The advancement of deep learning techniques has significantly propelled the development of the continuous sign language recognition (cSLR) task. However, the spatial feature extraction of sign language videos in the RGB space tends to focus on the overall image information while neglecting the perception of traits at different granularities, such as eye gaze and lip shape, which are more detailed, or posture and gestures, which are more macroscopic. Exploring the efficient fusion of visual information of different granularities is crucial for accurate sign language recognition. In addition, applying a vanilla Transformer to sequence modeling in cSLR exhibits weak performance because specific video frames could interfere with the attention mechanism. These limitations constrain the capability to understand potential semantic characteristics. We introduce a feature fusion method for integrating visual features of disparate granularities and refine the metric of attention to enhance the Transformer’s comprehension of video content. Specifically, we extract CNN feature maps with varying receptive fields and employ a self-attention mechanism to fuse feature maps of different granularities, thereby obtaining multi-scale spatial features of the sign language framework. As for video modeling, we first analyze why the vanilla Transformer failed in cSLR and observe that the magnitude of the feature vectors of video frames could interfere with the distribution of attention weights. Therefore, we utilize the Euclidean distance among vectors to measure the attention weights instead of scaled-dot to enhance dynamic temporal modeling capabilities. Finally, we integrate the two components to construct the model MSF-ET (Multi-Scaled feature Fusion–Euclidean Transformer) for cSLR and train the model end-to-end. We perform experiments on large-scale cSLR benchmarks—PHOENIX-2014 and Chinese Sign Language (CSL)—to validate the effectiveness.

Computational pathology (CPath) has significantly advanced the clinical practice of pathology. Despite the progress made, Multiple Instance Learning (MIL), a promising paradigm within CPath, continues to face challenges, particularly related to incomplete information utilization. Existing frameworks, such as those based on Convolutional Neural Networks (CNNs), attention, and selective scan space state sequential model (SSM), lack sufficient flexibility and scalability in fusing diverse features, and cannot effectively fuse diverse features. Additionally, current approaches do not adequately exploit order-related and order-independent features, resulting in suboptimal utilization of sequence information. To address these limitations, we propose a novel MIL framework called Mamba2MIL. Our framework utilizes the state space duality model (SSD) to model long sequences of patches of whole slide images (WSIs), which, combined with weighted feature selection, supports the fusion processing of more branching features and can be extended according to specific application needs. Moreover, we introduce a sequence transformation method tailored to varying WSI sizes, which enhances sequence-independent features while preserving local sequence information, thereby improving sequence information utilization. Extensive experiments demonstrate that Mamba2MIL surpasses state-of-the-art MIL methods. We conducted extensive experiments across multiple datasets, achieving improvements in nearly all performance metrics. Specifically, on the NSCLC dataset, Mamba2MIL achieves a binary tumor classification AUC of 0.9533 and an accuracy of 0.8794. On the BRACS dataset, it achieves a multiclass classification AUC of 0.7986 and an accuracy of 0.4981. The code is available at https://github.com/YuqiZhang-Buaa/Mamba2MIL.

Sign language translation aims to transform continuous sign language videos into text, thereby mapping sign language actions onto natural language. Current research primarily focuses on the Transformer’s encoder-decoder architecture, employing an autoregressive (AR) approach for SLT. This one-by-one token generation method has a slow speed to translate. We design a Non-Autoregressive Sign Language Translation (NA-SLT) model based on Transformer Encoder. The proposed model consists of three components: the text-encoder, sign-encoder, and sign-text-encoder. Both the text-encoder and sign-encoder utilize the Encoder to model single-modality data. The sign-text-encoder effectively engages with multi-modal data, producing high-quality corresponding text after minimal refinement iterations. Moreover, during model training, we devise a double-mask strategy. In addition to traditional random masking, the schedule allocates mask probabilities based on each word’s frequency, ensuring long-tail tokens have an increased chance of being masked and receiving extensive training. We evaluate the model on PHOENIX-2014T, and the results indicate that NA-SLT approaches the performance of AR models on BLEU (21.06) and ROUGE(49.59) metrics while significantly improving translation speed (5.4x-7.9x).