Abstract: A novel unified neural network framework, the dynamic memory network, is disclosed. This unified framework reduces every task in natural language processing to a question answering problem over an input sequence. Inputs and questions are used to create and connect deep memory sequences. Answers are then generated based on dynamically retrieved memories.

Abstract: A method for training parameters of a first domain adaptation model includes evaluating a cycle consistency objective using a first task specific model associated with a first domain and a second task specific model associated with a second domain. The evaluating the cycle consistency objective is based on one or more first training representations adapted from the first domain to the second domain by a first domain adaptation model and from the second domain to the first domain by a second domain adaptation model, and one or more second training representations adapted from the second domain to the first domain by the second domain adaptation model and from the first domain to the second domain by the first domain adaptation model. The method further includes evaluating a learning objective based on the cycle consistency objective, and updating parameters of the first domain adaptation model based on learning objective.

Abstract: Approaches to zero-shot learning include partitioning training data into first and second sets according to classes assigned to the training data, training a prediction module based on the first set to predict a cluster center based on a class label, training a correction module based on the second set and each of the class labels in the first set to generate a correction to a cluster center predicted by the prediction module, presenting a new class label for a new class to the prediction module to predict a new cluster center, presenting the new class label, the predicted new cluster center, and each of the class labels in the first set to the correction module to generate a correction for the predicted new cluster center, augmenting a classifier based on the corrected cluster center for the new class, and classifying input data into the new class using the classifier.

Abstract: Approaches for determining a response for an agent in an undirected dialogue are provided. The approaches include a dialogue generating framework comprising an encoder neural network, a decoder neural network, and a language model neural network. The dialogue generating framework generates a sketch sentence response with at least one slot. The sketch sentence response is generated word by word and takes into account the undirected dialogue and agent traits of the agent making the response. The dialogue generating framework generates sentence responses by filling the slot with words from the agent traits. The dialogue generating framework ranks the sentence responses according to perplexity by passing the sentence responses through a language model and selects a final response which is a sentence response that has a lowest perplexity.

Abstract: The technology disclosed provides a quasi-recurrent neural network (QRNN) encoder-decoder model that alternates convolutional layers, which apply in parallel across timesteps, and minimalist recurrent pooling layers that apply in parallel across feature dimensions.

Abstract: Approaches for the translation of structured text include an embedding module for encoding and embedding source text in a first language, an encoder for encoding output of the embedding module, a decoder for iteratively decoding output of the encoder based on generated tokens in translated text from previous iterations, a beam module for constraining output of the decoder with respect to possible embedded tags to include in the translated text for a current iteration using a beam search, and a layer for selecting a token to be included in the translated text for the current iteration. The translated text is in a second language different from the first language. In some embodiments, the approach further includes scoring and pointer modules for selecting the token based on the output of the beam module or copied from the source text or reference text from a training pair best matching the source text.

Abstract: The technology disclosed provides a so-called “joint many-task neural network model” to solve a variety of increasingly complex natural language processing (NLP) tasks using growing depth of layers in a single end-to-end model. The model is successively trained by considering linguistic hierarchies, directly connecting word representations to all model layers, explicitly using predictions in lower tasks, and applying a so-called “successive regularization” technique to prevent catastrophic forgetting. Three examples of lower level model layers are part-of-speech (POS) tagging layer, chunking layer, and dependency parsing layer. Two examples of higher level model layers are semantic relatedness layer and textual entailment layer. The model achieves the state-of-the-art results on chunking, dependency parsing, semantic relatedness and textual entailment.

Abstract: A method for using a neural network model for natural language processing (NLP) includes receiving training data associated with a source domain and a target domain; and generating one or more query batches. Each query batch includes one or more source tasks associated with the source domain and one or more target tasks associated with the target domain. For each query batch, class representations are generated for each class in the source domain and the target domain. A query batch loss for the query batch is generated based on the corresponding class representations. An optimization is performed on the neural network model by adjusting its network parameters based on the query batch loss. The optimized neural network model is used to perform one or more new NLP tasks.

Abstract: Embodiments described herein provide unsupervised density-based clustering to infer table structure from document. Specifically, a number of words are identified from a block of text in an noneditable document, and the spatial coordinates of each word relative to the rectangular region are identified. Based on the word density of the rectangular region, the words are grouped into clusters using a heuristic radius search method. Words that are grouped into the same cluster are determined to be the element that belong to the same cell. In this way, the cells of the table structure can be identified. Once the cells are identified based on the word density of the block of text, the identified cells can be expanded horizontally or grouped vertically to identify rows or columns of the table structure.

Abstract: The technology disclosed provides a quasi-recurrent neural network (QRNN) encoder-decoder model that alternates convolutional layers, which apply in parallel across timesteps, and minimalist recurrent pooling layers that apply in parallel across feature dimensions.

