Abstract: A system monitors a vehicle navigating in an environment using a pre-trained machine learning model that is configured to make navigation decisions during autonomous driving of the vehicle. The system identifies a first simulated trajectory generated by the pre-trained machine learning model and a first real trajectory driven by the vehicle. In response to determining that a trajectory type of the first real trajectory is one of a plurality of trajectory types that trigger fine-tuning of the pre-trained machine learning model, the system calculates a discrepancy value between the first real trajectory and the first simulated trajectory. In response to determining that the discrepancy value is greater than a threshold discrepancy value, the system updates weights of the pre-trained machine learning model and executes the pre-trained machine learning model with the updated weights.

Abstract: Disclosed herein are systems and methods for providing LLM embeddings. The method also includes receiving a query to a LLM service. The method also includes determining if a first embedding of the input query is provided by a main embedding model. The method further includes, based on a determination that the first embedding is provided by the main embedding model, identifying an embeddings cluster similar to the input query, and computing a main centroid embedding for the embeddings cluster based on the first embedding. The method further includes based on a determination that the first embedding is not provided by the main embedding model, identifying an embeddings cluster similar to the input query, and transmitting the main centroid embedding of the embeddings cluster to the LLM service.

Abstract: Systems and methods include detecting obstacles and drivable areas by an autonomous vehicle by inputting image and map data into a neural network to extract feature vectors. A transformer encoder converts these vectors from camera space to Bird's Eye View (BEV) space. A detection head identifies objects, and a segmentation head generates a BEV map showing objects and drivable surfaces. Attributes from both heads are compared, and the segmentation head's weights are updated accordingly, resulting in an updated BEV segmentation map output by the updated segmentation head.

Abstract: Disclosed herein are systems and method for repurposing a machine learning model. An exemplary method includes: receiving a first training dataset; determining an input portion and an output portion in an entry of the first training dataset; comparing the first training dataset to a second training dataset used to train a machine learning model, wherein the comparing includes determining a similarity score between the input portion and the output portion of the first training dataset and an input portion and an output portion of the second training dataset; in response to determining that the similarity score is greater than a threshold similarity score, re-training the machine learning model using the first training dataset; and executing the re-trained machine learning model on an input value to generate an output value corresponding to the first training dataset.

Abstract: Disclosed herein are systems and methods for providing LLM embeddings. The method also includes receiving a query to a LLM service. The method also includes determining if a first embedding of the input query is provided by a main embedding model. The method further includes, based on a determination that the first embedding is provided by the main embedding model, identifying an embeddings cluster similar to the input query, and computing a main centroid embedding for the embeddings cluster based on the first embedding. The method further includes based on a determination that the first embedding is not provided by the main embedding model, identifying an embeddings cluster similar to the input query, and transmitting the main centroid embedding of the embeddings cluster to the LLM service.

Abstract: A system obtains a weight matrix associated with a linear operation, the weight matrix comprising a first non-zero value and a second non-zero value that is a negative value. The system decomposes the weight matrix into a first binary matrix indicating positions of the first non-zero value and a second binary matrix indicating positions of the second non-zero value. The system encrypts vector elements of an input vector using an encryption scheme to obtain an encrypted input vector. For each row of the first and second binary matrices, the system homomorphically processes selected encrypted elements of the encrypted input vector to generate at least two intermediate encrypted aggregates corresponding to respective contribution signs. The system generates an output element for each row based on a combination of the intermediate encrypted aggregates.

Abstract: Disclosed herein are systems and methods for proctoring online examinations by determining a gaze of a user. The method includes: performing a calibration process to calibrate a field of view of a user taking an online examination on a computer; after the calibration process is completed, obtaining a gaze origin and a gaze vector of the user using a prepared neural network to obtain facial key points of the user and determining, by a webcam or camera, the gaze of the user in three-dimensional (3D) coordinates to determine whether the gaze vector of the user is within a defined boundary relative to the gaze origin; and based on a determination that the gaze vector of the user is outside the defined boundary for a period of time, transmitting a message that the gaze of the user is outside of the defined boundary.

Abstract: A system receives, via a user interface (UI), an input specification indicating a topic and properties for generating a custom course comprising a plurality of content blocks. For each respective content block of the plurality of content blocks, the system generates, using a machine learning, the respective content block that describes a portion of the topic; prior to generating a subsequent content block of the plurality of content blocks, the system analyzes a set of properties associated with the respective content block to assess whether the respective content block includes information beyond a scope of the topic as indicated in the input specification, in response to determining that the respective content block includes the information, the system modifies the respective content block; and in response to determining that the respective content block after modification does not include the information, the system generates the subsequent content block.

Abstract: Systems and methods for web UI development, including supporting and rendering UI using multiple frameworks and migrating between different frameworks. Systems and methods can train different adapters for converting UI models in declarative abstract view into framework-specific code and render UI from the code using multiple or optimal frameworks.

Abstract: Application integration within a cloud platform system based on declarative callback configuration including registering an application at a vendor portal module to acquire a unique vendor and application identification, declaring callbacks associated with application events, each with a callback type identifier and corresponding data structure, and initiating an API callback by a client module in response to an event, and forwarding the callback to an API Callback Gateway, which verifies the API callback for compliance, encapsulates it with authentication data, and transmits it to a designated callback handler within an ISV infrastructure, processing the callback by the handler, sending a structured response back to the Gateway, which validates and provides the response to a client module, and invoking the cloud platform API using the registered application identifier to trigger the API callback.

