Dan Ye

Liangyi Kang, Jie Liu, Lingqiao Liu et al.

Charge prediction, determining charges for criminal cases by analyzing the textual fact descriptions, is a promising technology in legal assistant systems. In practice, the fact descriptions could exhibit a significant intra-class variation due to factors like non-normative use of language, which makes the prediction task very challenging, especially for charge classes with too few samples to cover the expression variation. In this work, we explore to use the charge definitions from criminal law to alleviate this issue. The key idea is that the expressions in a fact description should have corresponding formal terms in charge definitions, and those terms are shared across classes and could account for the diversity in the fact descriptions. Thus, we propose to create auxiliary fact representations from charge definitions to augment fact descriptions representation. The generated auxiliary representations are created through the interaction of fact description with the relevant charge definitions and terms in those definitions by integrated sentence- and word-level attention scheme. Experimental results on two datasets show that our model achieves significant improvement than baselines, especially for classes with few samples.

Yingqi Qu, Jie Liu, Liangyi Kang et al.

With the rapid growth of knowledge bases (KBs), question answering over knowledge base, a.k.a. KBQA has drawn huge attention in recent years. Most of the existing KBQA methods follow so called encoder-compare framework. They map the question and the KB facts to a common embedding space, in which the similarity between the question vector and the fact vectors can be conveniently computed. This, however, inevitably loses original words interaction information. To preserve more original information, we propose an attentive recurrent neural network with similarity matrix based convolutional neural network (AR-SMCNN) model, which is able to capture comprehensive hierarchical information utilizing the advantages of both RNN and CNN. We use RNN to capture semantic-level correlation by its sequential modeling nature, and use an attention mechanism to keep track of the entities and relations simultaneously. Meanwhile, we use a similarity matrix based CNN with two-directions pooling to extract literal-level words interaction matching utilizing CNNs strength of modeling spatial correlation among data. Moreover, we have developed a new heuristic extension method for entity detection, which significantly decreases the effect of noise. Our method has outperformed the state-of-the-arts on SimpleQuestion benchmark in both accuracy and efficiency.

Cheng-yi Zhang, Dan Ye, Cong-lei Zhong et al.

It is well known that as a famous type of iterative methods in numerical linear algebra, Gauss-Seidel iterative methods are convergent for linear systems with strictly or irreducibly diagonally dominant matrices, invertible $H-$matrices (generalized strictly diagonally dominant matrices) and Hermitian positive definite matrices. But, the same is not necessarily true for linear systems with nonstrictly diagonally dominant matrices and general $H-$matrices. This paper firstly proposes some necessary and sufficient conditions for convergence on Gauss-Seidel iterative methods to establish several new theoretical results on linear systems with nonstrictly diagonally dominant matrices and general $H-$matrices. Then, the convergence results on preconditioned Gauss-Seidel (PGS) iterative methods for general $H-$matrices are presented. Finally, some numerical examples are given to demonstrate the results obtained in this paper.