Commonsense reasoning aims to empower machines with the human ability to make presumptions about ordinary situations in our daily life. In this paper, we propose a textual inference framework for answering commonsense questions, which effectively utilizes external, structured commonsense knowledge graphs to perform explainable inferences. The framework first grounds a question-answer pair from the semantic space to the knowledge-based symbolic space as a schema graph, a related sub-graph of external knowledge graphs. It represents schema graphs with a novel knowledge-aware graph network module named KA GNE T, and finally scores answers with graph representations. Our model is based on graph convolutional networks and LSTMs, with a hierarchical path-based attention mechanism. The intermediate attention scores make it transparent and interpretable, which thus produce trustworthy inferences. Using ConceptNet as the only external resource for BERT-based models, we achieved state-of-the-art performance on the CommonsenseQA, a large-scale dataset for commonsense reasoning. We open-source our code 1 to the community for future research in knowledge-aware commonsense reasoning.
Existing work that augment question answering (QA) models with external knowledge (e.g., knowledge graphs) either struggle to model multi-hop relations efficiently, or lack transparency into the model's prediction rationale. In this paper, we propose a novel knowledge-aware approach that equips pretrained language models (PTLMs) with a multi-hop relational reasoning module, named multi-hop graph relation network (MHGRN). It performs multi-hop, multi-relational reasoning over subgraphs extracted from external knowledge graphs. The proposed reasoning module unifies path-based reasoning methods and graph neural networks and results in better interpretability and scalability. We also empirically show its effectiveness and scalability on CommonsenseQA and OpenbookQA datasets, and interpret its behaviors with case studies, with the code for experiments released 1 .
Recently, large-scale pretrained language models have demonstrated impressive performance on several commonsense-reasoning benchmark datasets. However, building machines with commonsense to compose realistically plausible sentences remains challenging. In this paper, we present a constrained text generation task, COMMONGEN associated with a benchmark dataset, to explicitly test machines for the ability of generative commonsense reasoning. Given a set of common concepts (e.g., {dog, frisbee, catch, throw}); the task is to generate a coherent sentence describing an everyday scenario using these concepts (e.g., "a man throws a frisbee and his dog catches it").The COMMONGEN task is challenging because it inherently requires 1) relational reasoning with background commonsense knowledge, and 2) compositional generalization ability to work on unseen concept combinations. Our dataset, constructed through a combination of crowdsourced and existing caption corpora, consists of 77k commonsense descriptions over 35k unique concept-sets. Experiments show that there is a large gap between state-of-the-art text generation models (e.g., T5) and human performance (31.6% v.s. 63.5% in SPICE metric). Furthermore, we demonstrate that the learned generative commonsense reasoning capability can be transferred to improve downstream tasks such as CommonsenseQA (76.9% to 78.4 in dev accuracy) by generating additional context.
CategoryRelations 1-hop 2-hop
Spatial knowledgeAtLocation, LocatedNear 9.40% 39.31%
Alzheimer disease (AD) is a primary cause of cognitive dysfunction in the elderly population worldwide. Despite the allocation of enormous amounts of funding and resources to studying this brain disorder, there are no effective pharmacological treatments for reducing the severity of pathology and restoring cognitive function in affected people. Recent reports on the failure of multiple clinical trials for AD have highlighted the need to diversify further the search for new therapeutic strategies for cognitive dysfunction. Thus, studies detailing the neuroprotective effects of physical activity (PA) on the brain in AD were reviewed, and mechanisms by which PA might mitigate AD-related cognitive decline were explored. A MEDLINE database search was used to generate a list of studies conducted between January 2007 and September 2014 (n=394). These studies, along with key references, were screened to identify those that assessed the effects of PA on AD-related biomarkers and cognitive function. The search was not limited on the basis of intensity, frequency, duration, or mode of activity. However, studies in which PA was combined with another intervention (eg, diet, pharmacotherapeutics, ovariectomy, cognitive training, behavioral therapy), and studies not written in English were excluded. Thirty-eight animal and human studies met entry criteria. Most of the studies suggested that PA attenuates neuropathology and positively affects cognitive function in AD. Although the literature lacked sufficient evidence to support precise PA guidelines, convergent evidence does suggest that the incorporation of regular PA into daily routines mitigates AD-related symptoms, especially when deployed earlier in the disease process. Here the protocols used to alter the progression of AD-related neuropathology and cognitive decline are highlighted, and the implications for physical therapist practice are discussed.
Humans can learn a new language task efficiently with only few examples, by leveraging their knowledge obtained when learning prior tasks. In this paper, we explore whether and how such cross-task generalization ability can be acquired, and further applied to build better few-shot learners across diverse NLP tasks. We introduce CROSSFIT , a problem setup for studying cross-task generalization ability, which standardizes seen/unseen task partitions, data access during different learning stages, and the evaluation protocols. To instantiate different seen/unseen task partitions in CROSS-FIT and facilitate in-depth analysis, we present the NLP Few-shot Gym, a repository of 160 diverse few-shot NLP tasks created from openaccess NLP datasets and converted to a unified text-to-text format. Our analysis reveals that the few-shot learning ability on unseen tasks can be improved via an upstream learning stage using a set of seen tasks. We also observe that the selection of upstream learning tasks can significantly influence few-shot performance on unseen tasks, asking further analysis on task similarity and transferability. 1
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.