发表机构
Beijing Institute of Technology（北京理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出免训练框架CQTR，通过反事实尺度干预激活冻结检测器中的潜在小目标知识，利用解码器轨迹评估可靠性，在27种组合上一致提升AP和APs。
AI 中文摘要
小目标检测仍然具有挑战性，因为有限的像素导致信息丢失，并抑制了预训练检测器中编码的尺度知识。现有方法主要通过多尺度训练、架构重新设计或参数自适应来改进表示，隐含地假设冻结模型缺乏所需能力。我们挑战这一假设，并推测小目标知识已存在于冻结检测器中，但在查询演化过程中未被充分激活且不稳定。为验证这一假设，我们提出反事实查询轨迹可靠性（CQTR），一个免训练框架，通过反事实尺度干预激发潜在响应，并从解码器内部的空间收敛性、语义持久性和跨尺度冲突中解释候选可靠性。一个小的无标注训练子集为每个模型-数据流选择适当的校正机制，无需参数更新或目标域标注。在九个冻结检测器与三个数据集的27种组合中，CQTR一致地提高了平均精度（AP）和小目标平均精度（APs）。闭环分析进一步表明，尺度干预激活潜在响应，轨迹证据预测真实标注支持，无标注路由选择更有效的分支。因此，CQTR将小目标检测从外部尺度增强重新定义为潜在尺度知识的激活与可靠性评估。
英文摘要
Small-object detection remains challenging because limited pixels cause information loss and suppress the scale knowledge encoded in pretrained detectors. Existing approaches mainly improve representations through multiscale training, architecture redesign, or parameter adaptation, implicitly assuming that frozen models lack the required capability. We challenge this assumption and hypothesize that small-object knowledge already exists in frozen detectors but remains underactivated and unstable during query evolution. To test this hypothesis, we propose Counterfactual Query-Trajectory Reliability (CQTR), a training-free framework that elicits latent responses through counterfactual scale interventions and interprets candidate reliability from decoder-internal spatial convergence, semantic persistence, and cross-scale conflicts. A small unlabeled training subset selects the appropriate correction mechanism for each model-data stream, without parameter updates or target-domain annotations. Across 27 combinations of nine frozen detectors and three datasets, CQTR consistently improves average precision (AP) and average precision for small objects (APs). Closed-loop analyses further show that scale intervention activates latent responses, trajectory evidence predicts ground-truth support, and unlabeled routing selects the more effective branch. CQTR therefore reframes small-object detection from external scale augmentation to the activation and reliability assessment of latent scale knowledge.
