An approach is described for addressing propagation of inaccurate information in a social networking environment. An associated method may include identifying inaccurate information within the social networking environment, facilitating creation of countering content to address the inaccurate information, and disseminating the countering content. The countering content may be determined by identifying behavior of one or more users among a plurality of users within the social networking environment. Identifying the inaccurate information within the social networking environment may include receiving information provided within the social networking environment. Upon determining that the received information is factual and thus objectively verifiable, it may be determined whether the received information matches analogous information verified as accurate. Upon determining that the received information does not match the analogous information verified as accurate, the received information may be marked as inaccurate.

An approach is described for addressing propagation of inaccurate information in a social networking environment. An associated method may include identifying inaccurate information within the social networking environment, facilitating creation of countering content to address the inaccurate information, and disseminating the countering content. The countering content may be determined by identifying behavior of one or more users among a plurality of users within the social networking environment. Identifying the inaccurate information within the social networking environment may include receiving information provided within the social networking environment. Upon determining that the received information is factual and thus objectively verifiable, it may be determined whether the received information matches analogous information verified as accurate. Upon determining that the received information does not match the analogous information verified as accurate, the received information may be marked as inaccurate.

Systems, devices, and methods discussed herein provide improved autonomous agent applications that are configured to provide explanations in response to user-submitted questions. Training data comprising a question, and an explanation pair may be accessed. A discourse tree and an explanation chain can be constructed from the explanation. The explanation chain may identify logical relationships between two entities of elementary discourse units identified from the discourse tree. A query may be submitted for the two entities, and a set of search results can be mined to identify text linking the two entities. An additional discourse tree can be generated from the text of a search result. The additional discourse tree can be combined with the original discourse tree to generate a complete discourse tree. A model may be trained using this augmented data (e.g., the complete discourse tree) to improve the quality of explanations provided by the autonomous agent application.

Systems, devices, and methods discussed herein provide improved autonomous agent applications that are configured to provide explanations in response to user-submitted questions. Training data comprising a question, and an explanation pair may be accessed. A discourse tree and an explanation chain can be constructed from the explanation. The explanation chain may identify logical relationships between two entities of elementary discourse units identified from the discourse tree. A query may be submitted for the two entities, and a set of search results can be mined to identify text linking the two entities. An additional discourse tree can be generated from the text of a search result. The additional discourse tree can be combined with the original discourse tree to generate a complete discourse tree. A model may be trained using this augmented data (e.g., the complete discourse tree) to improve the quality of explanations provided by the autonomous agent application.

Systems and methods for parsing natural language sentences using an artificial neural network (ANN) are described. Embodiments of the described systems and methods may generate a plurality of word representation matrices for an input sentence, wherein each of the word representation matrices is based on an input matrix of word vectors, a query vector, a matrix of key vectors, and a matrix of value vectors, and wherein a number of the word representation matrices is based on a number of syntactic categories, compress each of the plurality of word representation matrices to produce a plurality of compressed word representation matrices, concatenate the plurality of compressed word representation matrices to produce an output matrix of word vectors, and identify at least one word from the input sentence corresponding to a syntactic category based on the output matrix of word vectors.

An embodiment of a spoken intent detection device includes technology to detect a phrase in an electronic representation of an audio stream based on a pre-defined vocabulary, associate a time stamp with the detected phrase, and classify a spoken intent based on a sequence of detected phrases and the respective associated time stamps. Other embodiments are disclosed and claimed.

An embodiment of a spoken intent detection device includes technology to detect a phrase in an electronic representation of an audio stream based on a pre-defined vocabulary, associate a time stamp with the detected phrase, and classify a spoken intent based on a sequence of detected phrases and the respective associated time stamps. Other embodiments are disclosed and claimed.

Embodiments of the present disclosure include a method for token-position handling comprising: processing a first sequence of tokens to produce a second sequence of tokens, wherein the second sequence of tokens has a smaller number of tokens than the first sequence of tokens; masking at least some tokens in the second sequence to produce masked tokens; moving the masked tokens to the beginning of the second sequence to produce a third sequence; encoding tokens in the third sequence into a set of numeric vectors in a first array; and processing the first array in a transformer neural network to determine correlations among the third sequence, the processing the first array producing a second array.

Embodiments of the present disclosure include a method for token-position handling comprising: processing a first sequence of tokens to produce a second sequence of tokens, wherein the second sequence of tokens has a smaller number of tokens than the first sequence of tokens; masking at least some tokens in the second sequence to produce masked tokens; moving the masked tokens to the beginning of the second sequence to produce a third sequence; encoding tokens in the third sequence into a set of numeric vectors in a first array; and processing the first array in a transformer neural network to determine correlations among the third sequence, the processing the first array producing a second array.

Sentiment classification can be implemented by an entity-level multimodal sentiment classification neural network. The neural network can include left, right, and target entity subnetworks. The neural network can further include an image network that generates representation data that is combined and weighted with data output by the left, right, and target entity subnetworks to output a sentiment classification for an entity included in a network post.