Vegetation protection is a crucial task in agricultural production. Due to the diversity of tree species and their varied growth conditions, the demand for protective ash spray varies greatly, posing multifaceted challenges of experience and technology for manual operations. This paper proposes a framework that integrates an attention mechanism with RetNet for predicting tree growth conditions. It extracts and fuses core visual features from both tree species and growth conditions to achieve high-quality classification of tree growth status, thereby guiding the concentration of ash spray. Quantitative experimental results demonstrate that the proposed approach outperforms the baseline in accuracy, recall, and other metrics. These results indicate that the proposed framework is helpful for the optimization of vegetation protection workflow in the context of human-machine collaboration.

With the rapid development of Artificial Intelligence and Human-Computer Interaction technology, the quality of agricultural production is significantly enhanced. This paper presents a meta-study on expert knowledge-based interactive smart agricultural systems from the vantage point of human-machine collaboration. By means of quantitative and trend analysis, this study identifies the research challenges, dominant technologies, and data sources in this field, as well as the promotion of agricultural production. The findings are intended to provide valuable insights for the construction of expert knowledge-based interactive, distributed, and collaborative smart agricultural systems. Through rigorous quantitative and trend analyses, this investigation discerns the prevailing research impediments, prevailing technological paradigms, and pivotal data repositories within the domain, in addition to their implications for agricultural enhancement, offering substantive guidance for the advancements of expert knowledge-based smart agricultural systems.

Backdoor attacks have posed a significant threat to the security of deep neural networks (DNNs). Despite considerable strides in developing defenses against backdoor attacks in the visual domain, the specialized defenses for the audio domain remain empty. Furthermore, the defenses adapted from the visual to audio domain demonstrate limited effectiveness. To fill this gap, we propose Gradient Norm-based Fine-Tuning (GN-FT), a novel defense strategy against the attacks in the audio domain, based on the observation from the corresponding backdoored models. Specifically, we first empirically find that the backdoored neurons exhibit greater gradient values compared to other neurons, while clean neurons stay the lowest. On this basis, we fine-tune the backdoored model by incorporating the gradient norm regularization, aiming to weaken and reduce the backdoored neurons. We further approximate the loss computation for lower implementation costs. Extensive experiments on two speech recognition datasets across five models demonstrate the superior performance of our proposed method. To the best of our knowledge, this work is the first specialized and effective defense against backdoor attacks in the audio domain.

Backdoor learning is a critical research topic for understanding the vulnerabilities of deep neural networks. While the diffusion model (DM) has been broadly deployed in public over the past few years, the understanding of its backdoor vulnerability is still in its infancy compared to the extensive studies in discriminative models. Recently, many different backdoor attack and defense methods have been proposed for DMs, but a comprehensive benchmark for backdoor learning on DMs is still lacking. This absence makes it difficult to conduct fair comparisons and thoroughly evaluate existing approaches, thus hindering future research progress. To address this issue, we propose BackdoorDM, the first comprehensive benchmark designed for backdoor learning on DMs. It comprises nine state-of-the-art (SOTA) attack methods, four SOTA defense strategies, and three useful visualization analysis tools. We first systematically classify and formulate the existing literature in a unified framework, focusing on three different backdoor attack types and five backdoor target types, which are restricted to a single type in discriminative models. Then, we systematically summarize the evaluation metrics for each type and propose a unified backdoor evaluation method based on multimodal large language model (MLLM). Finally, we conduct a comprehensive evaluation and highlight several important conclusions. We believe that BackdoorDM will help overcome current barriers and contribute to building a trustworthy artificial intelligence generated content (AIGC) community. The codes are released in https://github.com/linweiii/BackdoorDM.