Deep neural networks (DNNs) are known to be vulnera-ble to both backdoor attacks as well as adversarial attacks. In the literature, these two types of attacks are commonly treated as distinct problems and solved separately, since they belong to training-time and inference-time attacks respectively. However, in this paper we find an intriguing connection between them: for a model planted with backdoors, we observe that its adversarial examples have similar behaviors as its triggered images, i.e., both activate the same subset of DNN neurons. It indicates that planting a back-door into a model will significantly affect the model's adversarial examples. Based on these observations, a novel Progressive Backdoor Erasing (PBE) algorithm is proposed to progressively purify the infected model by leveraging un-targeted adversarial attacks. Different from previous back-door defense methods, one significant advantage of our approach is that it can erase backdoor even when the clean extra dataset is unavailable. We empirically show that, against 5 state-of-the-art backdoor attacks, our PBE can effectively erase the backdoor without obvious performance degradation on clean samples and outperforms existing de-fense methods.

Urban-rural integration (URI) is a complex system engineering problem involving the interaction of multiple aspects of population-land-industry. Identifying the driving mechanisms of the drivers of urban-rural integration development is crucial for predicting the future trends of urban-rural integration and identifying the key promotion paths. In this study, 27 representative factors were synthetically identified using literature research and expert interview methods. Then, the explanatory structural model (ISM) is combined with the Impact Matrix Cross-Reference Multiplication Applied to a Classification (MICMAC), which can be used to analyze and model the complex relationships and interactions in URI systems, to establish integrated ISM- MICMAC model. The model was applied to quantify the causal driving relationships among factors and to develop a multilevel recursive structural model for understanding the underlying mechanisms. The analysis revealed that all 27 drivers can be classified into seven layers. Four source factors such as land use planning and policies, three outcome factors such as urban and rural residents' income, and 20 process factors such as population size are included. And accordingly, corresponding suggestions are made for the government. This study contributes to the advancement of theoretical and model development in understanding the driving mechanism of urban-rural integration. Moreover, it provides valuable insights for policymakers in formulating evidence-based development strategies.

Comprehensive land consolidation (CLC) is a more integrated and sustainable land management model that can better meet the requirements of socio-economic development and ecological environmental protection. However, CLC involves multiple fields, stakeholders, and complex socio-economic environments, which make it challenging to grasp the main points and difficulties in understanding CLC, resulting in insufficient scientific and targeted work. Therefore, identifying the key influencing factors (KIFs) of CLC is crucial for its successful implementation. This study adopts a multivariate system theory perspective and identifies 43 factors that can be categorized into three groups: consolidation subject, consolidation element, and consolidation system and environment. Furthermore, the Hamming distance optimization traditional fuzzy best and worst method (FBWM) is introduced to identify eight KIFs, including A6 comprehensive quality of professionals, A7 importance of local government, B9 financial support, B18 land use planning, B20 comprehensive consolidation cost, C2 benefit distribution mechanism, C5 whole process supervision and control mechanism, and C8 multi-departmental coordination mechanism. This study provides a new theoretical framework and methodology for research on the influencing factors of CLC, and helps policy makers and decision makers understand the interrelationships and influences of different factors in the complex consolidation work, clarifying the management priorities for CLC implementation. This can lead to the formulation of more scientific and reasonable policies and measures to promote the sustainable use and comprehensive management of land resources. Moreover, the introduction of Hamming distance optimization to the traditional FBWM enhances the reliability and scientific rigor of the research results, and can serve as a reference for similar studies.

As an organic whole that consists of multiple links and subjects coupled and coordinated, the introduction of the platform economy with network synergy is of great significance to the upgrading of the agricultural industry chain, so it is important to investigate the key influencing factors (KIFs) of the platform economy to enhance the value of the whole agricultural industry chain. Through literature research, meta-analysis and expert interviews, this study identifies 34 initial influencing factors for the value enhancement of the whole agricultural industry chain enabled by the platform economy, and divides the initial influencing factors into three dimensions based on the general system theory idea, namely, industry chain link elements, industry chain participating subjects, and industry chain system and environment. After that, we constructed the key influencing factors identification model based on Fuzzy-AHP and Fuzzy-DEMATEL methods and finally identified seven key influencing factors, which are intelligent management of production, organization of agricultural production, digital literacy and skills of agricultural subjects, cooperation willingness and synergy ability of various subjects, rural digital infrastructure construction, agricultural data and information sharing system, and Legal regulations and policy support, and corresponding policy suggestions are put forward accordingly. This study not only enriches the theory of platform economy development in agriculture, but also provides management decision-makers with corresponding practical basis.

This paper proposes a research framework for social impact assessment of the construction land reduction project, a unique brownfield redevelopment project in China. This paper explains the steps of analyzing the social impact assessment of the construction land reduction project, including the identification of the indicator dimension by the word frequency statistical method, the initial indicator system determined by the literature analysis method, the final indicator system obtained by the Delphi method screening, and the indicator weight determined by the entropy combination weight method, etc. Then a grey clustering model for social impact assessment of the construction land reduction project is established. Finally, a numerical example is applied to show feasibility of the framework. This research is expected to put forward suggestions and countermeasures for the implementation of similar projects through the social impact assessment of the construction land reduction project. And the assessment indicator system and evaluation model constructed in this paper can provide reference for the social impact evaluation of similar projects in the future.