There is a recent interest in investigating fewshot NER, where the low-resource target domain has different label sets compared with a resource-rich source domain. Existing methods use a similarity-based metric. However, they cannot make full use of knowledge transfer in NER model parameters. To address the issue, we propose a template-based method for NER, treating NER as a language model ranking problem in a sequence-to-sequence framework, where original sentences and statement templates filled by candidate named entity span are regarded as the source sequence and the target sequence, respectively. For inference, the model is required to classify each candidate span based on the corresponding template scores. Our experiments demonstrate that the proposed method achieves 92.55% F1 score on the CoNLL03 (rich-resource task), and significantly better than fine-tuning BERT 10.88%, 15.34%, and 11.73% F1 score on the MIT Movie, the MIT Restaurant, and the ATIS (low-resource task), respectively.

No abstract

Machine reading is a fundamental task for testing the capability of natural language understand- ing, which is closely related to human cognition in many aspects. With the rising of deep learning techniques, algorithmic models rival human performances on simple QA, and thus increasingly challenging machine reading datasets have been proposed. Though various challenges such as evidence integration and commonsense knowledge have been integrated, one of the fundamental capabilities in human reading, namely logical reasoning, is not fully investigated. We build a comprehensive dataset, named LogiQA, which is sourced from expert-written questions for testing human Logical reasoning. It consists of 8,678 QA instances, covering multiple types of deductive reasoning. Results show that state-of-the-art neural models perform by far worse than human ceiling. Our dataset can also serve as a benchmark for reinvestigating logical AI under the deep learning NLP setting. The dataset is freely available at https://github.com/lgw863/LogiQA-dataset.

Heparan sulfate (HS) is a polysaccharide known to modulate many important biological processes, including Wnt signaling. However, the biochemical interaction between HS and Wnt molecules is not well characterized largely due to the lack of suitable methods. To determine the Wnt binding domain in HS, we used a Wnt signaling-inhibitory antibody (HS20) and a panel of synthetic HS oligosaccharides with distinct lengths and sulfation modifications. We found that the binding of HS20 to heparan sulfate required sulfation at both the C2 position (2-O-sulfation) and C6 position (6-O-sulfation). The oligosaccharides with the greatest competitive effect for HS20 binding were between six and eight saccharide residues in length. Additionally, a four residue-long oligosaccharide could also be recognized by HS20 if an additional 3-O-sulfation modification was present. Furthermore, similar oligosaccharides with 2-O, 6-O and 3-O-sulfations showed inhibition for Wnt activation. These results have revealed that HS20 and Wnt recognize a HS structure containing IdoA2S and GlcNS6S, and that the 3-O-sulfation in GlcNS6S3S significantly enhances the binding of both HS20 and Wnt. This study provides the evidence for identifying the Wnt binding domain in HS and suggests a therapeutic approach to target the interaction of Wnt and HS in cancer and other diseases.

The present review aimed to summarize the effectiveness and features of traditional Chinese medicine (TCM) for the treatment of infectious diseases and to discuss the limitation of the development of TCM. The personalized medicine with TCM exerts a curative effect on viral and bacterial infectious diseases with unique advantages on the improvement of clinical manifestation, pathogen inhibition, and organ recovery during severe and drug‐resistant infection. The deficiency of personalized medicine with TCM lies in that the current research design of TCM primarily focuses on the study of the effective components and material basis of Chinese herbs at the cellular, molecular, and genetic level, while ignoring the guidance of the TCM syndrome differentiation theory, which is the core concept of individualized treatment. Personalized medicine with TCM has a broad prospective for infectious diseases due to the specific efficacy and advantages. While the curative effect of individualized treatment with TCM cannot be excluded from the TCM syndrome differentiation theory, the study of personalized medicine with TCM for infectious diseases urgently requires a unified standardization of the clinical syndrome differentiation and the evolution rule of infectious diseases by TCM theory.