Deepfake technology has great potential in the field of media and entertainment, but it also brings serious risks, including privacy disclosure and identity fraud. To counter these threats, proactive forensic methods have become a research hotspot by embedding invisible watermark signals to build active protection schemes. However, existing methods are vulnerable to watermark destruction under malicious distortions, which leads to insufficient robustness. Moreover, embedding strong signals may degrade image quality, making it challenging to balance robustness and imperceptibility. Although watermarked images look natural, their underlying structures are often different from the original images, which is ignored by traditional watermarking methods. To address these issues, this paper proposes a proactive watermarking framework called WaveGuard, which explores frequency domain embedding and graph-based structural consistency optimization. In this framework, the watermark is embedded into the high-frequency sub-bands by dual-tree complex wavelet transform (DT-CWT) to enhance the robustness against distortions and deepfake forgeries. By leveraging joint sub-band correlations and selected sub-band combinations, the framework enables robust source tracing and semi-robust deepfake detection. To enhance imperceptibility, we propose a Structural Consistency Graph Neural Network (SC-GNN) that constructs graph representations of the original and watermarked images to ensure structural consistency and reduce perceptual artifacts. Experimental results show that the proposed method performs exceptionally well in face swap and face replay tasks. The code has been published at https://github.com/vpsg-research/WaveGuard.

Unstructured data, such as images and texts, are increasingly represented as high-dimensional vectors for emerging AI applications like retrieval-augmented generation. A key operation in these applications is querying for vectors that are both semantically similar and temporally relevant. This operation can be formulated as Timestamp Approximate Nearest Neighbor Search (TANNS), where both the vectors and the query incorporate temporal attributes, aiming to retrieve the approximate nearest neighbors valid at the given timestamp. A naive solution is to create separate indexes for each timestamp, which enables accurate and fast searches but incurs high update latency and excessive storage demands. In this paper, we introduce the timestamp graph, a novel structure that supports rapid index updates while minimizing storage costs. Exploiting the temporal locality of changes in valid vectors, our timestamp graph effectively manages a unified index across all historical timestamps, thereby substantially reducing storage overhead. Moreover, we design the historic neighbor tree, which further compresses the space complexity to that of a single-timestamp index. Extensive evaluations on four standard datasets show that our method achieves over 99% accuracy while improving the query efficiency by 4.4× to 138.1× than existing solutions.

High-precision semantic segmentation is an important research topic in the communities of computer vision and intelligent transportation. The existing unsupervised domain adaptation methods based on image translation often lead to artifacts and structural distortions. To overcome this problem, a novel adaptive semantic segmentation method of traffic scenes via frequency domain analysis is proposed. Firstly, we leverage the frequency domain space to decouple style and semantic features. The Fast Fourier Transform is applied to achieve structural-preserving style alignment. Subsequently, a content enhancement module is proposed based on the Wavelet transform, which utilizes the original source images to correct and enhance high-frequency structural and semantic details. Furthermore, a convolutional enhancement attention module is proposed, which utilizes depthwise separable convolution to capture more local details. The experimental results based on the GTA5→Cityscapes and SYNTHIA→Cityscapes tasks have respectively attained state-of-the-art mIoU of 76.4 and 67.7, convincingly demonstrating the effectiveness of the methods.

Large language models (LLMs) are equipped with increasingly extended context windows recently, yet their long context understanding capabilities over long dependency tasks remain fundamentally limited and underexplored. This gap is especially significant in many real-world long-context applications that were rarely benchmarked. In this paper, we introduce $\textbf{LooGLE v2}$, a novel benchmark designed to evaluate LLMs' long context ability in real-world applications and scenarios. Our benchmark consists of automatically collected real-world long texts, ranging from 16k to 2M tokens, encompassing domains in law, finance, game and code. Accordingly, we delicately design 10 types of domain-specific long-dependency tasks and generate 1,934 QA instances with various diversity and complexity in a scalable data curation pipeline for further practical needs. We conduct a comprehensive assessment of 6 locally deployed and 4 API-based LLMs. The evaluation results show that even the best-performing model achieves only a 59.2\% overall score on our benchmark. Despite the extensive context windows, popular LLMs are only capable of understanding a much shorter length of context than they claim to be, revealing significant limitations in their ability to handle real-world tasks with long dependencies and highlighting substantial room for model improvement in practical long-context understanding.

MOTIVATION: For immune system monitoring in large-scale studies at the single-cell resolution using CyTOF, (semi-)automated computational methods are applied for annotating live cells of mixed cell types. Here, we show that the live cell pool can be highly enriched with undefined heterogeneous cells, i.e. 'ungated' cells, and that current semi-automated approaches ignore their modeling resulting in misclassified annotations. RESULT: We introduce 'CyAnno', a novel semi-automated approach for deconvoluting the unlabeled cytometry dataset based on a machine learning framework utilizing manually gated training data that allows the integrative modeling of 'gated' cell types and the 'ungated' cells. By applying this framework on several CyTOF datasets, we demonstrated that including the 'ungated' cells can lead to a significant increase in the precision of the 'gated' cell types prediction. CyAnno can be used to identify even a single cell type, including rare cells, with higher efficacy than current state-of-the-art semi-automated approaches. AVAILABILITY AND IMPLEMENTATION: The CyAnno is available as a python script with a user-manual and sample dataset at https://github.com/abbioinfo/CyAnno. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.