Xiao Yang

Kunpeng Song, Yizhe Zhu, Bingchen Liu et al.

In this paper, we present MoMA: an open-vocabulary, training-free personalized image model that boasts flexible zero-shot capabilities. As foundational text-to-image models rapidly evolve, the demand for robust image-to-image translation grows. Addressing this need, MoMA specializes in subject-driven personalized image generation. Utilizing an open-source, Multimodal Large Language Model (MLLM), we train MoMA to serve a dual role as both a feature extractor and a generator. This approach effectively synergizes reference image and text prompt information to produce valuable image features, facilitating an image diffusion model. To better leverage the generated features, we further introduce a novel self-attention shortcut method that efficiently transfers image features to an image diffusion model, improving the resemblance of the target object in generated images. Remarkably, as a tuning-free plug-and-play module, our model requires only a single reference image and outperforms existing methods in generating images with high detail fidelity, enhanced identity-preservation and prompt faithfulness. Our work is open-source, thereby providing universal access to these advancements.

Shilong Liu, Lei Zhang, Xiao Yang et al.

This paper presents a simple and effective approach to solving the multi-label classification problem. The proposed approach leverages Transformer decoders to query the existence of a class label. The use of Transformer is rooted in the need of extracting local discriminative features adaptively for different labels, which is a strongly desired property due to the existence of multiple objects in one image. The built-in cross-attention module in the Transformer decoder offers an effective way to use label embeddings as queries to probe and pool class-related features from a feature map computed by a vision backbone for subsequent binary classifications. Compared with prior works, the new framework is simple, using standard Transformers and vision backbones, and effective, consistently outperforming all previous works on five multi-label classification data sets, including MS-COCO, PASCAL VOC, NUS-WIDE, and Visual Genome. Particularly, we establish $91.3\%$ mAP on MS-COCO. We hope its compact structure, simple implementation, and superior performance serve as a strong baseline for multi-label classification tasks and future studies. The code will be available soon at https://github.com/SlongLiu/query2labels.

Xiao Yang, Roland Kwitt, Martin Styner et al.

This paper introduces Quicksilver, a fast deformable image registration method. Quicksilver registration for image-pairs works by patch-wise prediction of a deformation model based directly on image appearance. A deep encoder-decoder network is used as the prediction model. While the prediction strategy is general, we focus on predictions for the Large Deformation Diffeomorphic Metric Mapping (LDDMM) model. Specifically, we predict the momentum-parameterization of LDDMM, which facilitates a patch-wise prediction strategy while maintaining the theoretical properties of LDDMM, such as guaranteed diffeomorphic mappings for sufficiently strong regularization. We also provide a probabilistic version of our prediction network which can be sampled during the testing time to calculate uncertainties in the predicted deformations. Finally, we introduce a new correction network which greatly increases the prediction accuracy of an already existing prediction network. We show experimental results for uni-modal atlas-to-image as well as uni- / multi- modal image-to-image registrations. These experiments demonstrate that our method accurately predicts registrations obtained by numerical optimization, is very fast, achieves state-of-the-art registration results on four standard validation datasets, and can jointly learn an image similarity measure. Quicksilver is freely available as an open-source software.

Tingbing Yan, Wenzheng Zeng, Yang Xiao et al.

Most existing one-shot skeleton-based action recognition focuses on raw low-level information (e.g., joint location), and may suffer from local information loss and low generalization ability. To alleviate these, we propose to leverage text description generated from large language models (LLM) that contain high-level human knowledge, to guide feature learning, in a global-local-global way. Particularly, during training, we design $2$ prompts to gain global and local text descriptions of each action from an LLM. We first utilize the global text description to guide the skeleton encoder focus on informative joints (i.e.,global-to-local). Then we build non-local interaction between local text and joint features, to form the final global representation (i.e., local-to-global). To mitigate the asymmetry issue between the training and inference phases, we further design a dual-branch architecture that allows the model to perform novel class inference without any text input, also making the additional inference cost neglectable compared with the base skeleton encoder. Extensive experiments on three different benchmarks show that CrossGLG consistently outperforms the existing SOTA methods with large margins, and the inference cost (model size) is only $2.8$\% than the previous SOTA. CrossGLG can also serve as a plug-and-play module that can substantially enhance the performance of different SOTA skeleton encoders with a neglectable cost during inference. The source code will be released soon.

Xiao Yang, Dengbo He, Jiyao Wang et al.

Remote physiological measurement gained wide attention, while it requires collecting users' privacy-sensitive information, and existing contactless measurements still rely on labeled client data. This presents challenges when we want to further update real-world deployed models with numerous user data lacking labels. To resolve these challenges, we instantiate a new protocol called Federated Unsupervised Domain Generalization (FUDG) in this work. Subsequently, the \textbf{Fed}erated \textbf{H}eterogeneous \textbf{U}nsupervised \textbf{G}eneralization (\textbf{FedHUG}) framework is proposed and consists of: (1) Minimal Bias Aggregation module dynamically adjusts aggregation weights based on prior-driven bias evaluation to cope with heterogeneous non-IID features from multiple domains. (2) The Global Distribution-aware Learning Controller parameterizes the label distribution and dynamically manipulates client-specific training strategies, thereby mitigating the server-client label distribution skew and long-tail issue. The proposal shows superior performance across state-of-the-art techniques in estimation with either RGB video or mmWave radar. The code will be released.

