在本文中，我们研究了一个可以说是最少的限制设置的领域自适应在实际部署的意义上，只有源模型的接口是可用的目标域，并允许两个域之间的标签空间关系是不同的和未知的。我们将这样的设置称为通用黑盒领域自适应（UB $^2$ DA）。然而，由于领域自适应只能通过访问源模型的接口，依赖于部分重叠标签空间中未标记目标数据的预测，因此UB $^2$ DA带来了巨大的学习挑战。为了应对这些挑战，我们首先注意到学习任务可以转换为类内\footnote{In this paper we use in-class (out-class) to describe the classes observed (not observed) in the source black-box model.}判别和类外检测两个子任务，它们可以分别通过模型蒸馏和熵分离来学习。我们建议将它们统一到一个自训练框架中，通过目标样本局部邻域中预测的一致性进行正则化。我们的框架是简单的，强大的，易于优化。在领域自适应基准上的实验表明了该方法的有效性。值得注意的是，通过仅访问源模型的接口，我们的框架优于现有的通用领域自适应方法，这些方法利用源数据和/或源模型，具有新提出的（并且可以说是更合理的）H分数度量，并且与平均类准确度度量的性能相当。
英文摘要：
In this paper, we study an arguably least restrictive setting of domain adaptation in a sense of practical deployment, where only the interface of source model is available to the target domain, and where the label-space relations between the two domains are allowed to be different and unknown. We term such a setting as Universal Black-Box Domain Adaptation (UB$^2$DA). The great promise that UB$^2$DA makes, however, brings significant learning challenges, since domain adaptation can only rely on the predictions of unlabeled target data in a partially overlapped label space, by accessing the interface of source model. To tackle the challenges, we first note that the learning task can be converted as two subtasks of in-class\footnote{In this paper we use in-class (out-class) to describe the classes observed (not observed) in the source black-box model.} discrimination and out-class detection, which can be respectively learned by model distillation and entropy separation. We propose to unify them into a self-training framework, regularized by consistency of predictions in local neighborhoods of target samples. Our framework is simple, robust, and easy to be optimized. Experiments on domain adaptation benchmarks show its efficacy. Notably, by accessing the interface of source model only, our framework outperforms existing methods of universal domain adaptation that make use of source data and/or source models, with a newly proposed (and arguably more reasonable) metric of H-score, and performs on par with them with the metric of averaged class accuracy.
Prototypical Distillation and Debiased Tuning for Black-box Unsupervised Domain Adaptation
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；University of Science and Technology of China（中国科学技术大学）；School of Intelligence Science and Technology, University of Science and Technology Beijing（北京科技大学智能科学与技术学院）
Adaptive Dual-Teacher Distillation with Subnetwork Rectification for Bridging Semantic Gaps in Black-Box Domain Adaptation
School of Computer Science and Engineering, Tianjin University of Technology（天津理工大学计算机科学与工程学院）；School of Artificial Intelligence, Tianjin University（天津大学人工智能学院）
Box Pose and Shape Estimation and Domain Adaptation for Large-Scale Warehouse Automation
Laboratory for Information & Decision Systems (LIDS)（信息与决策系统实验室）；Massachusetts Institute of Technology（麻省理工学院）；Symbotic（Symbotic公司）；Vector Institute University of Toronto（多伦多大学向量研究所）
Joint Distribution Alignment for Universal Domain Adaptation
School of Software Engineering, South China University of Technology（华南理工大学软件工程学院）；School of Future Technology, South China University of Technology（华南理工大学未来技术学院）；College of Computer Science, Chongqing University（重庆大学计算机学院）
Unsupervised Domain Adaptation for Segmentation with Black-box Source Model
Gordon Center for Medical Imaging, Massachusetts General Hospital and Harvard Medical School（戈登医学影像中心，麻省总医院与哈佛医学院）；Ewha Womans University（梨花女子大学）；University of Southern California（南加州大学）
Black-box Source-free Domain Adaptation via Two-stage Knowledge Distillation
Tsinghua University（清华大学）；Southern University of Science and Technology（南方科技大学）；The Hong Kong Polytechnic University（香港理工大学）；Southeast University（东南大学）
AdaptGuard: Defending Against Universal Attacks for Model Adaptation
University of Science and Technology of China（中国科学技术大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Nanjing University（南京大学）
Pseudo Label Refinery for Unsupervised Domain Adaptation on Cross-dataset 3D Object Detection
State Key Lab of CAD&CG, Zhejiang University（浙江大学CAD&CG国家重点实验室）；Fabu Inc.（Fabu公司）；Alibaba Group（阿里巴巴集团）；School of Software Technology, Zhejiang University（浙江大学软件学院）；School of Computer Sciene and Technology, Hangzhou Dianzi University（杭州电子科技大学计算机科学与技术学院）；Fullong Inc.（Fullong公司）
本文针对跨数据集三维目标检测的无监督域自适应中伪标签不可靠问题，提出伪标签精炼框架，通过互补增强策略和跨域 RoI 特征对齐提升伪标签质量，在六个自动驾驶基准上超越现有方法。
Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias
National University of Singapore (NUS)（新加坡国立大学）；Centre for Frontier AI Research (CFAR), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局前沿人工智能研究中心）；Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局信息通信研究所）
Target Semantics Clustering via Text Representations for Robust Universal Domain Adaptation
University of Science and Technology of China（中国科学技术大学）；Institute of Artificial Intelligence, Hefei Comprehensive National Science Center（合肥综合性国家科学中心人工智能研究院）
Memory-Assisted Sub-Prototype Mining for Universal Domain Adaptation
School of Computer Science and Engineering, Southeast University（东南大学计算机科学与工程学院）；School of Computer Science and Engineering, Nanjing University of Science and Technology（南京理工大学计算机科学与工程学院）
