Abstract: Disclosed herein are systems and methods for graph-based dynamic information retrieval and synthesis, particularly for enhancing the response generation of chatbot applications using large language models (LLMs). The methods involve receiving a user query, generating a query embedding, and accessing a knowledge graph comprising nodes representing documents with associated document embeddings. A set of documents is retrieved based on similarity scores, and a distance matrix is constructed to calculate shortest-path distances between nodes. Documents are clustered to identify a largest cluster from which content is retrieved to formulate a response. The disclosed embodiments improve upon retrieval augmented generation (RAG) by leveraging the hierarchical relationships of a knowledge graph and node interconnectedness to provide contextually coherent information and mitigate the inclusion of information from irrelevant or superficially related documents in LLM responses.

Abstract: Embodiments are disclosed for an automatic identification of a driver. In one example, an in-vehicle computing system of a vehicle includes a sensor interface communicatively coupled to one or more data collection devices, a processor, and a storage device storing instructions executable by the processor to generate a two-dimensional current driver profile matrix for a current driver of the vehicle, the current driver profile matrix indicating a vehicle operating status in terms of a pair of driving parameters at sampled times during a current vehicle trip, and determine a probability that the current driver is one of a plurality of known drivers by comparing the current driver profile matrix to one or more stored driver profile matrices associated with the plurality of known drivers.

Abstract: Device component status detection and illustration apparatuses, methods, and systems determine and generate visualizations of updates timelines indicating device components associated with a remote connected device at different times. A device component status detection and illustration apparatus may include a memory and processor with instructions to: obtain device selection parameters, determine one or more remote connected devices that satisfy the device selection parameters, and identify a remote connected device selected from one or more remote connected devices by a user. The instructions may further be executed to generate visualizations of updates timelines associated with the remote connected device, including information regarding device components associated with the identified remote connected device as of selected update times. This allows the apparatus to present information regarding the status of a particular device at different points in time.

Abstract: The Remote Embedded Device Update Platform Apparatuses, Methods and Systems (“REDUP”) transforms telemetry inputs via REDUP components into remote embedded updates outputs. The REDUP may include a memory and processor with instructions to: obtain a remote embedded device connection request message from a remote embedded device and analyze the message to determine a version of embedded instructions on the remote embedded device. With that, the REDUP may determine if other remote embedded devices similar to the remote embedded device have provided request messages by searching a remote embedded device connection request message database. This allows the REDUP to determine if a potential issue requiring updates on the remote embedded device exists. With that, the REDUP may determine and provide an update for the remote embedded device.