Abstract: Large language models (LLMs) are able to provide robust results based on specified formatting and organization. Traditionally, however, users must form detailed queries to obtain desired results in a desired format. Accordingly, although LLMs are designed to receive natural language input, users often lack the skill, knowledge, or patience to utilize LLMs to their full potential. Ambient information and user history associated with device screenshots are leveraged to provide proactive artificial-intelligence (AI) assistance and query resolution in an LLM environment. In particular, screenshots associated with a computer display are continuously captured and analyzed to detect activity triggers for plugins, for example. In response to detecting an activity trigger, local context associated with one or more prior screenshots is collected. The collected context is then used to inform the plugin for performing the task, thereby reducing the burden placed on the user to input the required information.

Abstract: Securing and optimizing communications for a cloud service provider includes collecting connection summary information at network interface devices associated with host computing devices for a group of resources allocated to a customer of the cloud computing environment. The connection summary information includes local address information, remote address information, and data information, each connection established via the network interface devices. At least one communication graph is generated for the group of resources using the connection summary information. The graph includes nodes that represent communication resources of the group of resources and edges extending between nodes that characterize communication between the nodes. At least one analytics process is performed on data from the graph to identify at least one of a micro-segmentation strategy, a communication pattern, and a flow prediction for the group of resources.

Abstract: The technology relates to systems and methods for dynamically generating prompts for a generative artificial intelligence (AI) model. An example method includes receiving input content for evaluation by a generative AI model; receiving an input-content embedding for the input content; receiving trait data and trait-data embeddings for the trait data; identifying similar trait data by comparing the input-content embedding with the trait-data embeddings, wherein the similar trait data is a subset of the trait data that is similar to the input content; generating a prompt including the input content and the identified similar trait data; providing the prompt to the generative AI model; and receiving, from the generative AI model in response to the prompt, an output payload including an evaluation of the input content.

Abstract: Techniques of memory tiering in computing devices are disclosed herein. One example technique includes retrieving, from a first tier in a first memory, data from a data portion and metadata from a metadata portion of the first tier upon receiving a request to read data corresponding to a system memory section. The method can then include analyzing the data location information to determine whether the first tier currently contains data corresponding to the system memory section in the received request. In response to determining that the first tier currently contains data corresponding to the system memory section in the received request, transmitting the retrieved data from the data portion of the first memory to the processor in response to the received request. Otherwise, the method can include identifying a memory location in the first or far memory that contains data corresponding to the system memory section and retrieving the data from the identified memory location.

Abstract: The disclosed concepts relate to leveraging a language model to identify data health issues in a data set. One example method involves accessing a data set. The example method also involves, using an automated evaluation planning agent, inputting a prompt to generate a data evaluation plan for the data set to a generative language model, the prompt including context describing the data set. The example method also involves receiving the data evaluation plan generated by the generative language model and identifying one or more data health issues in the data set by performing the data evaluation plan using an automated evaluation plan execution agent.

Abstract: Systems and methods are provided for implementing automatic prompt optimization using textual gradients. In various embodiments, a feedback prompt, input into a large language model (“LLM”), is used to generate textual gradients that criticize a current prompt. The feedback prompt includes the current prompt and predictions that are incorrect compared with corresponding labels associated with minibatch data processed by the LLM using the current prompt. The textual gradients and current prompt are used in an editing prompt to the LLM to obtain a set of optimized prompts, which may be expanded using a paraphrasing prompt that is input into the LLM to generate a set of paraphrased prompts. A selection algorithm is used to select one or more optimized prompts from the set of optimized prompts and/or the set of paraphrased prompts, and the process is repeated with the selected one or more optimized prompts replacing the current prompt.

Abstract: Examples of the present disclosure describe systems and methods for heterogeneous scheduling for processors with multiple core types. In some examples, a scheduler assigns thread policies to respective threads. The scheduler then allocates the threads to heterogeneous cores in accordance with the thread policies assigned to the respective threads. The heterogeneous cores include one or more power efficient cores, one or more intermediate cores, and one or more performance-oriented cores, among other core types. In some examples, a core parking engine determines how many cores should be unparked for one or more power efficient cores, one or more intermediate cores, and one or more performance-oriented cores, among other core types.

Abstract: The disclosed concepts relate to implementation of application and application engine functionality using machine learning. One example method involves obtaining a seed image representing a seeded application state and mapping the seed image to at least one seed image token using an image encoder. The example method also involves inputting the at least one seed image token as a prompt to a neural dreaming model that has been trained to predict training sequences obtained from one or more executions of one or more applications, the training sequences including images output by the one more applications during the one or more executions and inputs to the one or more applications during the one or more executions. The example method also involves generating subsequent image tokens with the neural dreaming model, and decoding the subsequent image tokens with an image decoder to obtain subsequent images.

Abstract: A high-power multiplexer/demultiplexer (“mux/demux”) and a three-dimensional (“3D”) printed phase mask are provided for hollow-core optical fiber applications. The high-power mux/demux includes hollow core optical fiber interfaces configured to couple with free-space optical fiber cables, a diffraction grating, a 3D printed phase mask, and a set of lenses. The diffraction grating is configured, based on different wavelengths, either to at least diffract each optical signal of a plurality of optical signals having different wavelengths into two or more optical signals or to at least diffract a single optical signal having multiple wavelengths into a plurality of optical signals. The phase mask includes reflective features configured to reflect optical signals at different optical path lengths to provide reflected optical signals with different phases. The set of lenses is configured to collimate optical signals onto or from the diffraction grating or to focus optical signals onto or from the phase mask.

