InsCal：面向目标检测的校准式多源全测试时提示词调优
InsCal: Calibrated Multi-Source Fully Test-Time Prompt Tuning for Object Detection
把测试时提示词调优从分类扩展到文本驱动的开放词表检测器，并用实例级校准修正熵最小化带来的误校准，可直接用于跨域检测的免训练适配。
原文依据 we extend TPT to object detection, unlocking new capabilities for cross-domain adaptation
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论测试时提示词调优（TPT）已成为将预训练视觉语言模型（VLM）适配到图像分类、视觉推理等多种下游任务的有力技术。随着文本驱动目标检测器的兴起，我们将 TPT 扩展到目标检测，为跨域适配开辟了新的能力。然而，TPT 中的一个关键挑战是熵最小化所固有的误校准：域偏移往往导致错误预测，而强制高置信度会加剧误校准，最终降低性能。为解决这一问题，我们提出 InsCal，一个通过三项关键创新来增强跨域目标检测的新框架：（1）将 TPT 扩展到多源范式，实现跨多个领域的知识聚合；（2）通过一种新的文本驱动风格迁移策略缩小域差距，该策略无需参考图像即可将特征对齐到源域；（3）以实例级校准改进熵最小化目标，确保稳健且良好校准的适配。我们的方法不仅缓解了误校准，还显著提升了跨域目标检测性能，为 VLM 中的测试时自适应树立了新的基准。
英文原文摘要 1287 字符
Test-time prompt tuning (TPT) has emerged as a powerful technique for adapting pre-trained vision-language models (VLMs) to diverse downstream tasks, including image classification and visual reasoning. With the rise of text-driven object detectors, we extend TPT to object detection, unlocking new capabilities for cross-domain adaptation. However, a critical challenge in TPT is the inherent miscalibration caused by entropy minimization: domain shifts often lead to incorrect predictions, and enforcing high confidence exacerbates miscalibration, ultimately degrading performance. To tackle this, we introduce InsCal, a novel framework designed to enhance cross-domain object detection through three key innovations: (1) extending TPT to a multi-source paradigm, enabling knowledge aggregation across diverse domains; (2) reducing domain gaps via a novel text-driven style transfer strategy that aligns features to the source domain without requiring reference images; and (3) refining the entropy minimization objective with instance-specific calibration, ensuring robust and well-calibrated adaptation. Our approach not only mitigates miscalibration but also significantly improves cross-domain object detection performance, setting a new benchmark for test-time adaptation in VLMs.