While recent work has convincingly showed that sequence-to-sequence models struggle to generalize to new compositions (termed compositional generalization), little is known on what makes compositional generalization hard on a particular test instance. In this work, we investigate what are the factors that make generalization to certain test instances challenging. We first substantiate that indeed some examples are more difficult than others by showing that different models consistently fail or succeed on the same test instances. Then, we propose a criterion for the difficulty of an example: a test instance is hard if it contains a local structure that was not observed at training time. We formulate a simple decision rule based on this criterion and empirically show it predicts instance-level generalization well across 5 different semantic parsing datasets, substantially better than alternative decision rules. Last, we show local structures can be leveraged for creating difficult adversarial compositional splits and also to improve compositional generalization under limited training budgets by strategically selecting examples for the training set.
A rapidly growing body of research has demonstrated the inability of NLP models to generalize compositionally and has tried to alleviate it through specialized architectures, training schemes, and data augmentation, among other approaches. In this work, we study a different relatively under-explored approach: sampling diverse train sets that encourage compositional generalization. We propose a novel algorithm for sampling a structurally diverse set of instances from a labeled instance pool with structured outputs. Evaluating on 5 semantic parsing datasets of varying complexity, we show that our algorithm performs competitively with or better than prior algorithms in not only compositional template splits but also traditional IID splits of all but the least structurally diverse datasets. In general, we find that diverse train sets lead to better generalization than random training sets of same size in 9 out of 10 dataset-split pairs, with over 10% absolute improvement in 5, providing further evidence to their sample efficiency. Moreover, we show that structural diversity also makes for more comprehensive test sets that require diverse training to succeed on. Finally, we use information theory to show that reduction in spurious correlations between substructures may be one reason why diverse training sets improve generalization.
Introduction:Lower respiratory tract infection is an infection beneath the larynx which includes: Pneumonia, Wheeze associated Lower respiratory tract infection, Bronchiolitis and Empyema. The aim of this study was to find out the association among hyponatremia and LRTI in tertiary care center. Materials and Methods: The sample size was calculated to be a minimum of 50 subjects. Based on clinical signs and symptoms (as defined by WHO) and infiltrates present on chest X-ray diagnosis of LRTI such as Pneumonia, Bronchiolitis was made. Then patients were subjected to routine investigations such as: Complete blood count, Serum electrolyte (serum sodium) at the time of admission, on day 2 and day 3 by Ion selective electrodes. Results: Mild hyponatremia was found among 11 patients (16.92%), moderate among 9 (13.85%) and severe among 2 (3.08%) patients. Hyponatremia was found to be more common among 1 to 5 years age group as compared to ≥ 2months to 12 months and > 5 to 12 years age groups. Conclusion: Hyponatremia is a significantly common association among hospitalized children with lower respiratory tract infections and it is mainly due to syndrome of inappropriate antidiuretic hormone secretion (SIADH).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.