoRA, which seamlessly integrates Low-Rank Adaptation (LoRA) for domain adaptation with Adaptive Mutual Guidance (AMG) to dynamically refine and align attention maps across modalities. We then depart from conventional local fusion approaches by introducing a Hierarchical Mixture of Experts (HMoE) that enables efficient global relation modeling, effectively balancing expressiveness and computational efficiency in cross-modal fusion. Equipped with these innovations, SEATrack advances notable progress over state-of-the-art methods in balancing performance with efficiency across RGB-T, RGB-D, and RGB-E tracking tasks. Code is available.