Text-based question answering (TBQA) has been studied extensively in recent years. Most existing approaches focus on finding the answer to a within single paragraph. However, many difficult questions require multiple supporting evidence from scattered text among two or more documents. In this paper, we propose Dynamically Fused Graph Network (DFGN), novel method those requiring and reasoning over them. Inspired by human’s step-by-step behavior, DFGN includes dynamic fusion layer that starts...

Survival analysis is a hotspot in statistical research for modeling time-to-event information with data censorship handling, which has been widely used many applications such as clinical research, system and other fields survivorship bias. Many works have proposed survival ranging from traditional statistic methods to machine learning models. However, the existing methodologies either utilize counting-based statistics on segmented data, or pre-assumption event probability distribution w.r.t....

Lihua Qian, Hao Zhou, Yu Bao, Mingxuan Wang, Lin Qiu, Weinan Zhang, Yong Yu, Lei Li. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

The accurate estimation of Muskingum model parameter is essential to give flood routing for control in water resources management. continues be a popular method routing, and its global optimization problem with the main objective find set optimal values that attains best fit between observed computed flow. Although some techniques have been employed estimate parameters model, an efficient still required improve computational precision. Therefore, this paper, differential evolution (DE)...

ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTKinetics and product channels of the reactions perhydroxyl with oxygen hydrogen atoms at 296 KU. C. Sridharan, L. X. Qiu, F. KaufmanCite this: J. Phys. Chem. 1982, 86, 23, 4569–4574Publication Date (Print):November 1, 1982Publication History Published online1 May 2002Published inissue 1 November 1982https://pubs.acs.org/doi/10.1021/j100220a023https://doi.org/10.1021/j100220a023research-articleACS PublicationsRequest reuse permissionsArticle...

Distributed representations of words (aka word embedding) have proven helpful in solving natural language processing (NLP) tasks. Training distributed with neural networks has lately been a major focus researchers the field. Recent work on embedding, Continuous Bag-of-Words (CBOW) model and Skip-gram (Skip-gram) model, produced particularly impressive results, significantly speeding up training process to enable representation learning from large-scale data. However, both CBOW do not pay...

Lihua Qian, Lin Qiu, Weinan Zhang, Xin Jiang, Yong Yu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint (EMNLP-IJCNLP). 2019.

In assisted reproduction techniques, oocytes encounter elevated levels of reactive oxygen species (ROS) during in vitro maturation (IVM). Oxidative stress adversely affects oocyte quality, hampering their maturation, growth, and subsequent development. Thus, mitigating excessive ROS to safeguard less viable IVM stands as a strategy. Numerous antioxidants have been explored for IVM, yielding considerable effects; however, several aspects, including solubility, stability, safety, demand...

Text-based question answering (TBQA) has been studied extensively in recent years. Most existing approaches focus on finding the answer to a within single paragraph. However, many difficult questions require multiple supporting evidence from scattered text among two or more documents. In this paper, we propose Dynamically Fused Graph Network(DFGN), novel method those requiring and reasoning over them. Inspired by human's step-by-step behavior, DFGN includes dynamic fusion layer that starts...

Research on monitoring forest disturbances and analyzing its driving factors is crucial for the sustainable management of ecosystems. To quantitatively identify spatial distribution dynamic changes disturbance in Guangdong Province from 1990 to 2019, long-term Landsat time series imagery LandTrendr change detection algorithm were utilized. The impact four types landscape fragmentation (attrition, perforation, shrinkage, subdivision) was analyzed using Forman index. Geodetector model used...

Large Language Model (LLM) agents frameworks often employ modular architectures, incorporating components such as planning, reasoning, action execution, and reflection to tackle complex tasks. However, quantifying the contribution of each module overall system performance remains a significant challenge, impeding optimization interpretability. To address this, we introduce CapaBench (Capability-level Assessment Benchmark), an evaluation framework grounded in cooperative game theory's Shapley...

Abstract This article proposes a novel asymmetric chamfering consequent-pole Vernier permanent magnet (CPVPM) rotor topology of torque enhancement and ripple reduction for electric propulsion vessels. The structure is featured by the chamfered applied on CP adjacent to PMs. It demonstrated that without any modification high price limited storage rare-earth PM materials employed rotor, density machine with an can be effectively increased. influence key parameters CPVPM characteristics are...

