Text-based question answering (TBQA) has been studied extensively in recent years. Most existing approaches focus on finding the answer to a question within a single paragraph. However, many difficult questions require multiple supporting evidence from scattered text across two or more documents. In this paper, we propose the Dynamically Fused Graph Network (DFGN), a novel method to answer those questions requiring multiple scattered evidence and reasoning over them. Inspired by human's step-by-step reasoning behavior, DFGN includes a dynamic fusion layer that starts from the entities mentioned in the given query, explores along the entity graph dynamically built from the text, and gradually finds relevant supporting entities from the given documents. We evaluate DFGN on HotpotQA, a public TBQA dataset requiring multi-hop reasoning. DFGN achieves competitive results on the public board. Furthermore, our analysis shows DFGN could produce interpretable reasoning chains.
2015): Effects of fermented rapeseed meal on antioxidant functions, serum biochemical parameters and intestinal morphology in broilers, Food and Agricultural Immunology, ABSTRACT This trial was conducted to determine the effects of solid-state fermented rapeseed meal (FRSM) on antioxidant functions, serum biochemical parameters and intestinal morphology of broilers. The rapeseed meal (RSM) was fermented with Bacillus subtilis, Candida utilis and Enterococcus faecalis. One hundred and eighty-day-old Arbor Acres broilers were randomly divided into three treatments: a corn-soybean meal based diet and two experimental diets in which the control diet was supplemented with 10.9% RSM or 9.41% FRSM. Results showed that the fermentation process can effectively increase crude protein or small peptides level, and decrease crude fiber level, glucosinolate, isothiocyanate, tannin and phytic acid level in RSM. The levels of serum total antioxidative capacity, total superoxide dismutase, total protein, albumin and glucose of birds fed FRSM were higher than birds fed RSM on days 21 and 42. FRSM also improved the intestinal morphology of broilers. The results indicate that FRSM can be effectively applied in broiler diets. ARTICLE HISTORY
Rate constants for the reactions H02 + O (1) and H02 + H (2) were measured in a discharge-flow apparatus fitted with back-to-back laser-induced fluorescence and vacuum UV resonance fluorescence detectors. The decays of [O] and [H] were monitored under conditions of large excess H02, generated by F + H202 and detected as OH after conversion with added excess NO. kl and k2 were found to be (5.4 ± 0.9) X 10"11 and (7.4 ± 1.2) X 10~n cm1 23 s'1, respectively. The branching ratios of (2), whose three sets of products are OH + OH (2a), H20 + O (2b), and H2 + 02 (2c) were determined by reacting small, known concentrations of H02 with large excess of H and measuring the [OH] and [0] formed. They were found to be 0.87 ± 0.04, 0.04 ± 0.02, and 0.09 ± 0.045, respectively. These results are compared with published data and discussed in terms of the likely course of the molecular interactions.
Survival analysis is a hotspot in statistical research for modeling time-to-event information with data censorship handling, which has been widely used in many applications such as clinical research, information system and other fields with survivorship bias. Many works have been proposed for survival analysis ranging from traditional statistic methods to machine learning models. However, the existing methodologies either utilize counting-based statistics on the segmented data, or have a pre-assumption on the event probability distribution w.r.t. time. Moreover, few works consider sequential patterns within the feature space. In this paper, we propose a Deep Recurrent Survival Analysis model which combines deep learning for conditional probability prediction at finegrained level of the data, and survival analysis for tackling the censorship. By capturing the time dependency through modeling the conditional probability of the event for each sample, our method predicts the likelihood of the true event occurrence and estimates the survival rate over time, i.e., the probability of the non-occurrence of the event, for the censored data. Meanwhile, without assuming any specific form of the event probability distribution, our model shows great advantages over the previous works on fitting various sophisticated data distributions. In the experiments on the three realworld tasks from different fields, our model significantly outperforms the state-of-the-art solutions under various metrics.
Recent work on non-autoregressive neural machine translation (NAT) aims at improving the efficiency by parallel decoding without sacrificing the quality. However, existing NAT methods are either inferior to Transformer or require multiple decoding passes, leading to reduced speedup. We propose the Glancing Language Model (GLM) for single-pass parallel generation models. With GLM, we develop Glancing Transformer (GLAT) for machine translation. With only single-pass parallel decoding, GLAT is able to generate high-quality translation with 8×-15× speedup. Note that GLAT does not modify the network architecture, which is a training method to learn word interdependency. Experiments on multiple WMT language directions show that GLAT outperforms all previous single pass non-autoregressive methods, and is nearly comparable to Transformer, reducing the gap to 0.25-0.9 BLEU points.
Paraphrasing plays an important role in various natural language processing (NLP) tasks, such as question answering, information retrieval and sentence simplification. Recently, neural generative models have shown promising results in paraphrase generation. However, prior work mainly focused on single paraphrase generation, while ignoring the fact that diversity is essential for enhancing generalization capability and robustness of downstream applications. Few works have been done to solve diverse paraphrase generation. In this paper, we propose a novel approach with two discriminators and multiple generators to generate a variety of different paraphrases. A reinforcement learning algorithm is applied to train our model. Our experiments on two realworld datasets demonstrate that our model not only gains a significant increase in diversity but also improves generation quality over several state-of-the-art baselines.
With the increasing demand for non-contact temperature sensing, the development of optical thermometer with excellent performance is more and more compelling. Cr3+-doped InTaO4 phosphor was prepared for the implementation of...
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.