Cryo-electron microscopy (cryo-EM) has achieved nearatomic level resolution of biomolecules by reconstructing 2D micrographs. However, the resolution and accuracy of the reconstructed particles are significantly reduced due to the extremely low signal-to-noise ratio (SNR) and complex noise structure of cryo-EM images. In this paper, we introduce a diffusion model with post-processing framework to effectively denoise and restore single particle cryo-EM images. Our method outperforms the state-of-the-art (SOTA) denoising methods by effectively removing structural noise that has not been addressed before. Additionally, more accurate and high-resolution three-dimensional reconstruction structures can be obtained from denoised cryo-EM images.

Frequent activities in marine energy exploration and transportation have led to the ongoing presence of continuous objects, such as oil spills and radioactive waste, in the ocean. This article focuses on enhancing the understanding of these objects’ boundaries for accurate assessment of their shape, coverage, and evolution. We introduce a continuous object tracking algorithm named JGL-COT, based on joint global-local binary tree topological transformations and specifically designed for underwater acoustic sensor networks. The contribution of JGL-COT lies in its ability to leverage the correlation between the morphologies of a continuous object's boundaries over time, alternating between global and local binary tree topological transformations. When the present boundary features a strong resemblance to its previous form, JGL-COT switches to a local transformation by establishing a semi-infinite region. Otherwise, it transitions to a global transformation. Following this, JGL-COT chooses a group of binary tree-structured cells for boundary mapping, creating virtual boundary nodes as sampling points for boundary fitting. Building upon graph theory, we derive the lower bound on the effectiveness and time complexity of the proposed joint global and local binary tree topological transformations when applied to object boundary tracking. Experiments in both realistic and simulated settings confirm that JGL-COT provides highly accurate tracking and significantly reduces network energy consumption.

The vulnerability of artificial intelligence has emerged as a bottleneck, with adversarial attacks posing a significant threat to natural language processing. Although multiple defense mechanisms have been proposed, they often suffer from strict constraints, weak generalization, and low scalability. To address these challenges, we propose leveraging perplexity to quantify the difference between clean and adversarial examples based on the observation of numerous cases. We then statistically prove the substantial difference between them using Bayesian hypothesis testing. Subsequently, we develop an adversarial defense framework named UMPS, which contains two branches: “Uncovering the Mask”(UM) and “Perplexity-guided Sampling”(PS). UM utilizes a masked language model and Jaro-Winkler distance constraint to recover out-of-vocabulary words, while PS employs perplexity to locate the optimal sample within a convex hull which is constructed with integrated gradients. Theoretically, the proposed framework fulfills three requirements: effectiveness, universality, and portability. The experimental results demonstrate that UMPS effectively enhances the robustness of language models including BERT, against advanced attacks and outperforms three strong baseline methods. Furthermore, we conduct an instance analysis to illustrate how UMPS functions and what it outputs, an ablation study to support the validity and necessity of the two branches, and an post-hoc test on the difference in perplexity to explains the defense performance of our framework.

The research of adversarial attacks in the text domain attracts many interests, and many methods in a black-box scenario with a high attack success rate have been proposed. However, these attack methods are inefficient as they require lots of queries for the victim models when crafting text adversarial examples. In this paper, a novel attack model based on an improved beam search algorithm with different attack strategies in a black-box scenario is proposed, its attack success rate surpasses the benchmark attack methods, but more importantly, it is more efficient than the benchmark attack methods. The novel method is empirically evaluated by attacking WordCNN, LSTM, BiLSTM, and BERT on four benchmark datasets. For instance, it achieves a 100% attack success rate and surpasses all the baseline methods when attacking BERT and BiLSTM on IMDB, but the number of queries for the victim models is much lower than the baseline methods. Also, further experiments show the novel method has a good transferability to the generated adversarial examples.