Abstract: A computer implemented method for generating a response to a received electronic message. A generative AI is prompted to analyse the electronic message and to generate intent data indicative of an intent associated with the electronic message and data-identifier data indicative of one or more predetermined data identifiers present in the electronic message. The electronic message is analysed to identify and retrieve client data associated with one or more parties associated with the electronic message. The method includes Verifying the retrieved client data includes data corresponding to the generated data-identifier data. If so, query an ERP system with the verified data-identifier data to extract ERP data corresponding to the predetermined data identifiers of which the data-identifier data is indicative, select a response-message template using the intent data, populate the selected response-message template with the ERP data, and present the populated response-message template to a user for approval.

Abstract: A computer-implemented chatbot including a prompt generation module, a large language model (LLM) module, and an answer generation module. The prompt generation modules generates an initial prompt based on a prompt template combined with a received user query. The initial prompt includes information source data specifying sources of factual information, conversation history data, and failed response data. The initial prompt is input to the LLM module, which is configured to generate an output and communicate the output to the answer generation module. The answer generation modules determines if the output is a plan to answer a user query. If so, relevant data is retrieved from external database or more APIs specified in the information source data. A further prompt is generated for answering the user query and input to the LLM module. The answer generation module repeats its tasks until a suitable answer to the user query is output.

Abstract: A computer implemented method of detecting hallucination in a large language model (LLM) output. A message is received. A prompt is generated for an LLM including the message and an instruction to generate an output identifying predetermined content in the message. The prompt is passed through an LLM to generate the output. The output is processed in accordance with a hallucination detection process to identify if any predetermined content identified by the LLM in the output is potentially hallucinated.

Abstract: A system and computer implemented method for generating validated prompt-templates for generating prompts for instructing large language models (LLMs) to perform specific tasks. An initial prompt is generated instructing an LLM to produce a plurality of candidate prompt-templates. The initial prompt includes an input data type to be included with a prompt generated using a candidate prompt-template, and output data to be produced by an LLM that has processed a prompt generated using a candidate prompt-template. The initial prompt is passed through an LLM to generate candidate prompt-templates. Test prompts are generated, each constructed from a candidate prompt-templates using input data from a set of pre-labelled input data having items of input data and corresponding labels. Each test prompt is passed through a further LLM to generate an output. The output is assessed, and candidate-prompt-templates are selected for subsequent generation of prompts.

Abstract: A system and computer implemented method for detecting hallucination in output of a generative AI system. User input is received specifying a query or task relating to information contained in a data object. A first vector representation of the user input and a second vector representation of the data object are generated. The first and second vector representations are compared to identify parts of the data object which match the query or task. An input is generated for a generative AI system with the user input and the parts of the data object. The input is input to a generative AI system. An output produced by the generative AI system is analysed to determine if the output contains information also present in the data object. If not, an error process is initiated. If so, the output is produced by the generative AI system.

Abstract: The present disclosure, in some embodiments, relates to a metal-insulator-metal (MIM) capacitor structure. The MIM capacitor structure includes one or more lower interconnects disposed within a lower dielectric structure over a substrate. A first dielectric layer is over the lower dielectric structure and includes sidewalls defining a plurality of openings extending through the first dielectric layer. A lower electrode is arranged along the sidewalls and over an upper surface of the first dielectric layer, a capacitor dielectric is arranged along sidewalls and an upper surface of the lower electrode, and an upper electrode is arranged along sidewalls and an upper surface of the capacitor dielectric. A spacer is along opposing outermost sidewalls of the upper electrode. The spacer has an outermost surface extending from a lowermost surface of the spacer to a top of the spacer. The outermost surface is substantially aligned with an outermost sidewall of the lower electrode.

Abstract: The present disclosure, in some embodiments, relates to a capacitor structure. The capacitor structure includes one or more lower interconnects disposed within a lower dielectric structure over a substrate. A lower electrode is arranged along sidewalls and an upper surface of the lower dielectric structure, a capacitor dielectric is arranged along sidewalls and an upper surface of the lower electrode, and an upper electrode is arranged along sidewalls and an upper surface of the capacitor dielectric. A spacer is arranged along outermost sidewalls of the upper electrode. The spacer includes a first upper surface arranged along a first side of the upper electrode and a second upper surface arranged along an opposing second side of the upper electrode. The first upper surface has a different width than the second upper surface.

