Federated learning is a distributed learning approach designed to train models across multiple client devices with local data while preserving the privacy of that data. However, traditional federated learning suffers from three major f laws: 1.Vulnerability to Byzantine Attacks. 2.Privacy Leakage Risks. 3.Assumption of Third-Party Server Reliability. Few solutions currently address all three of these flaws at the same time. To address the above issues simultaneously, this paper introduces a multi-server federated learning approach that enhances security and privacy. We first set up a multi-server architecture to address the performance issues and single point of failure in traditional federated learning frameworks. And we set up and update different servers reputation on both the server and participant sides, which work together to resist aggregation attacks that can be implemented by malicious servers. It is also worth noting that we split models to be uploaded into two segments, each of which is uploaded to two different servers in conjunction with the last model of the previous round to defend against privacy attacks and keep model as complete as possible. Experiments on MNIST, IMDB, CIFAR10, and Adult indicate this method maintains accuracy, defends against label-flipping and Gaussian noise, and protects privacy.

The rapid advancement of artificial intelligence has propelled the healthcare industry into a new era of diagnostic precision. A pivotal component of this evolution is the accurate classification of 3D medical images, which necessitates extracting robust feature representations capable of effectively modeling long-range dependencies within the data. This paper introduces the Mamba Token Turing Machine (MTTM), a novel architecture that integrates the efficiency of Mamba with the memory mechanisms of the Token Turing Machine (TTM), effectively addressing limitations of Transformers in long-range dependency modeling. MTTM’s Memory-Augmented Processing Unit (MAPU) employs four blending methods, achieving state-of-the-art accuracy and efficiency on the MedMNIST v2 dataset, thereby advancing diagnostic precision in 3D medical image analysis. The code is available at https://github.com/hongkai-wei/MTTM.

Monocular 3D Visual Grounding (Mono3DVG) aims to predict the 3D localization of objects in monocular RGB images based on natural language descriptions. This task has broad applications in areas such as autonomous driving, human-computer interaction and robotic manipulation, making it both a significant and challenging task. To tackle this challenge, we propose a novel network architecture, the Monocular Selective Attention Learning Network (MSALNet). This network enhances the understanding and localization of objects by introducing an Adaptive Learning Module (ALM) and a Vision-Text Interaction Encoder. Specifically, the ALM further learns the features extracted from the 3D scene and textual descriptions, capturing the contextual relationships within the features. This enables the model to better understand the meaning of both the scene and the text. Meanwhile, the Vision-Text Interaction Encoder facilitates refined cross-modal interaction and fusion, promoting alignment between visual and textual information and providing more discriminative feature representations. Experimental results demonstrate that our method achieves competitive performance on the Mono3DRefer dataset.

Image-Text Retrieval (ITR) serves as a fundamental task in multimodal information retrieval, aiming to bridge the semantic gap between visual and textual modalities. The core challenge resides in overcoming intrinsic modality discrepancies while effectively capturing cross-modal semantic relationships. Fine-grained retrieval approaches can model associations between salient visual regions and their linguistic counterparts, but are prone to interference from irrelevant visual embeddings and non-predicative textual components, leading to misalignments and degraded embedding quality for cross-modal retrieval. To address these challenges, we propose a Multi-Granularity Selective Graph Mamba (MGSGM) framework for discriminative embeddings. It uses the Selective Relationship Reasoning Module (SRRM) to suppress intra-modal redundancy, followed by graph refinement for medium-grained feature extraction. The Cross-modal Mamba Alignment (CMA) establishes selective semantic bridges between inter-modal entities. A hierarchical fusion mechanism ultimately integrates multi-granularity representations into discriminative embeddings. Extensive experiments on Flickr30K and MS-COCO datasets demonstrate the superiority of our method.