Abstract: Due to the high language use variability in real-life, manual construction of semantic resources to cover all synonyms is prohibitively expensive and may result in limited coverage. Described herein are systems and methods that automate the process of synonymy resource development, including both formal entities and noisy descriptions from end-users. Embodiments of a multi-task model with hierarchical task relationship are presented that learn more representative entity/term embeddings and apply them to synonym prediction. In model embodiments, a skip-gram word embedding model is extended by introducing an auxiliary task “neighboring word/term semantic type prediction” and hierarchically organize them based on the task complexity. In one or more embodiments, existing term-term synonymous knowledge is integrated into the word embedding learning framework.

Abstract: The present disclosure provides a radio frequency filter, including: a substrate; a supporting electrode protruded on a front surface of the substrate; and a thin film structure formed on the substrate and spaced with the substrate by the supporting electrode. An end surface of a top end of the supporting electrode is in sealing contact with a front surface of the thin film structure.

Abstract: Methods, systems, and devices for wireless communications are described. A user equipment (UE) may communicate with a base station via a first instance of a periodic data burst, where the periodic data burst may include one or more uplink or downlink transmissions scheduled on a first bandwidth part (BWP). The UE may start a bandwidth switching timer at the beginning of the first instance of the periodic data burst, and determine whether the bandwidth switching timer has expired upon expiration of an inactivity timer that was triggered by receipt of one or more downlink control channel messages. The UE may operate, based on the determination, on one of the first BWP or a second BWP during at least a portion of a time period that extends from the expiration of the inactivity timer and a beginning of a second instance of the periodic data burst.

Abstract: The present disclosure relates to methods and apparatus for display processing. For example, disclosed techniques facilitate region of interest tracking for in-motion frames. Aspects of the present disclosure can identify a layer of interest of a plurality of layers associated with a frame. Aspects of the present disclosure can also determine coordinates for a region of interest based on characteristics associated with the identified layer of interest. Further, aspects of the present disclosure can perform display processing on the region of interest of the frame based on the coordinates for the region of interest.

Abstract: A method of wireless communication, by a user equipment (UE), includes receiving, from a base station, a medium access control-control element (MAC-CE) including an offset change command. The method also includes adjusting an offset within a periodic pattern based on the offset change command. A method of wireless communication, by a base station, includes detecting a misalignment between an arrival time of data relative to a periodic pattern. The method also includes transmitting, to a user equipment (UE), a medium access control-control element (MAC-CE) including an offset change command to adjust an offset within the periodic pattern.

Abstract: Efficient inner product search is important for many data ranking services, such as recommendation and Information Retrieval. Efficient retrieval via inner product dramatically influences the performance of such data searching and retrieval systems. To resolve deficiencies of prior approaches, embodiments of a new index graph construction approach, referred to generally as Norm Adjusted Proximity Graph (NAPG), for approximate Maximum Inner Product Search (MIPS) are presented. With adjusting factors estimated on sampled data, NAPG embodiments select more meaningful data points to connect with when constructing a graph-based index for inner product search. Extensive experiments verify that the improved graph-based index pushes the state-of-the-art of inner product search forward greatly, in the trade-off between search efficiency and effectiveness.

Abstract: An elastic contact element of the electrical connector is provided and has an upper ring, a lower ring, and a plurality of elastic flat bodies connected to and located between the upper ring and the lower ring. Each one of the elastic flat bodies is strip-shaped and forms a contact portion. The elastic flat bodies are spaced apart from each other and divided into several groups according to different height positions of the contact portions. Each one of the contact portions protrudes toward a center axis. Because the elastic flat bodies are divided into several groups according to different height positions of the contact portions, the elastic flat bodies contact an inserted male probe at various positions.