Zonglei Jing, Xiao Yang, Xiaoqian Li et al.

Text-to-image (T2I) models have demonstrated remarkable generative capabilities but remain vulnerable to producing not-safe-for-work (NSFW) content, such as violent or explicit imagery. While recent moderation efforts have introduced soft prompt-guided tuning by appending defensive tokens to the input, these approaches often rely on large-scale curated image-text datasets and apply static, one-size-fits-all defenses at inference time. However, this results not only in high computational cost and degraded benign image quality, but also in limited adaptability to the diverse and nuanced safety requirements of real-world prompts. To address these challenges, we propose PromptSafe, a gated prompt tuning framework that combines a lightweight, text-only supervised soft embedding with an inference-time gated control network. Instead of training on expensive image-text datasets, we first rewrite unsafe prompts into semantically aligned but safe alternatives using an LLM, constructing an efficient text-only training corpus. Based on this, we optimize a universal soft prompt that repels unsafe and attracts safe embeddings during the diffusion denoising process. To avoid over-suppressing benign prompts, we introduce a gated mechanism that adaptively adjusts the defensive strength based on estimated prompt toxicity, thereby aligning defense intensity with prompt risk and ensuring strong protection for harmful inputs while preserving benign generation quality. Extensive experiments across multiple benchmarks and T2I models show that PromptSafe achieves a SOTA unsafe generation rate (2.36%), while preserving high benign fidelity. Furthermore, PromptSafe demonstrates strong generalization to unseen harmful categories, robust transferability across diffusion model architectures, and resilience under adaptive adversarial attacks, highlighting its practical value for safe and scalable deployment.

Xiao Yang, Daren Sun, Ruiwei Zhu et al.

Sponsored search has more than 20 years of history, and it has been proven to be a successful business model for online advertising. Based on the pay-per-click pricing model and the keyword targeting technology, the sponsored system runs online auctions to determine the allocations and prices of search advertisements. In the traditional setting, advertisers should manually create lots of ad creatives and bid on some relevant keywords to target their audience. Due to the huge amount of search traffic and a wide variety of ad creations, the limits of manual optimizations from advertisers become the main bottleneck for improving the efficiency of this market. Moreover, as many emerging advertising forms and supplies are growing, it's crucial for sponsored search platform to pay more attention to the ROI metrics of ads for getting the marketing budgets of advertisers. In this paper, we present the AiAds system developed at Baidu, which use machine learning techniques to build an automated and intelligent advertising system. By designing and implementing the automated bidding strategy, the intelligent targeting and the intelligent creation models, the AiAds system can transform the manual optimizations into multiple automated tasks and optimize these tasks in advanced methods. AiAds is a brand-new architecture of sponsored search system which changes the bidding language and allocation mechanism, breaks the limit of keyword targeting with end-to-end ad retrieval framework and provides global optimization of ad creation. This system can increase the advertiser's campaign performance, the user experience and the revenue of the advertising platform simultaneously and significantly. We present the overall architecture and modeling techniques for each module of the system and share our lessons learned in solving several key challenges.

Zhipeng Ding, Greg Fleishman, Xiao Yang et al.

Deformable image registration and regression are important tasks in medical image analysis. However, they are computationally expensive, especially when analyzing large-scale datasets that contain thousands of images. Hence, cluster computing is typically used, making the approaches dependent on such computational infrastructure. Even larger computational resources are required as study sizes increase. This limits the use of deformable image registration and regression for clinical applications and as component algorithms for other image analysis approaches. We therefore propose using a fast predictive approach to perform image registrations. In particular, we employ these fast registration predictions to approximate a simplified geodesic regression model to capture longitudinal brain changes. The resulting method is orders of magnitude faster than the standard optimization-based regression model and hence facilitates large-scale analysis on a single graphics processing unit (GPU). We evaluate our results on 3D brain magnetic resonance images (MRI) from the ADNI datasets.

Xiao Yang, Roland Kwitt, Marc Niethammer

We present a method to predict image deformations based on patch-wise image appearance. Specifically, we design a patch-based deep encoder-decoder network which learns the pixel/voxel-wise mapping between image appearance and registration parameters. Our approach can predict general deformation parameterizations, however, we focus on the large deformation diffeomorphic metric mapping (LDDMM) registration model. By predicting the LDDMM momentum-parameterization we retain the desirable theoretical properties of LDDMM, while reducing computation time by orders of magnitude: combined with patch pruning, we achieve a 1500x/66x speed up compared to GPU-based optimization for 2D/3D image registration. Our approach has better prediction accuracy than predicting deformation or velocity fields and results in diffeomorphic transformations. Additionally, we create a Bayesian probabilistic version of our network, which allows evaluation of deformation field uncertainty through Monte Carlo sampling using dropout at test time. We show that deformation uncertainty highlights areas of ambiguous deformations. We test our method on the OASIS brain image dataset in 2D and 3D.