Abstract: A method of providing data communication between a first device and a second device includes, establishing a first communication link with a downstream device connected to the second device using a first mode via a USB-type interface, wherein in the first mode the USB-type interface utilizes a first set of USB communication lanes; establishing a second communication link with the first device via the USB-C port using an Alternate mode wherein the Alt-mode utilizes the first set of USB communication lanes; and, in accordance with establishing the second communication link, changing a mode of the first communication link so that the first communication link does not communicate via the first set of USB communication lanes.

Abstract: In non-limiting examples of the present disclosure, systems, methods, and devices for matching device configurations to games are presented. A set of device configuration tiers may be generated from gameplay telemetry data generated by a plurality of client devices executing a plurality of games. A device configuration for a specific client device may be determined based at least on the specific client device's GUI type. When the specific client device accesses a software game library a determination may be made based on a performance tier corresponding to the device configuration for the specific client device as to whether the specific client device can adequately execute each game. One or more recommendations may be rendered and displayed in the game library based on the determination of whether the specific client device can adequately execute each game.

Abstract: Aspects relate to observing various activities, interactions, behaviors, and other factors associated with a data exchange and creating one or more markers based on significant details associated with the observance. The one or more markers are retained and selectively rendered as a function of one or more conditions that should be satisfied before the marker is presented to the user. Some markers can contain parameters that should be satisfied in order for the marker to be considered complete. If a parameter is not satisfied, subsequent markers can be created as a function of the rendered marker. The subsequent markers can be rendered when a condition associated with the subsequent marker is satisfied.

Abstract: Methods, systems, and computer programs are presented for eliminating bias while training an ML model using training data that includes past experimental data. One method includes accessing experiment results, for A/B testing of a first model, that comprise information regarding engagement with a first set of items presented to users, each item being presented within an ordered list of results. A position bias is calculated for positions within the ordered list of results where the items were presented. A machine-learning program is trained to obtain a second model using a training set comprising values for features that include the calculated position bias. The method includes detecting a second set of items to be ranked for presentation to a first user, and calculates, using the second model, a relevance score for the second set of items, which are ranked based on the respective relevance score and presented on a display.

Abstract: A method for automated device disassembly includes, at a computerized demanufacturing system, receiving a target electronic device for disassembly. Using a set of one or more sensors of the computerized demanufacturing system, a set of sensor data is collected that quantifies one or more physical properties of the target electronic device. Based at least in part on the set of sensor data, correspondences are identified between one or more effectors of the computerized demanufacturing system and one or more interactable elements of the target electronic device. The interactable elements are automatically manipulated using the one or more effectors to at least partially disassemble the target electronic device.

Abstract: Methods, systems, and computer program products for direct assignment of physical devices to confidential virtual machines (VMs). At a first guest privilege context of a guest partition, a direct assignment of a physical device associated with a host computer system to the guest partition is identified. The guest partition includes the first guest privilege context and a second guest privilege context, which is restricted from accessing memory associated with the first guest privilege context. The guest partition corresponds to a confidential VM, such that a memory region associated with the guest partition is inaccessible to a host operating system. It is determined, based on a policy, that the physical device is allowed to be directly assigned to the guest partition. Communication between the physical device and the second guest privilege context is permitted, such as by exposing the physical device on a virtual bus and/or forwarding an interrupt.

Abstract: A solution is disclosed for computer hardware replacement using performance analytics that selects replacement computer hardware based on actual user needs and enterprise priorities. Key performance data is collected and compared with various baselines, thereby identifying hardware that is performing below acceptable levels. Enterprise data and collected data are received from an instrumented operating system on a computing device. The collected data includes boot performance, application performance, and hardware performance. Based at least on the collected data, a usability score is determined by performing a weighted calculation on the collected data. Based at least on the usability score and the enterprise data, it is determined whether a score improvement is required. Based at least on the enterprise data, a score improvement selection is determined. The score improvement selection is reported based at least on determining that a score improvement is required.

Abstract: In implementations of the subject matter as described herein, there is provided a method for forgery detection of a face image. Subsequent to inputting a face image, it is detected whether a blending boundary due to the blend of different images exists in the face image, and then a corresponding grayscale image is generated based on a result of the detection, where the generated grayscale image can reveal whether the input face image is formed by blending different images. If a visible boundary corresponding to the blending boundary exists in the generated grayscale image, it indicates that the face image is a forged image; on the contrary, if the visible boundary does not exist in the generated grayscale image, it indicates that the face image is a real image.

Abstract: A reference to a digital item is stored as a digital card. The digital card can also be contained in, and/or refer to, other digital cards. The digital card can also include properties or attributes that may be added from the digital item that is being referred to. The digital card can be tagged with a tag that may also be applied to other digital cards. The digital card can be stored in a data pod within a de-centralized data storage system architecture so the tag is surfaced along with the digital card in different contexts.

Abstract: A method enhances authentication requirements to documents of a document repository based, at least in part, on a security policy associated with a branch under which the documents are organized. The method implements an approval service that is identified in a branch policy. The approval service determines whether a user is authorized to modify documents included in the branch. The method further selectively requires multiple authentications from multiple authentication systems in order to access one or more particular branches in a document repository. Further, the multiple authentication systems are based on separate and independent sets of authentication credentials.

Abstract: The disclosed technology is generally directed to a distributed query-and-command system. In one example of the technology, in a trusted execution environment (TEE) of a first node, database code of the first node and distributed ledger code of the first node is executed, such that execution of the distributed ledger code of the first node instantiates a first instance of a distributed ledger of a consortium blockchain, and such that execution of the query-and-command code of the first node instantiates a first instance of a query-and-command system. The consortium blockchain is distributed among a plurality of nodes, and the query-and-command system is distributed among the plurality of nodes. A first transaction that is associated with modifying the query-and-command system is received. The first transaction is executed. Changes associated with the first transaction to the distributed ledger are persisted.