Abstract: The present application describes systems and methods for analyzing hypertext. An element detector receives input information (e.g., hypertext) that is included in a web page. The element detector identifies one or more elements in the input information that are relevant to a classification task. The classification task may include classifying the one or more elements. An element extractor generates text tokens, positional tokens, and/or path tokens based on the identified one or more elements. A machine learning (ML) model outputs one or more vector representations based on the text tokens, positional tokens, and/or path tokens. An output classifier, which may be an additional layer of the ML model, classifies the one or more elements into classifications based on the one or more vector representations. A tool may perform one or more tasks based on the classifications.

Abstract: A question answering system includes: a first encoder module configured to receive a question, the question including a first plurality of words, and encode the question into a first vector representation; a second encoder module configured to encode a document into a second vector representation, the document including a second plurality of words; a first reading module configured to generate a third vector representation based on the first and second vector representations; a first reformulation module configured to generate a first reformulated vector representation based on the first vector representation; a second reading module configured to generate a fifth vector representation based on the second vector representation and the first reformulated vector representation; a second reformulation module configured to generate a second reformulated vector representation based on first reformulated vector representation; and an answer module configured to determine an answer to the question based on the secon

Abstract: A summarization system includes: K embedding modules configured to: receive K blocks of text, respectively, of a document to be summarized; and generate K first representations based on the K blocks of text, respectively, where K is an integer greater than 2; a first propagation module configured to generate second representations based on the K first representations; a second propagation module configured to generate third representations based on the second representations; an output module configured to select ones of the K blocks based on the third representations; and a summary module configured to generate a summary of the document from text of the selected ones of the K blocks.

Abstract: There is disclosed a computer implemented method that includes accessing a dataset having (1) a first set of questions including at least one pair of relational questions that correspond respectively with a pair of binary answers and (2) a second set of questions including at least another pair of relational questions that correspond respectively with a binary answer and a scalar answer. A question answering network is used to compute both a relational loss for the at least one pair of relational questions, and a relational loss for the at least another pair of relational questions. Both the relational loss for the at least one pair of relational questions and the relational loss for the at least another pair of relational questions are optimized, and a neural network model is trained with the optimized relational losses.

Abstract: A question answering system includes: a first encoder module configured to receive a question, the question including a first plurality of words, and encode the question into a first vector representation; a second encoder module configured to encode a document into a second vector representation, the document including a second plurality of words; a first reading module configured to generate a third vector representation based on the first and second vector representations; a first reformulation module configured to generate a first reformulated vector representation based on the first vector representation; a second reading module configured to generate a fifth vector representation based on the second vector representation and the first reformulated vector representation; a second reformulation module configured to generate a second reformulated vector representation based on first reformulated vector representation; and an answer module configured to determine an answer to the question based on the secon

Abstract: There is disclosed a computer implemented method for improving an efficiency of training of a question answering network, the question answering network including a neural network. The computer implemented method includes accessing a dataset having first and second sets of questions, The first set of questions includes at least one pair of relational questions with each of the at least one pair of relational questions corresponding respectively with a pair of binary answers. The second set of questions includes at least another pair of relational questions with each of the at least another pair of relational questions corresponding respectively with a binary answer and a scalar answer, The question answering network is used to compute both a relational loss for the at least one pair of relational questions; and a relational loss for the at least another pair of relational questions.

Abstract: A method of using a first neural network includes: by the first neural network, receiving a text; by the first neural network, receiving a question concerning the text; and by the first neural network, determining an answer to the question using the text, where the first neural network is trained to answer the question about the text adversarially by a second neural network that is trained to maximize a likelihood of failure of the first neural network to correctly answer questions.