Abstract: An interactive area includes a first display disposed along a path for a queue for an attraction. The interactive area also includes a recognition and tracking system configured to recognize or to identify a guest upon entering the queue and to track the guest as they move along the path. The interactive area further includes a controller coupled to the recognition and tracking system. The controller is configured to cause first display to exhibit a user interface that moves with the guest along the first display as the guest moves along the path.

Abstract: A structural member for a battery cell may be provided in the form of an I-beam having a central support structure that extends through the middle of the battery cell. One or more electrode stacks may be provided within one or more cavities defined by the structural member. The structural member may provide mechanical, structural, and/or thermal benefits for the battery cell and/or one or more battery sub-assemblies, battery modules, and/or battery packs. A battery cell includes a contoured housing with curved lateral surfaces conforming to a shape of one or more wound electrode layers. When the wound electrode layers are inserted into the contoured housing, space between the electrode layers and the contoured housing is reduced, as compared to that of square or rectangular housings, leading to higher wetting rate and lower wetting times of the wound electrode layers, higher structural loading, wider tolerances, and/or flexible cooling strategies.

Abstract: In an industrial control system (ICS), latent vectors are generated to represent identity or behaviors of host devices coupled to the ICS. A computing system captures communications transmitted by a host device across a network associated with the ICS. A set of values are extracted from one or more respective fields in the communication, then applied to a trained neural network. Values of a first set of fields are applied at an input layer of the trained neural network, while values of a second set of fields are applied at an output layer of the neural network. Based on the application of the neural network to the values extracted from the communication, the computing system generates a latent vector.

Abstract: Aspects of the disclosure relate to a battery cell that includes a rolled electrode layer having a coated region, and multiple uncoated extensions that are radially aligned with each other, such as along a direction extending radially from a center of the rolled electrode layer.

Abstract: A learning environment presents, at a first skill level, first content of a first learning requirement to a learner via a user interface on a client device. The learning environment detects behavior of the learner selecting assistive content during interaction with the first content and determines a proficiency of the learner at a decision point of the first learning requirement. The learning environment determines a path directive based on the behavior, the proficiency, and a learner history, where the path directive defines a direction of a next step in the learning path for the learner. The learning environment determines a next content and a next skill level based on the path directive and presents the next content at the next skill level to the learner via the user interface.

Abstract: An example method includes providing, to a sequence model (i) a plurality of few-shot prompts, wherein each prompt comprises a demonstration passage, a demonstration task, and a demonstration query, wherein the demonstration task describes a type of retrieval, and wherein the demonstration query is relevant to the demonstration task, and (ii) a plurality of passages sampled from a corpus of passages. The method also includes receiving, from the sequence model and for the plurality of passages and based on the plurality of few-shot prompts, a respective plurality of predicted task-query pairs, the sequence model having been prompted to predict a task based on an input passage, and predict an output query relevant to the predicted task. The method further includes generating a synthetic training dataset comprising the plurality of passages and the respective plurality of predicted task-query pairs. The method also includes providing the synthetic training dataset.

Abstract: An interstitial recessed cantilever latch mechanism for fiber optic and electrical connectors is disclosed. This mechanism enables the design of more spatially efficient multi-position connectors while retaining the reliability and tactile feedback of traditional connector latch mechanisms. The mechanism also allows the receptacle interface to be substantially simplified, and reduces the number of unique components required by means of a hermaphroditically assembled backshell half.

Abstract: A method of operating a cleaning and transferring system in a venue with a first location and a second location includes receiving one or more accessories onto a conveyor at the second location. The method also includes transferring the one or more accessories from the second location to the first location via the conveyor. The method further includes cleaning the one or more accessories during the transferring of the one or more accessories from the second location to the first location via the conveyor. The method further includes receiving one or more guest items onto the conveyor at the first location. The method further includes transferring the one or more guest items from the first location to the second via the conveyor.

Abstract: An attraction system has a plurality of show effects, a handheld or wearable device, and a control system. The control system is configured to perform operations that include monitoring a parameter of the handheld or wearable device, determining that the parameter is indicative of an undesirable utilization of the attraction system, and controlling the handheld or wearable device, at least one show effect of the plurality of show effects, or both. The control system is also configured to provide guidance to change the parameter in response to the determination that the parameter is indicative of the undesirable utilization of the attraction system.