During the process of sperm cryopreservation, overproduction reactive oxygen species (ROS) triggers oxidative stress thereby leading to a reduction in motility and quality. Therefore, it is feasible strategy mitigate damage during cryopreservation by adding antioxidants freezing media. In this study, we explored potential melatonin protect from stress-induced evaluating sperm-related parameters after thawing through self-assembly with hyaluronic acid-bilirubin conjugate into nanoparticles...

Patients discuss complementary and alternative medicine (CAM) in online health communities. Sometimes, patients' conflicting opinions toward CAM-related issues trigger debates the community. The objectives of this paper are to identify such debates, controversial CAM therapies a popular breast cancer community, as well stances towards them. To scale our analysis, we trained set classifiers. We first constructed supervised classifier based on long short-term memory neural network (LSTM)...

ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTRate constant of the hydroxyl + perhydroxyl (HO2) reaction from 252 to 420 KU. C. Sridharan, L. X. Qiu, and F. KaufmanCite this: J. Phys. Chem. 1984, 88, 7, 1281–1282Publication Date (Print):March 1, 1984Publication History Published online1 May 2002Published inissue 1 March 1984https://pubs.acs.org/doi/10.1021/j150651a006https://doi.org/10.1021/j150651a006research-articleACS PublicationsRequest reuse permissionsArticle...

Knowledge base is one of the main forms to represent information in a structured way. A knowledge typically consists Resource Description Frameworks (RDF) triples which describe entities and their relations. Generating natural language description an important task NLP, has been formulated as conditional generation tackled using sequence-to-sequence framework. Current works mostly train models by maximum likelihood estimation, tends generate lousy sentences. In this paper, we argue that such...

Word embedding has been widely studied and proven helpful in solving many natural language processing tasks.However, the ambiguity of is always a problem on learning high quality word embeddings.A possible solution sense which trains for each words instead word.Some recent work uses context clustering methods to determine senses words, heuristic nature.Other creates probabilistic model performs disambiguation iteratively.However, most previous problems embeddings based imperfect as well...

A design of 2-D forward and inverse integer transform processor is presented, which suitable for MPEG-4 AVC/H.264 visual profile. The comparability between the symmetry their arithmetic has been utilized in architecture. According to this design, implemented by using duplicated 1-D transform. Parallel register array are used realize transpose operation. Under 0.35um technology, logic gate count only 3524 when maximum frequency more than 120MHz.

Active learning for sentence understanding aims at discovering informative unlabeled data annotation and therefore reducing the demand labeled data. We argue that typical uncertainty sampling method active is time-consuming can hardly work in real-time, which may lead to ineffective sample selection. propose adversarial discrete space (AUSDS) retrieve samples more efficiently. AUSDS maps sentences into latent generated by popular pre-trained language models, discover text via attack. The...

Document-level relation extraction aims to identify relations between entities in a whole document. Prior efforts capture long-range dependencies have relied heavily on implicitly powerful representations learned through (graph) neural networks, which makes the model less transparent. To tackle this challenge, paper, we propose LogiRE, novel probabilistic for document-level by learning logic rules. LogiRE treats rules as latent variables and consists of two modules: rule generator extractor....

ADVERTISEMENT RETURN TO ISSUEArticleNEXTKinetics of the reaction OH + HO2 .fwdarw. H2O O2U. C. Sridharan, L. X. Qiu, and F. KaufmanCite this: J. Phys. Chem. 1981, 85, 23, 3361–3363Publication Date (Print):November 1, 1981Publication History Published online1 May 2002Published inissue 1 November 1981https://doi.org/10.1021/j150623a001RIGHTS & PERMISSIONSArticle Views118Altmetric-Citations20LEARN ABOUT THESE METRICSArticle Views are COUNTER-compliant sum full text article downloads since 2008...

Accurate flood routing is essential for control in water resources planning and management. The Muskingum model continues to be popular method routing. Its parameter estimation a global optimization problem with the main objective find set of optimal values that attains best fit between observed computed flow. In order improve precision, hybrid chaotic genetic algorithm (HCGA) based on sequence GA proposed model. Empirical results involve historical data from existed paper reveal HCGA...