Abstract: A semiconductor image sensing structure includes a semiconductor substrate having a front side and a back side, a pixel sensor disposed in the semiconductor substrate, a transistor disposed over the front side of the semiconductor substrate, and a reflective structure disposed over the front side of the semiconductor substrate. A gate structure of the transistor and the reflective structure include a same material. A top surface of the gate structure of the transistor and a top surface of the reflective structure are aligned with each other.

Abstract: Disclosed are methods, systems, and non-transitory computer-readable medium for analysis of images including wearable items. For example, a method may include obtaining a first set of images, each of the first set of images depicting a product; obtaining a first set of labels associated with the first set of images; training an image segmentation neural network based on the first set of images and the first set of labels; obtaining a second set of images, each of the second set of images depicting a known product; obtaining a second set of labels associated with the second set of images; training an image classification neural network based on the second set of images and the second set of labels; receiving a query image depicting a product that is not yet identified; and performing image segmentation of the query image and identifying the product in the image by performing image analysis.

Abstract: The present disclosure, in some embodiments, relates to a metal-insulator-metal (MIM) capacitor structure. The MIM capacitor structure includes one or more lower interconnects disposed within a lower dielectric structure over a substrate. A first dielectric layer is over the lower dielectric structure and includes sidewalls defining a plurality of openings extending through the first dielectric layer. A lower electrode is arranged along the sidewalls and over an upper surface of the first dielectric layer, a capacitor dielectric is arranged along sidewalls and an upper surface of the lower electrode, and an upper electrode is arranged along sidewalls and an upper surface of the capacitor dielectric. A spacer is along opposing outermost sidewalls of the upper electrode. The spacer has an outermost surface extending from a lowermost surface of the spacer to a top of the spacer. The outermost surface is substantially aligned with an outermost sidewall of the lower electrode.

Abstract: The present disclosure, in some embodiments, relates to a method of forming a capacitor structure. The method includes forming a capacitor dielectric layer over a lower electrode layer, and forming an upper electrode layer over the capacitor dielectric layer. The upper electrode layer is etched to define an upper electrode and to expose a part of the capacitor dielectric layer. A spacer structure is formed over horizontally extending surfaces of the upper electrode layer and the capacitor dielectric layer and also along sidewalls of the upper electrode. The spacer structure is etched to remove the spacer structure from over the horizontally extending surfaces of the upper electrode layer and the capacitor dielectric layer and to define a spacer. The capacitor dielectric layer and the lower electrode layer are etched according to the spacer to define a capacitor dielectric and a lower electrode.

Abstract: A semiconductor device includes an image sensor structure and a periphery device structure. The image sensor structure includes a first semiconductor substrate, a first interconnect structure, a radiation device, a transfer gate transistor electrically coupled to the radiation device, a floating diffusion region electrically coupled to the transfer gate, and a first capacitor disposed in the first interconnect structure. The transfer gate transistor electrically interconnects and disconnects the radiation device and the floating diffusion region. The periphery device structure includes a second interconnect structure disposed on the first interconnect structure, a second semiconductor substrate disposed on the second interconnect structure, a plurality of logic devices disposed in the second semiconductor substrate, and a second capacitor disposed in the second interconnect structure. The first capacitor and the second capacitor are electrically coupled to the floating diffusion region.

Abstract: The present disclosure, in some embodiments, relates to a method of forming a capacitor structure. The method includes forming a capacitor dielectric layer over a lower electrode layer, and forming an upper electrode layer over the capacitor dielectric layer. The upper electrode layer is etched to define an upper electrode and to expose a part of the capacitor dielectric layer. A spacer structure is formed over horizontally extending surfaces of the upper electrode layer and the capacitor dielectric layer and also along sidewalls of the upper electrode. The spacer structure is etched to remove the spacer structure from over the horizontally extending surfaces of the upper electrode layer and the capacitor dielectric layer and to define a spacer. The capacitor dielectric layer and the lower electrode layer are etched according to the spacer to define a capacitor dielectric and a lower electrode.