Abstract: Described herein are embodiments for systems and methods for mutual machine learning with global topic discovery and local word embedding. Both topic modeling and word embedding map documents onto a low-dimensional space, with the former clustering words into a global topic space and the latter mapping word into a local continuous embedding space. Embodiments of Topic Modeling and Sparse Autoencoder (TMSA) framework unify these two complementary patterns by constructing a mutual learning mechanism between word co-occurrence based topic modeling and autoencoder. In embodiments, word topics generated with topic modeling are passed into auto-encoder to impose topic sparsity for the autoencoder to learn topic-relevant word representations. In return, word embedding learned by autoencoder is sent back to topic modeling to improve the quality of topic generations. Performance evaluation on various datasets demonstrates the effectiveness of the disclosed TMSA framework in discovering topics and embedding words.

Abstract: The present invention relates to the technical field of natural medicines, and in particular to a triterpenoid compound with an effect of improving glucose and lipid metabolism disorders, and preparation and application thereof. A novel triterpenoid compound was obtained by separation from Momordica charantia L. for the first time in the present invention. Studies on a mouse primary hepatocyte model and a 3T3-L1 preadipocyte model show that the novel triterpenoid compound in the present invention has activity of regulating glucose and lipid metabolism, and after combination with known triterpenoid compound Momordicoside G, the activity of regulating glucose and lipid metabolism is significantly higher than that of a single compound at the same dose, indicating that the novel triterpenoid compound and a complex thereof in the present invention can be used to improve glucose and lipid metabolism disorders and provide a new way for diabetes treatment.

Abstract: Wrapper for cigarette manufacture includes transversely extending band regions applied by a printing technique, such as gravure printing. The band regions comprise starch, an anti-wrinkling agent such as 1,2 propylene glycol or glycerin, and optionally calcium carbonate. Any suitable printing technique can be used to apply the aqueous solution to the banded regions. The pattern of banded regions may be bands, stripes, two-dimensional arrays, undulated regions, and the like along and/or around the tobacco rod. The pattern can be applied in one or more layers. The pattern may be configured so that when a smoking article is placed on a substrate, at least two longitudinal locations along the length of the tobacco rod have film-forming compound located only on sides of the smoking article not in contact with the substrate.

Abstract: A battery bending test system includes a fastening device having an inclined surface configured to hold a battery; a pressing device configured to press the battery to bend; and a load-bearing device configured to bear the fastening device, the load-bearing device having a reference surface for bearing the fastening device. A first angle Ø is formed between the inclined surface and the reference surface. The range of the first angle Ø is 0°<Ø<90°. The battery bending test system and method can precisely control the bending angle and bending morphology of the battery, and reasonably evaluate the safety performance of the battery when subjected to the bending force, thus meeting expectations of the safety test.

Abstract: Methods, systems, and devices for wireless communications are described. A base station may transmit, and a user equipment (UE) may receive a group physical control channel signaling, which may configure grant uplink transmissions for a set of UEs including the receiving UE. The group physical control channel signaling may indicate allotted resources for configured grant uplink transmissions. The group physical control channel signaling may further indicate a group physical control channel which a UE may monitor for control messages. The UE may transmit a configured grant uplink transmission in accordance with the group physical control channel signaling. The base station may receive the configured grant uplink transmission and may determine a control message which it may transmit in the group physical control channel based on the configured grant uplink transmission. The UE may receive the control message, and may determine whether to retransmit a configured grant uplink transmission based thereon.

Abstract: A three-dimensional (3D) memory device includes a 3D NAND memory array, an on-die static random-access memory (SRAM), and peripheral circuits formed on the same chip with the on-die SRAM. The peripheral circuits include a page buffer coupled to the on-die SRAM and a controller coupled to the on-die SRAM and the page buffer. The controller may be configured to load program data into the page buffer and cache the program data into the on-die SRAM as a backup copy of the program data. In response to a status of programming the program data from the page buffer into the 3D NAND memory array being failed, the controller may be further configured to transmit the backup copy of the program data in the on-die SRAM to the page buffer, and program the backup copy of the program data in the page buffer into the 3D NAND memory array.

