Abstract: The technology relates to methods for detecting and classifying emotions in textual communication, and using this information to suggest graphical indicia such as emoji, stickers or GIFs to a user. Two main types of models are fully supervised models and few-shot models. In addition to fully supervised and few-shot models, other types of models focusing on the back-end (server) side or client (on-device) side may also be employed. Server-side models are larger-scale models that can enable higher degrees of accuracy, such as for use cases where models can be hosted on cloud servers where computational and storage resources are relatively abundant. On-device models are smaller-scale models, which enable use on resource-constrained devices such as mobile phones, smart watches or other wearables (e.g., head mounted displays), in-home devices, embedded devices, etc.

Abstract: The technology relates to methods for detecting and classifying emotions in textual communication, and using this information to suggest graphical indicia such as emoji, stickers or GIFs to a user. Two main types of models are fully supervised models and few-shot models. In addition to fully supervised and few-shot models, other types of models focusing on the back-end (server) side or client (on-device) side may also be employed. Server-side models are larger-scale models that can enable higher degrees of accuracy, such as for use cases where models can be hosted on cloud servers where computational and storage resources are relatively abundant. On-device models are smaller-scale models, which enable use on resource-constrained devices such as mobile phones, smart watches or other wearables (e.g., head mounted displays), in-home devices, embedded devices, etc.

Abstract: The technology relates to methods for detecting and classifying emotions in textual communication, and using this information to suggest graphical indicia such as emoji, stickers or GIFs to a user. Two main types of models are fully supervised models and few-shot models. In addition to fully supervised and few-shot models, other types of models focusing on the back-end (server) side or client (on-device) side may also be employed. Server-side models are larger-scale models that can enable higher degrees of accuracy, such as for use cases where models can be hosted on cloud servers where computational and storage resources are relatively abundant. On-device models are smaller-scale models, which enable use on resource-constrained devices such as mobile phones, smart watches or other wearables (e.g., head mounted displays), in-home devices, embedded devices, etc.

Abstract: Implementations relate to automatic response suggestions based on images received in messaging applications. In some implementations, a computer-executed method includes detecting a first image included within a first message received at a second device over a communication network from a first device of a first user, and programmatically analyzing the first image to extract a first image content. The method includes retrieving a first semantic concept associated with the first image content, programmatically generating a suggested response to the first message based on the first semantic concept, and transmitting instructions causing rendering of the suggested response in the messaging application as a suggestion to a second user of the second device.

Abstract: A computing device may receive a communication sent from an external computing device. At least one processor of the computing device may determine, using an on-device machine-trained model and based at least in part on the communication, one or more candidate responses to the communication. The at least one processor may receive an indication of a user input that selects a candidate response from the one or more candidate responses. Responsive to receiving the indication of the user input that selects the candidate response, the at least one processor may send the candidate response to the external computing device.

Abstract: Implementations relate to automatic response suggestions based on images received in messaging applications. In some implementations, a computer-executed method includes detecting a first image included within a first message received at a second device over a communication network from a first device of a first user, and programmatically analyzing the first image to extract a first image content. The method includes retrieving a first semantic concept associated with the first image content, programmatically generating a suggested response to the first message based on the first semantic concept, and transmitting instructions causing rendering of the suggested response in the messaging application as a suggestion to a second user of the second device.

Abstract: Methods, apparatus, and computer readable media related to determining graphical element(s) (e.g., emojis, GIFs, stickers) for inclusion in an electronic communication being formulated by a user via a computing device of the user, and providing the graphical element(s) for inclusion in the electronic communication. For example, the graphical element(s) may be provided for presentation to the user via a display of the computing device of the user and, in response to user interface input directed to one of the graphical element(s), that graphical element may be incorporated in the electronic communication. In various implementations, the electronic communication is a communication to be submitted as part of a dialog that involves the user and one or more additional users.

Abstract: Implementations relate to automatic response suggestions based on images received in messaging applications. In some implementations, a computer-executed method includes detecting a first image included within a first message received at a second device over a communication network from a first device of a first user, and programmatically analyzing the first image to extract a first image content. The method includes retrieving a first semantic concept associated with the first image content, programmatically generating a suggested response to the first message based on the first semantic concept, and transmitting instructions causing rendering of the suggested response in the messaging application as a suggestion to a second user of the second device.

Abstract: Implementations relate to automatic response suggestions to images included in received messages. In some implementations, a computer-implemented method includes detecting an image posted within a first message by a first user, and programmatically analyzing the image to determine a feature vector representative of the image. The method programmatically generates one or more suggested responses to the first message based on the feature vector, each suggested response being a conversational reply to the first message. Generating the suggested responses includes determining probabilities associated with word sequences for the feature vector using a model trained with previous responses to previous images, and selecting one or more of the word sequences based on the associated probabilities. The suggested responses are determined based on the selected word sequences. The method causes the suggested responses to be rendered in the messaging application as one or more suggestions to a second user.

Abstract: Implementations relate to automatic response suggestions based on images received in messaging applications. In some implementations, a computer-executed method includes detecting a first image included within a first message received at a second device over a communication network from a first device of a first user, and programmatically analyzing the first image to extract a first image content. The method includes retrieving a first semantic concept associated with the first image content, programmatically generating a suggested response to the first message based on the first semantic concept, and transmitting instructions causing rendering of the suggested response in the messaging application as a suggestion to a second user of the second device.

