Abstract: Approaches, techniques, and mechanisms are disclosed for flexible and dynamic modification of elements within textual content to provide an enhanced interface for development or other activities with respect to those elements. In an embodiment, the elements may be augmented with links by which users may quickly access associated items in development tools or other resources. For example, in an embodiment, the elements may be specific lines or blocks of code mentioned in a stack trace or error message. The elements may be transformed into links that, for instance, open the file containing the mentioned line or block of code, and optionally scroll specifically to the mentioned line or block of code. In an embodiment, the elements are specific passages within web pages that match trigger conditions specified by rules encapsulated in a user-customizable client-side plug-in, extensions, or bookmarklet. Proxy-based or server-based implementations may also be used.

Abstract: Methods and systems are provided for automatically determining a need to invoke functionality of a feature of an application while using a primary application and for automatically invoking that functionality without user action. When input information is input into the primary application via a user interface of a user system, it can be received by an application server from a sender process at the user system, and a prefix that occurs before a text string can be identified. The text string that follows the prefix can be extracted by performing a matching analysis against a library of keywords and extracting a matching keyword. The text string comprises extracted text from the input information. The matching keyword can be matched to an entry that is mapped to a command that can be executed to generate an action that automatically invokes the functionality of the feature of the application without user action.

Abstract: Sentiment analysis is a task in natural language processing. The embodiments are directed to using a generative language model to extract an aspect term, aspect category and their corresponding polarities. The generative language model may be trained as a single, joint, and multi-task model. The single-task generative language model determines a term polarity from the aspect term in the sentence or a category polarity from an aspect category in the sentence. The joint-task generative language model determines both the aspect term and the term polarity or the aspect category and the category polarity. The multi-task generative language model determines the aspect term, term polarity, aspect category and category polarity of the sentence.

Abstract: A method for maintaining a dialogue state associated with a dialogue between a user and a digital system includes receiving, by a dialogue state tracker associated with the digital system, a representation of a user communication, updating, by the dialogue state tracker, the dialogue state and providing a system response based on the updated dialogue state. The dialogue state is updated by evaluating, based on the representation of the user communication, a plurality of member scores corresponding to a plurality of ontology members of an ontology set, and selecting, based on the plurality of member scores, zero or more of the plurality of ontology members to add to or remove from the dialogue state.

Abstract: Some embodiments of the current disclosure disclose methods and systems for analyzing root causes of an incident disrupting information technology services such as cloud services. In some embodiments, a set of problem review board (PRB) documents including information about said incidents may be parsed using a natural language processing (NLP) neural model to extract structured PRB data from the unstructured investigative information contained in the PRB documents. The structured PRB data may include symptoms of the incident, root causes of the incident, resolutions of the incidents, etc., and a causal knowledge graph causally relating the symptoms, root causes, resolutions of the incidents may be generated.

Abstract: Approaches for cross-lingual regularization for multilingual generalization include a method for training a natural language processing (NLP) deep learning module. The method includes accessing a first dataset having a first training data entry, the first training data entry including one or more natural language input text strings in a first language; translating at least one of the one or more natural language input text strings of the first training data entry from the first language to a second language; creating a second training data entry by starting with the first training data entry and substituting the at least one of the natural language input text strings in the first language with the translation of the at least one of the natural language input text strings in the second language; adding the second training data entry to a second dataset; and training the deep learning module using the second dataset.

Abstract: Embodiments described herein provide dynamic blocking, a decoding algorithm which enables large-scale pretrained language models to generate high-quality paraphrases in an un-supervised setting. Specifically, in order to obtain an alternative surface form, when the language model emits a token that is present in the source sequence, the language model is prevented from generating the next token that is the same as the subsequent source token in the source sequence at the next time step. In this way, the language model is forced to generate a paraphrased sequence of the input source sequence, but with mostly different wording.

Abstract: Some embodiments of the current disclosure disclose methods and systems for batch active learning using the Shapley values of data points. In some embodiments, Shapley values of a first subset of labeled data are used to measure the contributions of the first subset of data to the performance of neural network. Further, a regression model that correlates the first subset of data to their Shapley values is trained to predict the Shapley values of a second subset of data that are unlabeled. A portion of the second subset of data may then be selected for labeling based on the predicted Shapley values.

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: Embodiments described herein provide an Adapt-and-Adjust (A2) mechanism for multilingual speech recognition model that combines both adaptation and adjustment methods as an integrated end-to-end training to improve the models' generalization and mitigate the long-tailed issue. Specifically, a multilingual language model mBERT is utilized, and converted into an autoregressive transformer decoder. In addition, a cross-attention module is added to the encoder on top of the mBERT's self-attention layer in order to explore the acoustic space in addition to the text space. The joint training of the encoder and mBERT decoder can bridge the semantic gap between the speech and the text.

