Abstract: A method, a system, and a computer program product for content-based document access. A content of an electronic document is analyzed using a machine learning (ML) model. The ML model determines presence of a plurality sensitive data in the document. One or more document entity-based parameters are received. At least one sensitive data in the plurality of sensitive data is identified. At least one recipient computing device in a plurality of computing devices is prevented from receiving the document containing at least one sensitive data. The sensitive data is extracted from the document. The document is modified to redact the sensitive data and a modified electronic document is generated. The modified electronic document is transmitted to at least one recipient computing device.

Abstract: A method, a system, and a computer program product for federated document learning. A plurality of representations of a plurality of datasets from a plurality of client systems are received. Each representation corresponds to a dataset associated with a client system and is generated using a first machine learning (ML) model. A second ML model is applied to the plurality of representations to generate a combined representation of the plurality of datasets. Data from each dataset is not provided to the second ML model. The combined representation is filtered using one or more filtering parameters to generate a filtered representation. Using the second ML model, one or more model weights for training a third ML model in a plurality of third ML models are generated. Each third ML model is associated with a respective client system. The model weights are provided to the plurality of third ML models.

Abstract: A method, a system, and a computer program product for generating a benchmarking dataset. A type of an electronic document is determined. Using the type of the electronic document, a first request to generate one or more labels for the electronic document is generated. Using a content of the electronic document, a second request to generate one or more labels for the electronic document is generated. The electronic document and the first and second requests are sent to a generative artificial intelligence (AI) model. The generative AI model generates one or more first labels for the electronic document based on the first request and one or more second labels for the electronic document based on the second request. Using one or more first and second labels, one or more labels for the electronic document are generated.

Abstract: Techniques are disclosed for using a machine learning model to generate document templates from input templates. For example, a computing system receives an input document. A machine learning model of the computing system processes the input document to identify one or more text items corresponding to respective variable field types of the plurality of variable field types. The computing system creates, for each text item of the one or more text items, a variable field for the variable field type of the plurality of variable field types corresponding to the text item and creates, for each variable field of the one or more created variable fields, a mapping for the variable field to a corresponding data source. The computing system generates, based on the input document, a document template comprising the one or more created variable fields and the respective one or more mappings.

Abstract: Techniques are described for deploying software applications associated with document management. The techniques may include retrieving, by a computing system and from a software repository, configuration information for a software application for managing an electronic document, the configuration information specifying a plurality of widgets for deploying the software application. The techniques may further include receiving, by the computing system and from a client device, a request to access the software application. The techniques may further include generating, by the computing system, a response to the request comprising an indication of the configuration information and instructions configured to cause the client device to request, from the software repository, the plurality of widgets specified by the configuration information. The techniques may further include outputting, by the computing system, the response to the client device to provide the software application at the client device.

Abstract: A method, a system, and a computer program product for generating a benchmarking dataset. One or more queries for generation of one or more summaries of one or more electronic documents are received. The queries are modified using one or more parameters associated with the electronic documents to generate modified queries. The electronic documents are sent to a generative artificial intelligence (AI) model. The generative AI model generates summaries of the electronic documents based on at least one of: the initial queries and the modified queries. One or more labels for the electronic documents are generated using the summaries.

Abstract: A system, for example, a document management system stores documents and manages workflows associated with documents. The document management system allows discovery of templates based on explicit searches performed by users or automatic searches performed based on a context. The document management system allows generating new document templates based on selected versions of document component templates. The generated document template may be stored as a new version. The document management system instantiates documents based on templates by predicting values of variables used in the template based on various factors that describe the context in which the template is being used. The values used for instantiating variables may be generated using machine learning models that may be trained using historical data stored in the document management system.

Abstract: A method, an apparatus, and a computer-readable storage medium for searching of electronic documents. A search query is encoded to form a search vector. The search query includes at least one search term and seeks information within a plurality of electronic documents. A knowledge graph structure is generated using at least one search term. The search term is associated with at least another term in a plurality of terms in the knowledge graph structure. The search query is modified using the other term to generate a modified search vector. A response to the search query is generated based on one or more responsive document vectors retrieved from the plurality of electronic documents and that are semantically similar to the modified search vector.

Abstract: An electronic document system can allow users to upload a document package containing multiple individual component documents. Each component document includes a subset of a plurality of pages that are included in the document package. The electronic document system identifies a page of each component document by applying a machine learning model to the document package. The electronic document system partitions the document package into the individual component documents based on the identified pages. For each individual component document, the electronic document system identifies a document topic corresponding to the component document by applying another machine learning model. The electronic document system modifies a user interface to display each component document and corresponding document topic.

Abstract: An online document system can allow users to participate in collaborative negotiation of documents stored with in the online document system. To facilitate negotiations between multiple entities (each potentially including multiple users with different roles in the negotiation), the online document system includes a permissions system that allows per-clause control over user's access to propose and approve changes to a negotiated document. Similarly, the user interfaces for viewing a negotiated document provided by the online document system to users can depend on the role of that user in editing the document and the current stage of the negotiation of the document. Finally, the online document system can track agreement between sides in a negotiation on a per-clause basis and can otherwise help facilitate the negotiation of the document.