Abstract: Implementations relate to automatic response suggestions to images included in received messages. In some implementations, a computer-implemented method includes detecting an image posted within a first message by a first user, and programmatically analyzing the image to determine a feature vector representative of the image. The method programmatically generates one or more suggested responses to the first message based on the feature vector, each suggested response being a conversational reply to the first message. Generating the suggested responses includes determining probabilities associated with word sequences for the feature vector using a model trained with previous responses to previous images, and selecting one or more of the word sequences based on the associated probabilities. The suggested responses are determined based on the selected word sequences. The method causes the suggested responses to be rendered in the messaging application as one or more suggestions to a second user.

Abstract: Implementations relate to automatic response suggestions based on images received in messaging applications. In some implementations, a computer-executed method includes detecting a first image included within a first message received at a second device over a communication network from a first device of a first user, and programmatically analyzing the first image to extract a first image content. The method includes retrieving a first semantic concept associated with the first image content, programmatically generating a suggested response to the first message based on the first semantic concept, and transmitting instructions causing rendering of the suggested response in the messaging application as a suggestion to a second user of the second device.

Abstract: Methods for cryptographic delegation and enforcement of dynamic access to stored data are disclosed. An example method includes generating for a first modified data block, a new per-block hash value using as a hash function input data contained in the first modified data block or a new per-block hash message authentication code (HMAC) using as hash function inputs a new per-block hash key and data contained in the first modified data block, writing the new per-block hash value or the new per-block HMAC to data block metadata associated with the modified data block in the protected data object, and writing the first modified data block to one of the data blocks of the protected data object.

Abstract: Methods, apparatus, and computer readable media related to determining graphical element(s) (e.g., emojis, GIFs, stickers) for inclusion in an electronic communication being formulated by a user via a computing device of the user, and providing the graphical element(s) for inclusion in the electronic communication. For example, the graphical element(s) may be provided for presentation to the user via a display of the computing device of the user and, in response to user interface input directed to one of the graphical element(s), that graphical element may be incorporated in the electronic communication. In various implementations, the electronic communication is a communication to be submitted as part of a dialog that involves the user and one or more additional users.

Abstract: A computing device may receive a communication sent from an external computing device. At least one processor of the computing device may determine, using an on-device machine-trained model and based at least in part on the communication, one or more candidate responses to the communication. The at least one processor may receive an indication of a user input that selects a candidate response from the one or more candidate responses. Responsive to receiving the indication of the user input that selects the candidate response, the at least one processor may send the candidate response to the external computing device.

Abstract: Implementations relate to automatic response suggestions based on images received in messaging applications. In some implementations, a computer-executed method includes detecting a first image included within a first message received at a second device over a communication network from a first device of a first user, and programmatically analyzing the first image to extract a first image content. The method includes retrieving a first semantic concept associated with the first image content, programmatically generating a suggested response to the first message based on the first semantic concept, and transmitting instructions causing rendering of the suggested response in the messaging application as a suggestion to a second user of the second device.

Abstract: Methods for cryptographic delegation and enforcement of dynamic access to stored data are disclosed. An example method includes generating for a first modified data block, a new per-block hash value using as a hash function input data contained in the first modified data block or a new per-block hash message authentication code (HMAC) using as hash function inputs a new per-block hash key and data contained in the first modified data block, writing the new per-block hash value or the new per-block HMAC to data block metadata associated with the modified data block in the protected data object, and writing the first modified data block to one of the data blocks of the protected data object.

Abstract: Efficient methods for assigning, revoking, and realizing access to stored data involve a cryptographic key hierarchy and a set of operations performed on cryptographic keys and performed on the data objects to be protected. In addition to providing confidentiality and integrity for data objects, the methods allow access to selected data objects to be permanently revoked for all entities without requiring all instances of the data objects to be destroyed or overwritten. The methods also support access right modifications for a data object without requiring the re-encryption of the entire data object; instead, certain keys are selectively re-encrypted and re-authenticated to implement access control changes. The key hierarchy is parameterized to enable flexible performance tuning, and to provide efficient random access, keying and other security operations are performed for individual blocks within a data object rather than only for the entire data object.

Abstract: Efficient methods for assigning, revoking, and realizing access to stored data involve a cryptographic key hierarchy and a set of operations performed on cryptographic keys and performed on the data objects to be protected. In addition to providing confidentiality and integrity for data objects, the methods allow access to selected data objects to be permanently revoked for all entities without requiring all instances of the data objects to be destroyed or overwritten. The methods also support access right modifications for a data object without requiring the re-encryption of the entire data object; instead, certain keys are selectively re-encrypted and re-authenticated to implement access control changes. The key hierarchy is parameterized to enable flexible performance tuning, and to provide efficient random access, keying and other security operations are performed for individual blocks within a data object rather than only for the entire data object.

Abstract: Efficient methods for implementing security and manageability for stored data objects involve logical object reorganization, computation and injection of metadata, and specialized data access operations. Methods for utilization and incorporation of cryptographic key hierarchies and security functions for data objects are disclosed. Cryptographic keying and other data management operations may be performed for individual blocks within a data object rather than only for the entire data object in order to achieve performance objectives.