Abstract: An acoustic wave device including: a POI structure including: a material layer where a high acoustic velocity layer and a low acoustic velocity layer are alternate, a substrate is a lowermost high acoustic velocity layer; a first piezoelectric layer located above the material layer, wherein a layer adjacent to the first piezoelectric layer is referred to as a surface low acoustic velocity layer; wherein an acoustic velocity of a bulk wave propagated in the high acoustic velocity layer and the low high acoustic velocity layer is higher than and lower than an acoustic velocity of a bulk wave of the first piezoelectric layer, respectively. The POI structure includes at least two regions, a first device having a resonance of a first vibration mode is manufactured in the first region, and a second device having a resonance of a second vibration mode is manufactured in a second region.

Abstract: Different from prior works that model the internal distribution of patches within an image implicitly with a top-down latent variable model (e.g., generator), embodiments explicitly represent the statistical distribution within a single image by using an energy-based generative framework, where a pyramid of energy functions, each parameterized by a bottom-up deep neural network, are used to capture the distributions of patches at different resolutions. Also, embodiments of a coarse-to-fine sequential training and sampling strategy are presented to train the model efficiently. Besides learning to generate random samples from white noise, embodiments can learn in parallel with a self-supervised task (e.g., recover an input image from its corrupted version), which can further improve the descriptive power of the learned model. Embodiments does not require an auxiliary model (e.g.

Abstract: Learning latent variable models with deep top-down architectures typically requires inferring latent variables for each training example based on posterior distribution of these latent variables. The inference step relies on either time-consuming long-run Markov chain Monte Carlo (MCMC) sampling or a separate inference model for variational learning. Embodiments of a short-run MCMC, such as a short-run Langevin dynamics, are used herein as an approximate flow-based inference engine. Bias existing in the output distribution of non-convergent short-run Langevin dynamics may be corrected by optimal transport (OT), which aims at transforming the biased distribution produced by finite-step MCMC to the prior distribution with a minimum transport cost.

Abstract: A method of producing an aerosol-generating element for an aerosol-generating article or system is provided, the method including the steps of: preparing a matrix polymer solution including a matrix-forming polymer in water; adding a plurality of aerosol-generating formulation components to the matrix polymer solution to form an aerosol-generating solution, in which the aerosol-generating formulation components include a polyhydric alcohol and at least one alkaloid or cannabinoid, and in which the aerosol-generating solution includes at least 0.5 percent by weight of the at least one alkaloid or cannabinoid; forming a discrete portion of the aerosol-generating solution; adding the discrete portion of the aerosol-generating solution to a cross-linking solution of multivalent cations to cross-link the matrix-forming polymer; and removing the aerosol-generating element from the cross-linking solution and drying the aerosol-generating element.

Abstract: Embodiments of the present disclosure provide a method for generating an information assessment model, a method for determining the usefulness of comment information, apparatus, electronic device, and computer-readable medium.

Abstract: Current pretrained vision-language models for cross-modal retrieval tasks in English depend upon on the availability of many annotated image-caption datasets for pretraining to have English text. However, the texts are not necessarily in English. Although machine translation (MT) tools may be used to translate text to English, the performance largely relies on MT's quality and may suffer from high latency problems in real-world applications. Embodiments herein address these problems by learning cross-lingual cross-modal representations for matching images and their relevant captions in multiple languages. Embodiments seamlessly combine cross-lingual pretraining objectives and cross-modal pretraining objectives in a unified framework to learn image and text in a joint embedding space from available English image-caption data, monolingual corpus, and parallel corpus. Embodiments are shown to achieve state-of-the-art performance in retrieval tasks on multimodal multilingual image caption datasets.

Abstract: A hydraulic log splitter, comprises a rack, a main beam, an oil tank, an oil cylinder, a cutter, and a push plate, wherein the main beam is arranged at an upper end of the rack, the oil tank is arranged inside the main beam, one end of the main beam is connected to the rack, the other end of the main beam protrudes from the rack, the main beam is recessed inward to form a reinforcing rib structure having a V-shaped longitudinal section, the oil cylinder is arranged at one end of the main beam, the cutter is arranged at the other end of the main beam, the push plate is arranged between the cutter and the oil cylinder, the push plate is connected to a push rod of the oil cylinder.