RESUMEN
In the period 2022-2023, an analysis of fourteen phenotypic traits was conducted across 192 maize accessions in the Aral region of Xinjiang. The Shannon-Wiener diversity index was employed to quantify the phenotypic diversity among the accessions. Subsequently, a comprehensive evaluation of the index was performed utilizing correlation analysis, principal component analysis (PCA) and cluster analysis. The results highlighted significant findings: (1) A pronounced diversity was evident across the 192 maize accessions, accompanied by complex interrelationships among the traits. (2) The 14 phenotypic traits were transformed into 3 independent indicators through principal component analysis: spike factor, leaf width factor, and number of spikes per plant. (3) The 192 materials were divided into three groups using cluster analysis. The phenotypes in Group III exhibited the best performance, followed by those in Group I, and finally Group II. The selection of the three groups can vary depending on the breeding objectives. This study analysed the diversity of phenotypic traits in maize germplasm resources. Maize germplasm was categorised based on similar phenotypes. These findings provide theoretical insights for the study of maize accessions under analogous climatic conditions in Alar City, which lay the groundwork for the efficient utilization of existing germplasm as well as the development and selection of new varieties.
RESUMEN
Visual question answering with natural language explanation (VQA-NLE) is a challenging task that requires models to not only generate accurate answers but also to provide explanations that justify the relevant decision-making processes. This task is accomplished by generating natural language sentences based on the given question-image pair. However, existing methods often struggle to ensure consistency between the answers and explanations due to their disregard of the crucial interactions between these factors. Moreover, existing methods overlook the potential benefits of incorporating additional knowledge, which hinders their ability to effectively bridge the semantic gap between questions and images, leading to less accurate explanations. In this paper, we present a novel approach denoted the knowledge-based iterative consensus VQA-NLE (KICNLE) model to address these limitations. To maintain consistency, our model incorporates an iterative consensus generator that adopts a multi-iteration generative method, enabling multiple iterations of the answer and explanation in each generation. In each iteration, the current answer is utilized to generate an explanation, which in turn guides the generation of a new answer. Additionally, a knowledge retrieval module is introduced to provide potentially valid candidate knowledge, guide the generation process, effectively bridge the gap between questions and images, and enable the production of high-quality answer-explanation pairs. Extensive experiments conducted on three different datasets demonstrate the superiority of our proposed KICNLE model over competing state-of-the-art approaches. Our code is available at https://github.com/Gary-code/KICNLE.