Abstract: The technology employs soft knowledge prompts (KPs) to inject relevant world knowledge into language models. This includes training KPs via self-supervised learning on data from one or more knowledge bases. KPs are task independent and can function as an external memory of the language models. KPs may be entity-centric, meaning that each prompt primarily encodes information about one entity from a given knowledge base. A method includes identifying a KP in response to a received input text, concatenating that KP to a sequence of word embeddings of the input text, applying the concatenated information to a trained language model, predicting an object entity name, computing a cross-entropy loss, and updating the identified KP based on the computed cross-entropy loss.

Abstract: An inspection vehicle for an attraction system of an amusement park includes one or more bogies configured to engage with a path of the attraction system, a sensor configured to capture imagery of the path, and a controller communicatively coupled to the sensor. The controller is configured to receive data indicative of a detected marker, determine a path section at which the inspection vehicle is located based on the detected marker, determine target imagery associated with the path section, receive captured imagery from the sensor, and output a control signal in response to determining a difference between captured imagery received from the sensor and the target imagery is equal to or greater than a threshold value.

Abstract: A rotary manifold for a rotor assembly of a cohesion-type drive includes a manifold body extending along a drive axis for rotation thereabout, a first ductwork internal the body for fluid communication with a plurality of first chambers of the drive, and a second ductwork internal the body for fluid communication with a plurality of second chambers of the drive. The second ductwork is in fluid isolation of the first ductwork.

Abstract: The present disclosure provides an improved self-watering planter which makes use of osmotic forces by including a vertical support which is at least partially perforated or porous, and which doubles up as a reservoir for a set of tiered planter platforms. The soil held by the planter platforms comes into contact with the porous or perforated outer wall of the support, the dampness of the soil thus determining the rate at which water is able to leak out from the support to the platforms. The planter is constructed such that the reservoir of the vertical support extends above the height of the highest platform, so that gravitational and osmotic forces are all that is required to power and regulate the supply of water to housed plants.

Abstract: An aerial drone system includes a first drone comprising a first drone body and a first propeller system configured to propel the first drone through air, a second drone comprising a second drone body and a second propeller system configured to propel the second drone through the air, and an electronic media tether coupled to the first drone and the second drone, wherein the electronic media tether includes a plurality of light emitting diodes (LEDs) and is configured to transmit power to the plurality of LEDs.

Abstract: Present systems and methods are directed to an entertainment system for an environment. The entertainment system includes a container configured to hold one or more user devices and a sensor configured to detect a presence of the one or more user devices in the container and to generate data indicative of the presence of the one or more user devices in the container. The entertainment system also includes one or more processors configured to identify the presence of the one or more user devices in the container based on the data, and in response to identifying the presence of the one or more user devices in the container based on the data, instruct the one or more user devices to play one or more audio tracks.

Abstract: In some examples, an electronic device includes a transparent display panel including a first side facing a first direction and a second side facing a second direction. In some examples, the electronic device includes a first switchable opacity panel disposed in the first direction from the transparent display panel. In some examples, the electronic device includes a second switchable opacity panel disposed in the second direction from the transparent display panel.

Abstract: Systems and methods for prompt tuning can utilize previously-learned prompts for the initialization of tuning for prompts on different tasks that may differ from the task associated with the previously-learned prompt. The prompt being utilized for initialization can be a generic prompt and/or may be a prompt selected based on a determined similarity between two or more task embeddings.

Abstract: A show effect system may include an applicator configured to apply a detectable marker, a camera configured to generate image data representative of an environment with the detectable marker, and a control system communicatively coupled to the camera and comprising a processor and a memory, wherein the memory stores instructions executable by the processor. The detectable marker may include an ultra-violet (UV) component with a detectable wavelength range corresponding to UV-A, UV-B or UV-C. The processor may execute the instructions that cause the processor to determine a motion of the detectable marker based on the image data and determine that the motion of the detectable marker corresponds to a target motion. The processor may also identify a show effect that may be associated with the target motion and trigger generation of the show effect in the environment.

Abstract: The present disclosure provides a novel sentence-level representation learning method Conditional Masked Language Modeling (CMLM) for training on large scale unlabeled corpora. CMLM outperforms the previous state-of-the-art English sentence embedding models, including those trained with (semi-)supervised signals. For multilingual representations learning, it is shown that co-training CMLM with bitext retrieval and cross-lingual natural language inference (NL) fine-tuning achieves state-of-the-art performance. It is also shown that multilingual representations have the same language bias and principal component removal (PCR) can eliminate the bias by separating language identity information from semantics.