No abstract

No abstract

To fully evaluate the overall performance of different NLP models in a given domain, many evaluation benchmarks are proposed, such as GLUE, SuperGLUE and CLUE. The field of natural language understanding has traditionally focused on benchmarks for various tasks in languages such as Chinese, English, and multilingual, however, there has been a lack of attention given to the area of classical Chinese, also known as "wen yan wen (文言文)", which has a rich history spanning thousands of years and holds significant cultural and academic value.For the prosperity of the NLP community, in this paper, we introduce the WYWEB evaluation benchmark, which consists of nine NLP tasks in classical Chinese, implementing sentence classification, sequence labeling, reading comprehension, and machine translation. We evaluate the existing pre-trained language models, which are all struggling with this benchmark. We also introduce a number of supplementary datasets and additional tools to help facilitate further progress on classical Chinese NLU. The github repository is https://github.com/baudzhou/WYWEB.

Clogging is a major operational and maintenance issue associated with the use of constructed wetlands. In this study, four lab-scale vertical flow constructed wetlands (VFCW) were used to fully understand the development mechanisms of various types of clogging and their recovery characteristics. The VFCWs were fed with glucose solution, starch suspension with and without bacteriostat, glucose, and starch mixed solution, respectively, to simulate Bio-clogging, organic particle clogging (Op-clogging), inert particle clogging (Ip-clogging), and the combination of Bio-clogging and Op-clogging (C-clogging). Resting operations with water decline were applied to relieve the clogging in the VFCWs. The results indicate that Op-clogging occurred first, followed by C-clogging and Bio-clogging. Ip-clogging took the longest time to develop and did not occur by the end of this study. The microscope analysis found that the extracellular polymeric substances (EPS) bonded the starch particles together to form a dense membrane-like structure and promoted the clogging process. In addition, surface clogging was observed in all four experimental beds. Op-clogging occurred much closer to the surface than those caused by soluble organic matter and inert particles. Furthermore, the growth of biofilm caused significant decline in hydraulic conductivity, whereas its influence on porosity was relatively slight. Moreover, applying resting operation with water decline was effective for recovery from Bio-clogging, Op-clogging, and C-clogging in VFCWs except for Ip-clogging. The results also implied the recovery rates through applying resting operation with water decline were much higher than that with constant water level.

The horizontal subsurface constructed wetland (HSSF CW) is a highly effective technique for stormwater treatment. However, progressive clogging in HSSF CW is a widespread operational problem. The aim of this study was to understand the clogging development of HSSF CWs during stormwater treatment and to assess the influence of microorganisms and vegetation on the clogging. Moreover, the hydraulic performance of HSSF CWs in the process of clogging was evaluated in a tracer experiment. The results show that the HSSF CW can be divided into two sections, section I (circa 0-35 cm) and section II (circa 35-110 cm). The clogging is induced primarily by solid entrapment in section I and development of biofilm and vegetation roots in section II, respectively. The influence of vegetation and microorganisms on the clogging appears to differ in sections I and II. The tracer experiment shows that the hydraulic efficiency (λ) and the mean hydraulic retention time (t ) increase with the clogging development; although, the short-circuiting region (S) extends slightly. In addition, the presence of vegetation can influence the hydraulic performance of the CWs, and their impact depends on the characteristics of the roots.