Abstract: A system determines whether a first operation performed by an MLM is compatible with one of a first encryption scheme and a second encryption scheme, wherein the MLM is distributed over at least one client device and at least one server. In response to determining that the first operation is compatible with the first encryption scheme, the system: encrypts data associated with the first operation using the first encryption scheme; and transmits the data encrypted by the first encryption scheme to the at least one server configured to apply the first operation. In response to determining that the first operation is incompatible with the first encryption scheme, the system: encrypts the data associated with the first operation using the second encryption scheme; and transmits the data encrypted by the second encryption scheme to the at least one server configured to apply the first operation.

Abstract: Aspects of the present disclosure include a method for verifying live user presence in an online session, comprising receiving first and second video streams capturing a user during the session from a first position and a different second position, respectively. The method further comprises initiating a challenge by providing, for presentation on a different second display of a second device, a code, and providing, for presentation on a first display, an instruction to the user to respond to the challenge using the second device. The method further comprises detecting, based on at least one of the video streams, physical actions of the user during the challenge, receiving, from the second device, user input events representing a user response to the challenge, and determining, based on at least one of the physical actions or the user input events, whether the user successfully completed the challenge.

Abstract: A method and system is disclosed for generating video content where speech audio is accurately synchronized with facial movements. The method begins by converting input text into speech audio using a text-to-speech (TTS) system. An original video is then processed to identify and extract the mouth area of the human face. The method generates a plurality of frames of the mouth area corresponding to segments of the speech audio using a trained generative neural network. These frames are lip-synced with the corresponding audio segments. The extracted mouth area frames are overlaid onto the original video, followed by performing facial reconstruction using a trained facial reconstruction model to correct distortions and achieve a natural and expressive appearance around the mouth area. Finally, the generated speech audio is combined with the modified video to produce a final video with synchronized lip movements and phonetic sounds.

Abstract: Aspects of the present disclosure include a method for verifying live user presence in an online session, comprising obtaining at least one video stream of a user captured during at least one previous presentation of at least one user interface (UI) on a display, obtaining input events representing previous user interactions with the UI, generating a behavioral digital fingerprint profile based on the video stream and the input events, initiating an online exam during the online session by providing, for presentation on the display, at least one additional UI, receiving at least one additional video stream of the user during the exam, receiving additional input events representing user interactions with the additional UI, determining exam-specific behavioral features based on the additional video stream and the additional input events, and verifying whether the user is taking the exam based on at least one of the features or the profile.

Abstract: A system receives a user request to mount a child frontend application comprised in a primary frontend application. The system, in response to receiving the user request, checks whether the child frontend application is in a cache. The system, in response to determining that the child frontend application is not in the cache, mounts the child frontend application and initializes the child frontend application using a first lifecycle function that performs initial setup for the child frontend application. The system stores the child frontend application in the cache using a second lifecycle function for saving a current state of the child frontend application.

Abstract: Virtual file honeypots (VFHs) are deployed to protect real files in a computing system. The storage footprint is reduced because the virtual file honeypots do not require storage resources. VFHs are generated to appear the same as other files in a computing system, presenting an attractive attack surface as bait to reveal malicious processes and agents. The VFHs are generated on-demand and are not backed up in system storage. The use of VFHs requires less storage capacity than real honeypot files and conserves computing system resources.

Abstract: A system generates, for display on a computing device, an avatar based on an avatar model specifying visual characteristics and movement capabilities of the avatar. The system receives a request to reduce computational costs for generating the avatar by a specified amount. The system executes a machine learning model trained to select one or more of the visual characteristics and the movement capabilities of the avatar to adjust based on available computational resources and computational costs to utilize the available computational resources. The system adjusts the one or more of the visual characteristics and the movement capabilities of the avatar outputted by the machine learning model in response to the request.

Abstract: Disclosed herein are systems and methods for automated grading of written response tests. The method includes obtaining course material, test materials for a plurality of courses offered by an academic institution, a standardized scoring rubric for a plurality of tests and different courses, and grading guidelines comprising test scoring rules for the written response test or course. The method further includes analyzing the course materials and test materials using a trained rubric customizer LLM. The method also includes obtaining a written response from a learner for a test in a course taken by the learner. The method further includes analyzing the written response using two or more trained test grading LLM agents executing in parallel with each other. The method further includes combining, by a scoring engine, a plurality of criteria scores from the one or more trained test grading LLM agents. The method further includes generating a grading report.

Abstract: Systems and methods provide a multidimensional data structure to tailor data storage and corresponding user output. Tailored graph data structures reduce the level of hallucination in composing an answer to a user using a large language model. In response to a user query of documents, the user query is translated to a vector. From the vector, the most relevant pieces of documents that are as similar as possible in semantics to the question of a person can be determined and communicated to the user. In one aspect, using the vector base (vector query applied to the documents), a graph can be generated in the form of nodes and connections between data in the documents.

Abstract: Disclosed herein are systems and method for a dynamic knowledge graph (KG) augmentation of an LLM. The method includes obtaining a LLM chat history between a user and the LLM enhanced with a KG database. The LLM chat history includes user queries, relevant knowledge data retrieved from the KG database, and LLM answers for each user query. The LLM answers are based at least in part on relevant knowledge data retrieved from the KG database. The method also includes applying a KG update MLM agent trained to: analyze the LLM chat history, KG database and the graph schema and to identify: (i) a missing knowledge data from the KG database, and/or (ii) a missing relationship between knowledge data nodes in the KG database. The method further includes after identifying missing knowledge data and/or the missing relationship, updating the KG database with new knowledge data and/or new relationship between nodes.