Abstract: Embodiments described herein provide natural language processing (NLP) systems and methods that provide a customized summarization of scientific or technical articles, which disentangles background information from new contributions, and summarizes the background information or the new information (or both) based on a user's preference. Specifically, the systems and methods utilize machine learning classifiers to classify portions of sentences within the article as containing background information or as containing a new contribution attributable to the article. The systems and methods then incorporate the background information in the summary or incorporate the new contribution in the summary and output the summary. In this way, the systems and methods can provide summaries of scientific literatures, which largely accelerates literature review in scientific fields.

Abstract: Embodiments described herein a code generation and understanding model that builds on a Transformer-based encoder-decoder framework. The code generation and understanding model is configured to derive generic representations for programming language (PL) and natural language (NL) in code domain via pre-training on unlabeled code corpus, and then to benefit many code-related downstream tasks with fine-tuning. Apart from the denoising sequence-to-sequence objectives widely adopted for pre-training on natural language, identifier tagging and prediction pre-training objective is adopted to enable the model to better leverage the crucial token type information from PL, which specifically are the identifiers assigned by developers.

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: The system and method are directed to a prototypical contrastive learning (PCL). The PCL explicitly encodes the hierarchical semantic structure of the dataset into the learned embedding space and prevents the network from exploiting low-level cues for solving the unsupervised learning task. The PCL includes prototypes as the latent variables to help find the maximum-likelihood estimation of the network parameters in an expectation-maximization framework. The PCL iteratively performs an E-step for finding prototypes with clustering and M-step for optimizing the network on a contrastive loss.

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: Computing systems, database systems, and related methods are provided for managing data pertaining to electronic messages. A database system includes a database including a first object having a plurality of content fields corresponding to a first instance of an electronic message configured for a first communication channel and a server coupled to the database and a network to create a second database object corresponding to a second instance of the electronic message configured for a different communication channel, create a multichannel data structure in the database maintaining an association between the first and second objects, automatically populate a subset of content fields of the second object with values copied from the first object based on a mapping between the two communication channels, and thereafter use the second object to generate a version of the electronic message to be communicated to a recipient using the second communication channel.

Abstract: Devices, systems and methods are provided for remotely managing configuration of a networking device. One method of managing configuration of a networking device involves obtaining resources associated with one or more graphical user interface (GUI) displays of a web application for configuring the networking device, receiving a request to access the web application, and in response to the request, determining a representative value for the resources as a function of a current state of the resources, obtaining a reference value for the resources that reflects a validated state of the resources, and instantiating the web application using the resources when the representative value matches the reference value.

Abstract: A system for storing multiple petabytes of raw log data from cloud-based source data buckets into internal partitioned data storage in a data lake is provided. The system is configured to: deploy a log partitioner cluster comprising a plurality of log partitioner service instances for storing the raw log data in a partitioned manner; associate one or more of the source data buckets to each of a plurality of deployed log partitioner service instances by providing associated configuration metadata from a configuration service; fetch, via the log partitioner cluster, raw log data from associated source data buckets based on the associated configuration metadata, wherein the associated configuration metadata provides instructions for use by a log partitioner service instance to onboard raw log data; and place fetched raw log data, via the log partitioner cluster, in the internal partitioned log storage in accordance with the associated configuration metadata.

Abstract: In some embodiments, a method stores an executable contract in a ledger. The executable contract is for a test to be performed on a network and includes terms for the test. The method receives an indication of traffic that is received at a network device. It is determined that the traffic is associated with the executable contract that is stored in the ledger based on the terms of the test. The method responds to the network device with a response indicating that the traffic is associated with the executable contract. The network device allows the traffic to be sent to a destination on the network to perform an operation for the test.

Abstract: Disclosed herein are system, method, and computer program product embodiments for automatically detecting layout shifts or anomalies in a website during localization processes. A method includes generating a first rendering of a localization element on a website, extracting localized text from a localization descriptor for the localization element in code for the website, applying the localized text to the localization element on the website, generating a second rendering of the localization element on the website after the applying of the localized text, identifying an anomaly in the localization element displayed on the website by comparing the first and second renderings, and creating a visual indicator for display as an overlay outside a visible area of the website, wherein a location of the visual indicator corresponds to a location of the anomaly in the localization element displayed on the website.