Abstract: A framework has been created to filter queries for prompt injection attack detection. Application connectors collect data of SaaS interactions. The connectors associate, with the collected data, metadata that identifies the third-party application integrated with a primary SaaS application and an indication of whether collected data is sourced from an external user or an internal user. The filtering service periodically obtains a list of applications and risk scores and determines a risk score of the third-party application with the list. If the risk exceeds a threshold and/or the data origin is indicated as an external user, then the filtering service submits the data to a prompt injection attack detection service. Otherwise, the filtering service filters out the data from being submitted to the prompt injection attack detection service.

Abstract: The disclosed system generates “real-time” notifications to prevent cybersecurity violations while also effectively training users. The system captures text from a user interface and selects a set of task instructions based on whether the text corresponds to an outgoing or incoming communication. If the captured text is incoming, the system selects task instructions related to phishing. If the captured text is (intended) outgoing, then the system selects task instructions related to data leakage. The system forms a prompt with the selected task instructions and the captured text and then inputs the prompt to a generative language model. If the response from the generative language model indicates a cybersecurity violation, such as either phishing or potential data leakage, then the system generates a notification accordingly. The system also records generation of notifications per user to facilitate risk assessment.

Abstract: The disclosed system generates “real-time” notifications to prevent cybersecurity violations while also effectively training users. The system captures text from a user interface and selects a set of task instructions based on whether the text corresponds to an outgoing or incoming communication. If the captured text is incoming, the system selects task instructions related to phishing. If the captured text is (intended) outgoing, then the system selects task instructions related to data leakage. The system forms a prompt with the selected task instructions and the captured text and then inputs the prompt to a generative language model. If the response from the generative language model indicates a cybersecurity violation, such as either phishing or potential data leakage, then the system generates a notification accordingly. The system also records generation of notifications per user to facilitate risk assessment.

Abstract: The disclosed system generates “real-time” notifications to prevent cybersecurity violations while also effectively training users. The system captures text from a user interface and selects a set of task instructions based on whether the text corresponds to an outgoing or incoming communication. If the captured text is incoming, the system selects task instructions related to phishing. If the captured text is (intended) outgoing, then the system selects task instructions related to data leakage. The system forms a prompt with the selected task instructions and the captured text and then inputs the prompt to a generative language model. If the response from the generative language model indicates a cybersecurity violation, such as either phishing or potential data leakage, then the system generates a notification accordingly. The system also records generation of notifications per user to facilitate risk assessment.

Abstract: In implementations of systems for generating accessible color themes, a computing device implements an accessibility system to receive an input color palette including original colors defined in a color space. The accessibility system generates color vision deficiency simulations that correspond to pairs of the original colors and computes perceptual color differences between the color vision deficiency simulations. Candidate colors are determined for corresponding original colors based at least partially on the perceptual color differences and a conflicting perceptual color difference. The accessibility system outputs an output color palette including replacement colors defined in the color space that are generated at least partially based on distances between the candidate colors and the corresponding original colors computed in a CIELAB color space.

Abstract: In implementations of systems for generating accessible color themes, a computing device implements an accessibility system to receive an input color palette including original colors defined in a color space. The accessibility system generates color vision deficiency simulations that correspond to pairs of the original colors and computes perceptual color differences between the color vision deficiency simulations. Candidate colors are determined for corresponding original colors based at least partially on the perceptual color differences and a conflicting perceptual color difference. The accessibility system outputs an output color palette including replacement colors defined in the color space that are generated at least partially based on distances between the candidate colors and the corresponding original colors computed in a CIELAB color space.

Abstract: Apparatuses, methods, and systems disclosed herein that extracts grammatically meaningful phrases from property listing descriptions. In one example embodiment, a method is provided comprising receiving property descriptions, identifying phrase candidates by: parsing each phrase candidate into a set of word tokens, tagging each word token, and grouping the set of word tokens. Thereafter, computing a score for each phrase candidate and providing a list of property description recommendations comprising one or more phrase candidates with a high ranking score. The method further comprises predicting, via a machine learning algorithm, future trend value of each of the property description recommendations based on results from a plurality of historical property descriptions, and causing to display the list of property description recommendations in an order according to the predicted future trend value.

Abstract: An implantable medical device for releasing therapeutic agent having a medical device body and a plurality of reservoir-defining structures disposed on a surface of the body. A reservoir can be defined by the reservoir-defining structures and therapeutic agent may be located in the reservoir. A cover may extend over the reservoir so that the therapeutic agent is released from the reservoir when the medical device implanted. Methods for making the medical device may also include providing a medical device body, positioning a plurality of reservoir-defining structures on a surface of the body to form a reservoir, loading therapeutic agent into the reservoir, and covering the reservoir so that the therapeutic agent may release when the medical device is implanted. Alternatively, the reservoir may be covered with a cover and an opening formed in the cover so that the therapeutic agent may release when the medical device is implanted.

Abstract: According to an aspect of the present invention, implantable or insertable medical devices are provided which contain the following: (a) substrate having one or more depressions that contain at least one therapeutic agent and (b) a porous membrane disposed over the substrate and the one or more depressions, which regulate transport of chemical species between the therapeutic-agent-containing depressions and the exterior of the device. The substrate and the porous membrane are formed of different materials each with a different thermal expansion coefficient. Moreover, one of the substrate and the porous membrane at least partially surrounds the other. Other aspects of the present invention are directed to methods of making such medical devices, and methods of treatment using such medical devices.

Abstract: According to an aspect of the present invention, implantable or insertable medical devices are provided which contain the following: (a) substrate having one or more depressions that contain at least one therapeutic agent and (b) a porous membrane disposed over the substrate and the one or more depressions, which regulate transport of chemical species between the therapeutic-agent-containing depressions and the exterior of the device. The substrate and the porous membrane are formed of different materials each with a different thermal expansion coefficient. Moreover, one of the substrate and the porous membrane at least partially surrounds the other. Other aspects of the present invention are directed to methods of making such medical devices, and methods of treatment using such medical devices.

Abstract: The present invention relates to a device and method for delivery of a patch, graft, implant, therapeutic agent, or other device onto tissue using a rolled delivery mechanism. In one embodiment, a rolled delivery mechanism is provided for delivering a patch to tissue when the rolled delivery mechanism is unrolled. In another embodiment, a rolled delivery mechanism is provided for delivering therapeutic agent to a target tissue.

Abstract: A medical device comprising halofuginone is provided. The medical device is adapted for implantation or insertion into a blood vessel and it provides a cumulative, in vivo, 14 day release that lies between 0.02 ?g and 0.2 ?g of halofuginone per mm2 of stent surface area.