通过冻结视觉变换器中的电路复制实现推理路径优化用于海洋物种分类
Inference-Path Optimization via Circuit Duplication in Frozen Visual Transformers for Marine Species Classification
AI总结：
本文通过在冻结视觉变换器中应用电路复制方法，提升海洋物种分类的推理性能，无需微调模型权重，在类不平衡数据集上实现宏F1值提升。
摘要：
自动化的水下物种分类受到标注成本和环境变化的限制，限制了全监督模型的可迁移性。最近的研究表明，自监督视觉基础模型的冻结嵌入已经为海洋图像分类提供了强大的标签高效基线。本文探讨了在推理时是否能改进这种冻结嵌入模式，而无需微调或改变模型权重。我们应用了电路复制方法，该方法最初为大语言模型提出，其中在正向传递过程中会两次遍历选定的Transformer层。我们在类不平衡的AQUA20基准上使用冻结的DINOv3嵌入，采用两种设置：全局电路选择，即为整个数据集选择一个重复的电路；以及类特定电路选择，即每个物种可能接收不同的最优电路。两种设置均使用简单的半监督下游分类器。电路复制在标准冻结正向传递上始终表现更好。在最大标签预算下，类特定选择达到宏F1值0.875，接近全监督ConvNeXt基准（0.889）的差距缩小了1.4个百分点，无需任何基于梯度的训练。四个物种超过其全监督参考，其中乌贼的F1值提升了12.1个百分点。在所有预算下，大约75%的类别偏好类特定电路，表明真正的类依赖性收益。据我们所知，这是首次将电路复制应用于计算机视觉。
英文摘要：
Automated underwater species classification is constrained by annotation cost and environmental variation that limits the transferability of fully supervised models. Recent work has shown that frozen embeddings from self-supervised vision foundation models already provide a strong label-efficient baseline for marine image classification. Here we investigate whether this frozen-embedding regime can be improved at inference time, without fine-tuning or changing model weights. We apply Circuit Duplication, an inference-time method originally proposed for Large Language Models, in which a selected range of transformer layers is traversed twice during the forward pass. We evaluate on the class-imbalanced AQUA20 benchmark using frozen DINOv3 embeddings under two settings: global circuit selection, where a single duplicated circuit is chosen for the full dataset, and class-specific circuit selection, where each species may receive a different optimal circuit. Both settings use simple semi-supervised downstream classifiers. Circuit Duplication consistently improves over the standard frozen forward pass. At the maximum label budget, class-specific selection reaches a macro F1 of 0.875, closing the gap to the fully supervised ConvNeXt benchmark (0.889) to 1.4 points without any gradient-based training. Four species exceed their fully supervised reference, with octopus improving by +12.1 F1 points. Across all budgets, roughly 75% of classes prefer a class-specific circuit, indicating a genuinely class-dependent benefit. To our knowledge, this is the first application of Circuit Duplication to computer vision.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
