Multimodal object detection leverages the complementary characteristics of visible (RGB) and infrared (IR) imagery, making it well-suited for challenging scenarios such as low illumination, occlusion, and complex backgrounds. However, most existing fusion-based methods rely on static or heuristic strategies, limiting their adaptability to dynamic environments. To address this limitation, we propose CLSANet, a cognitive learning-based self-adaptive network that enhances detection performance by dynamically selecting and integrating modality-specific features. CLSANet consists of three key modules: (1) a Dominant Modality Identification Module that selects the most informative modality based on global scene analysis; (2) a Modality Enhancement Module that disentangles and strengthens shared and modality-specific representations; and (3) a Self-Adaptive Fusion Module that adjusts fusion weights spatially according to local scene complexity. Compared to existing methods, CLSANet achieves state-of-the-art detection performance with significantly fewer parameters and lower computational cost. Ablation studies further demonstrate the individual effectiveness of each module under different environmental conditions, particularly in low-light and occluded scenes. CLSANet offers a compact, interpretable, and practical solution for multimodal object detection in resource-constrained settings.

Federated learning has won a lot of interest in recent years, due to its capability in collaborative learning and privacy preservation. To ensure the accuracy of outsourced training tasks, task publishers prefer to assign tasks to task workers with a high reputation. However, existing reputation-based incentive mechanisms assume that task publishers are honest, and only task workers would probably behave dishonestly to pollute the federated learning model. Different from existing work, we argue that task publishers would also behave dishonestly, where they would benefit from colluding with task workers to help task workers obtain a high reputation. In this paper, we propose a collusion-resistant and reliable incentive mechanism for federated learning. First, to measure the credibility of both task publishers and task workers, we devise a novel metric named reliability. Second, we devise a new method to compute the task publisher reliability, which is obtained by computing the deviation of reputation scores given by different task publishers, i.e., low reliability is assigned to a task publisher once its deviation is far away from that of other publishers. Third, we propose a bidirectional reputation calculation method based on the basic uncertain information model to compute reputation and reputation reliability for task workers. Furthermore, by integrating an incentive mechanism, our proposed scheme not only effectively defends against collusion attacks but also ensures that only task workers with high reputation, reputation reliability, and the capability to accomplish complex tasks can win a high reward. Finally, we conduct extensive experiments to verify the efficiency and efficacy of our proposed schemes. The results demonstrate that our proposed schemes are not only collusion-resistant but also achieve 6.31% higher test accuracy compared with the state of the art on the MNIST dataset.

In modern high-demand applications, such as real-time video communication, cloud gaming, and high-definition live streaming, achieving both superior transmission speed and high visual fidelity is paramount. However, unstable networks and packet loss remain major bottlenecks, making accurate and low-latency video error concealment a critical challenge. Traditional error control strategies, such as Forward Error Correction (FEC) and Automatic Repeat Request (ARQ), often introduce excessive latency or bandwidth overhead. Meanwhile, receiver-side concealment methods struggle under high motion or significant packet loss, motivating the exploration of predictive models. SimVP-v2, with its efficient convolutional architecture and Gated Spatiotemporal Attention (GSTA) mechanism, provides a strong baseline by reducing complexity and achieving competitive prediction performance. Despite its merits, SimVP-v2’s reliance on 2D convolutions for implicit temporal aggregation limits its capacity to capture complex motion trajectories and long-term dependencies. This often results in artifacts such as motion blur, detail loss, and accumulated errors. Furthermore, its single-modality design ignores the complementary contextual cues embedded in the audio stream. To overcome these issues, we propose A3DSimVP (Audio- and 3D-Enhanced SimVP-v2), which integrates explicit spatio-temporal modeling with multimodal feature fusion. Architecturally, we replace the 2D depthwise separable convolutions within the GSTA module with their 3D counterparts, introducing a redesigned GSTA-3D module that significantly improves motion coherence across frames. Additionally, an efficient audio–visual fusion strategy supplements visual features with contextual audio guidance, thereby enhancing the model’s robustness and perceptual realism. We validate the effectiveness of A3DSimVP’s improvements through extensive experiments on the KTH dataset. Our model achieves a PSNR of 27.35 dB, surpassing the 27.04 of the SimVP-v2 baseline. Concurrently, our improved A3DSimVP model reduces the loss metrics on the KTH dataset, achieving an MSE of 43.82 and an MAE of 385.73, both lower than the baseline. Crucially, our LPIPS metric is substantially lowered to 0.22. These data tangibly confirm that A3DSimVP significantly enhances both structural fidelity and perceptual quality while maintaining high predictive accuracy. Notably, A3DSimVP attains faster inference speeds than the baseline with only a marginal increase in computational overhead. These results establish A3DSimVP as an efficient and robust solution for latency-critical video applications.