Search for: All records

Periphyton is a ubiquitous niche in aquatic environments and can be a significant source of dissolved organic matter (DOM) production and leaching, especially in such environment as the Everglades, a slow-water flow wetland in Florida, USA. We employed an array of methods, including compositional analysis, 3-dimensional excitation emission matrix (3-D EEM) fluorescence spectroscopy, and attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy, to perform quantitative and qualitative analyses on the DOM produced by periphyton and DOM in surrounding surface water and periphyton overlying water for comparison purposes. Higher dissolved organic carbon (DOC) and total dissolved nitrogen (TDN) contents in periphyton pore water than surface water and periphyton overlying water indicated the remarkable contribution from periphyton-produced DOM. Higher total protein, carbohydrate, and thiol contents in periphyton pore water than in surface water and periphyton overlying water underscored the possibility of periphyton pore water DOM leached from periphyton. These results agreed with 3-D EEM and ATR-FTIR analyses that showed the prevalence of possible microbial source of periphyton pore water DOM as indicated by higher fluorescence index (FI) than surface water and periphyton overlying water. Similarly, the size-fractionated DOM from surface water demonstrated terrestrial sources, and periphyton pore water demonstrated microbial sources regardless of their differences in size based on their FI values. The types of periphyton affect the production and composition of DOM, as evidenced by higher total protein, carbohydrate, and chlorophyll-a (Chl-a) contents in floating mat on the water surface than in epiphyton attached to submerged phytoplankton, probably because the former is photo-synthetically more productive than the latter due to different light availability. This study provided fundamental information on periphyton DOM that is essential for further investigating its role in carbon cycle and its biogeochemistry. 

Data augmentation has been a popular method for fine-tuning pre-trained language models to increase model robustness and performance. With augmentation data coming from modifying gold train data (in-sample augmentation) or being harvested from general domain unlabeled data (out-of-sample augmentation), the quality of such data is the key to successful fine-tuning. In this paper, we propose a dynamic data selection method to select effective augmentation data from different augmentation sources according to the model’s learning stage, by identifying a set of augmentation samples that optimally facilitates the learning process of the most current model. The method firstly filters out augmentation samples with noisy pseudo labels through a curriculum learning strategy, then estimates the effectiveness of reserved augmentation data by its influence scores on the current model at every update, allowing the data selection process tightly tailored to model parameters. And the two-stage augmentation strategy considers in-sample augmentation and out-of-sample augmentation in different learning stages. Experiments with both kinds of augmentation data on a variety of sentence classification tasks show that our method outperforms strong baselines, proving the effectiveness of our method. Analysis confirms the dynamic nature of the data effectiveness and the importance of model learning stages in utilization of augmentation data. 

Mild cognitive impairment is the prodromal stage of Alzheimer’s disease. Its detection has been a critical task for establishing cohort studies and developing therapeutic interventions for Alzheimer’s. Various types of markers have been developed for detection. For example, imaging markers from neuroimaging have shown great sensitivity, while its cost is still prohibitive for large-scale screening of early dementia. Recent advances from digital biomarkers, such as language markers, have provided an accessible and affordable alternative. While imaging markers give anatomical descriptions of the brain, language markers capture the behavior characteristics of early dementia subjects. Such differences suggest the benefits of auxiliary information from the imaging modality to improve the predictive power of unimodal predictive models based on language markers alone. However, one significant barrier to the joint analysis is that in typical cohorts, there are only very limited subjects that have both imaging and language modalities. To tackle this challenge, in this paper, we develop a novel crossmodal augmentation tool, which leverages auxiliary imaging information to improve the feature space of language markers so that a subject with only language markers can benefit from imaging information through the augmentation. Our experimental results show that the multi-modal predictive model trained with language markers and auxiliary imaging information significantly outperforms unimodal predictive models.