Abstract: Through a workflow interface, a document management system enables a user to establish a document workflow. The document workflow establishes a set of document actions for the document management system to perform automatically in response to a request from the user. The user defines the document workflow by selecting from a set of workflow operations and specifying parameters for the selected workflow operation. The user additionally defines an approval workflow operation and a rejection workflow operation that designate document actions in the event that a document is approved or rejected, respectively. The document management system performs at least a portion of the document workflow in an order determined from the workflow interface.

Abstract: A document management system sends a distinct copy of a set of documents to each of a plurality of recipients in a bulk sending process. Each copy is associated with one or more actions to be taken by a corresponding recipient. The document management system generates an interface that displays, for each copy of the set of documents, the recipient of the copy and a status of the copy. The document management system receives an identification of a subset of the copies of the set of documents. The document management system performs a bulk action on each of the subset of the copies without performing the bulk action on any remaining copies of the set of documents.

Abstract: A document execution engine that receives a document for execution within a document execution environment. The document execution engine may also detect activity within the document execution environment associated with the received document, and apply the trained machine learned model to the detected activity to determine if the detected activity is representative of malicious behavior and, in response to determining that the detected activity is representative of malicious behavior, to identify remedial actions that can mitigate the malicious behavior. The document execution engine may also provide, to a device of a user, a recommendation to perform the identified remedial actions.

Abstract: Techniques are described for deploying software applications associated with document management. The techniques may include receiving a request, by a computing system and from a client device, to execute a software application for managing an electronic document, wherein the software application is configured to execute using at least a first widget and a second widget. The techniques may further include generating, based on the first widget comprising instructions to initiate an execution of the second widget, instructions for executing the software application at the client device, wherein the instructions are configured to cause the client device to generate contextual information for the first widget and wherein the second widget causes the client device to load the contextual information when executing the second widget.

Abstract: A method, an apparatus, and a computer-readable storage medium for generating a document graph. A plurality of electronic documents is received. Each electronic document has a predetermined document type. A machine learning model is selected from the plurality of machine learning models based on the predetermined document type. The selected machine learning model is instructed to extract a plurality of document portions from each electronic document in the plurality of electronic documents in accordance with the predetermined document type. A relationship between two or more document portions is defined based on a content of each document portion, and the document portions are associated based on the relationship. A graph structure having a plurality of nodes is generated. Each node includes at least one document portion. Each node is connected to another node in accordance with the relationship between document portions included in the nodes. The graph structure is stored.

Abstract: A document management system trains a machine-learned model using a first training set of tagged documents to, when applied to a document, tag one or more portions of the document. The document management system applies the machine-learned model to a target document. One or more portions of the target document incorrectly tagged by the machine-learned model are identified. A feature vector representative of the target document is generated. Each entry of the feature vector is representative of a characteristic of the target document without including private information from the target document. The document management system queries a corpus of documents using the feature vector to identify a set of documents that correspond to the feature vector. A second training set of tagged documents is generated using the identified set of documents. The document management system retrains the machine-learned model using the second training set of tagged documents.

Abstract: A document management system surfaces changes to a portable document format (PDF) document to a user. The document management system converts each page of the PDF document into images, segments those images, and processes each segment of those images using computer vision and/or natural language processing. The document management system compares segments from an original copy of the PDF document with segments from a modified copy of the PDF document to identify significant changes to the PDF document.

Abstract: A method, a system, and a computer program product for generating a benchmarking dataset. One or more queries for generation of one or more summaries of one or more electronic documents are received. The queries are modified using one or more parameters associated with the electronic documents to generate modified queries. The electronic documents are sent to a generative artificial intelligence (AI) model. The generative AI model generates summaries of the electronic documents based on at least one of: the initial queries and the modified queries. One or more labels for the electronic documents are generated using the summaries.

Abstract: A method, a system, and a computer program product for generating a benchmarking dataset. A type of an electronic document is determined. Using the type of the electronic document, a first request to generate one or more labels for the electronic document is generated. Using a content of the electronic document, a second request to generate one or more labels for the electronic document is generated. The electronic document and the first and second requests are sent to a generative artificial intelligence (AI) model. The generative AI model generates one or more first labels for the electronic document based on the first request and one or more second labels for the electronic document based on the second request. Using one or more first and second labels, one or more labels for the electronic document are generated.

Abstract: An online document system can allow users to participate in collaborative negotiation of documents stored with in the online document system. To facilitate negotiations between multiple entities (each potentially including multiple users with different roles in the negotiation), the online document system includes a permissions system that allows per-clause control over user's access to propose and approve changes to a negotiated document. Similarly, the user interfaces for viewing a negotiated document provided by the online document system to users can depend on the role of that user in editing the document and the current stage of the negotiation of the document. Finally, the online document system can track agreement between sides in a negotiation on a per-clause basis and can otherwise help facilitate the negotiation of the document.