Search

Abstract Thanks to the strong representation capability of the pre-trained language models, supervised grammatical error correction has achieved promising performance. However, traditional model training depends significantly on the large scale of similar distributed samples. The model performance decreases sharply once the distributions of training and testing data are inconsistent. To address this issue, we propose an automatic sampling approach to effectively select high-quality samples from different corpora and filter out irrelevant or harmful ones. Concretely, we first provide a detailed analysis of error type and sentence length distributions on all datasets. Second, our corpus weighting approach is exploited to yield different weights for each sample automatically based on analysis results, thus emphasizing beneficial samples and ignoring the noisy ones. Finally, we enhance typical Seq2Seq and Seq2Edit grammatical error correction models with pre-trained language models and design a model ensemble algorithm for integrating the advantages of heterogeneous models and weighted samples. Experiments on the benchmark datasets demonstrate that the proper utilization of different corpora is extremely helpful in enhancing the accuracy of grammatical error correction. The detailed analysis gains more insights into the effect of different corpus weighting strategies.

Abstract Background Flower color plays a crucial role in attracting pollinators and facilitating environmental adaptation. Investigating the causes of flower color polymorphism and understanding their potential effects on both ecology and genetics can enhance our understanding of flower color polymorphism in wild plant. Results In this study, we examined the differences of potential male and female fitness between purple- and yellow- flower individuals in Iris potaninii on the Qinghai-Tibet Plateau, and screened key genes and positively selective genes involved in flower color change. Our results showed that yellow flower exhibited a higher pollen-to-ovule ratio. Yellow flowers were derived from purple flowers due to the loss of anthocyanins, and F3H could be an essential gene affecting flower color variation though expression regulation and sequence polymorphism in this species. Furthermore, our findings suggest that genes positively selected in yellow-flowered I. potaninii might be involved in nucleotide excision repair and plant-pathogen interactions. Conclusions These results suggest that F3H induces the flower color variation of Iris potaninii, and the subsequent ecological and additive positive selection on yellow flowers may further enhance plant adaptations to alpine environments.