Search for: All records

In this study, the amphiphilic salt lithium trifluoromethanesulfonylimide octadecane (C18LiTFSI) was used as a basis to investigate the effects of anion density and cation coordination sites within blended electrolytes with strong ionic aggregation. C18LiTFSI was previously reported as a single-component, ion-condensed electrolyte with a wide layered liquid crystalline phase regime. Three additive molecules with varyingly sized polar sulfonyl groups attached to an octodecane-tail were synthesized and mixed with C18LiTFSI. The thermal properties, morphology, and ionic conductivity of the blended electrolytes were characterized. It was found that the blended electrolytes exhibited layered liquid crystalline morphology over a narrower temperature range than the pure salt, and the ionic conductivity of the blended liquid crystalline electrolytes were generally lower than that of the pure salt. Surprising, the additives were found to have the greatest effect on the bulk ionic conductivity of the semicrystalline phase of the electrolytes. Addition of minor fractions of methylsulfonyloctadecane to C18LiTFSI resulted in increases in conductivity of over two orders of magnitude at room temperature, while addition of ethylsulfonyloctadecane or isopropylsulfonyloctadecane with the larger head group resulted in decreased ionic conductivity over the entire composition space and temperature range investigated. 

Everyone makes mistakes. So do human annotators when curating labels for named entity recognition (NER). Such label mistakes might hurt model training and interfere model comparison. In this study, we dive deep into one of the widely-adopted NER benchmark datasets, CoNLL03 NER. We are able to identify label mistakes in about 5.38% test sentences, which is a significant ratio considering that the state-of-the-art test F1 score is already around 93%. Therefore, we manually correct these label mistakes and form a cleaner test set. Our re-evaluation of popular models on this corrected test set leads to more accurate assessments, compared to those on the original test set. More importantly, we propose a simple yet effective framework, CrossWeigh, to handle label mistakes during NER model training. Specifically, it partitions the training data into several folds and train independent NER models to identify potential mistakes in each fold. Then it adjusts the weights of training data accordingly to train the final NER model. Extensive experiments demonstrate significant improvements of plugging various NER models into our proposed framework on three datasets. All implementations and corrected test set are available at our Github repo