Abstract: A fitness equipment control system is provided. The system includes a receiver and a mobile apparatus, and performs the operations: generating a first control signal based on a training course of a user; transmitting the first control signal to the receiver to make the receiver generates a signal which is readable by the fitness equipment according to the first control signal and transmit the readable signal to the fitness equipment; receiving at least one physiological signal from at least one of the wearable devices worn by the user; based on the training course and the at least one physiological signal, generating a second control signal; and when the second control signal is generated, transmitting the second control signal to the receiver, wherein the receiver generates a signal which is readable by the fitness equipment according to the second control signal and transmits it to the fitness equipment.

Abstract: Disclosed are methods, systems, and non-transitory computer-readable medium for analysis of images including wearable items. For example, a method may include obtaining a first set of images, each of the first set of images depicting a product; obtaining a first set of labels associated with the first set of images; training an image segmentation neural network based on the first set of images and the first set of labels; obtaining a second set of images, each of the second set of images depicting a known product; obtaining a second set of labels associated with the second set of images; training an image classification neural network based on the second set of images and the second set of labels; receiving a query image depicting a product that is not yet identified; and performing image segmentation of the query image and identifying the product in the image by performing image analysis.

Abstract: A model construction system, apparatus, and method are provided. The model construction system includes at least one first source apparatus, at least one second source apparatus, and a model construction apparatus. The model construction apparatus receives a de-identification data set from each first source apparatus, receives a parameter set of a source model from each second source apparatus, generates at least one aligned data set by aligning the de-identification data set according to a predetermined data format, trains an original model to an assisted training model with the aligned data set(s), generates at least one updated parameter set according to the parameter set(s) and an assisted training parameter set, updates the assisted training model with one of the updated parameter set(s), and transmits the updated parameter set(s) to the second source apparatus(es). Each second source apparatus updates the source model according to the corresponding updated parameter set.

Abstract: An apparatus can include a calibration circuit to determine a backlight status of a display device based on a detected difference in a noise level corresponding to the display device. The calibration circuit can perform an operation to calibrate a touchscreen coupleable to the display device based on the determined backlight status.

Abstract: Disclosed are methods, systems, and non-transitory computer-readable medium for analysis of images including wearable items. For example, a method may include obtaining a first set of images, each of the first set of images depicting a product; obtaining a first set of labels associated with the first set of images; training an image segmentation neural network based on the first set of images and the first set of labels; obtaining a second set of images, each of the second set of images depicting a known product; obtaining a second set of labels associated with the second set of images; training an image classification neural network based on the second set of images and the second set of labels; receiving a query image depicting a product that is not yet identified; and performing image segmentation of the query image and identifying the product in the image by performing image analysis.

Abstract: A computer-implemented method for providing a personalized user interface to a user may include obtaining customer data; obtaining customer article data; obtaining customer interface activity data of the at least one customer; training a prediction model; obtaining at least one of user data, user article data, or user interface activity data of a user of the apparel subscription application; determining a rank of one or more articles based on the prediction model; obtaining environmental data including values of one or more environmental factors associated with user article data; and providing, to the user, the personalized user interface associated with the apparel subscription application to the user based on the rank of the one or more articles and the environmental data.

Abstract: An example display device with a multiple privacy mode of operation includes a micro lens cell having a first liquid crystal layer with a first set of liquid crystal molecules and a controllable optical effective refractive index, to transmit light through the micro lens cell. A pair of electrodes receive a voltage to change an orientation of the first set of liquid crystal molecules and the optical effective refractive index of the first liquid crystal layer. A micro lens array directs the light at an interface of the micro lens array and first liquid crystal layer. The micro lens array directs the light at different angles based on the voltage. A second liquid crystal layer changes a polarization of the light. A color filter controls a color of the light. A common electrode controls an orientation of a second set of liquid crystal molecules in the second liquid crystal layer.