Abstract: Approaches for the translation of structured text include an embedding module for encoding and embedding source text in a first language, an encoder for encoding output of the embedding module, a decoder for iteratively decoding output of the encoder based on generated tokens in translated text from previous iterations, a beam module for constraining output of the decoder with respect to possible embedded tags to include in the translated text for a current iteration using a beam search, and a layer for selecting a token to be included in the translated text for the current iteration. The translated text is in a second language different from the first language. In some embodiments, the approach further includes scoring and pointer modules for selecting the token based on the output of the beam module or copied from the source text or reference text from a training pair best matching the source text.

Abstract: The technology disclosed relates to an end-to-end neural network for question answering, referred to herein as “dynamic coattention network (DCN)”. Roughly described, the DCN includes an encoder neural network and a coattentive encoder that capture the interactions between a question and a document in a so-called “coattention encoding”. The DCN also includes a decoder neural network and highway maxout networks that process the coattention encoding to estimate start and end positions of a phrase in the document that responds to the question.

Abstract: Systems and methods for dense captioning of a video include a multi-layer encoder stack configured to receive information extracted from a plurality of video frames, a proposal decoder coupled to the encoder stack and configured to receive one or more outputs from the encoder stack, a masking unit configured to mask the one or more outputs from the encoder stack according to one or more outputs from the proposal decoder, and a decoder stack coupled to the masking unit and configured to receive the masked one or more outputs from the encoder stack. Generating the dense captioning based on one or more outputs of the decoder stack. In some embodiments, the one or more outputs from the proposal decoder include a differentiable mask. In some embodiments, during training, error in the dense captioning is back propagated to the decoder stack, the encoder stack, and the proposal decoder.

Abstract: A system is provided for natural language processing. In some embodiments, the system includes an encoder for generating context-specific word vectors for at least one input sequence of words. The encoder is pre-trained using training data for performing a first natural language processing task. A neural network performs a second natural language processing task on the at least one input sequence of words using the context-specific word vectors. The first natural language process task is different from the second natural language processing task and the neural network is separately trained from the encoder.

Abstract: The technology disclosed provides a so-called “joint many-task neural network model” to solve a variety of increasingly complex natural language processing (NLP) tasks using growing depth of layers in a single end-to-end model. The model is successively trained by considering linguistic hierarchies, directly connecting word representations to all model layers, explicitly using predictions in lower tasks, and applying a so-called “successive regularization” technique to prevent catastrophic forgetting. Three examples of lower level model layers are part-of-speech (POS) tagging layer, chunking layer, and dependency parsing layer. Two examples of higher level model layers are semantic relatedness layer and textual entailment layer. The model achieves the state-of-the-art results on chunking, dependency parsing, semantic relatedness and textual entailment.

Abstract: Embodiments described herein provide a two-stage online detection of action start system including a classification module and a localization module. The classification module generates a set of action scores corresponding to a first video frame from the video, based on the first video frame and video frames before the first video frames in the video. Each action score indicating a respective probability that the first video frame contains a respective action class. The localization module is coupled to the classification module for receiving the set of action scores from the classification module and generating an action-agnostic start probability that the first video frame contains an action start.

Abstract: Approaches for multitask learning as question answering include an input layer for encoding a context and a question, a self-attention based transformer including an encoder and a decoder, a first bi-directional long-term short-term memory (biLSTM) for further encoding an output of the encoder, a long-term short-term memory (LSTM) for generating a context-adjusted hidden state from the output of the decoder and a hidden state, an attention network for generating first attention weights based on an output of the first biLSTM and an output of the LSTM, a vocabulary layer for generating a distribution over a vocabulary, a context layer for generating a distribution over the context, and a switch for generating a weighting between the distributions over the vocabulary and the context, generating a composite distribution based on the weighting, and selecting a word of an answer using the composite distribution.

Abstract: Systems and methods are provided for weakly supervised natural language localization (WSNLL), for example, as implemented in a neural network or model. The WSNLL network is trained with long, untrimmed videos, i.e., videos that have not been temporally segmented or annotated. The WSNLL network or model defines or generates a video-sentence pair, which corresponds to a pairing of an untrimmed video with an input text sentence. According to some embodiments, the WSNLL network or model is implemented with a two-branch architecture, where one branch performs segment sentence alignment and the other one conducts segment selection.

Abstract: Embodiments described herein provide a pipelined natural language question answering system that improves a BERT-based system. Specifically, the natural language question answering system uses a pipeline of neural networks each trained to perform a particular task. The context selection network identifies premium context from context for the question. The question type network identifies the natural language question as a yes, no, or span question and a yes or no answer to the natural language question when the question is a yes or no question. The span extraction model determines an answer span to the natural language question when the question is a span question.

Abstract: A method for evaluating robustness of one or more target neural network models using natural typos. The method includes receiving one or more natural typo generation rules associated with a first task associated with a first input document type, receiving a first target neural network model, and receiving a first document and corresponding its ground truth labels. The method further includes generating one or more natural typos for the first document based on the one or more natural typo generation rules, and providing, to the first target neural network model, a test document generated based on the first document and the one or more natural typos as an input document to generate a first output. A robustness evaluation result of the first target neural network model is generated based on a comparison between the output and the ground truth labels.