Abstract: A data-free post-training quantization method includes obtaining data distribution input by a user, where the data distribution is distribution to which an input activation value of each network layer of a floating-point model conforms; inputting random data into the floating-point model to obtain the input activation value of each network layer; performing statistical analysis on the input activation value of each network layer based on the data distribution, to obtain a data range of the input activation value of each network layer; determining a quantization parameter of the input activation value of each network layer based on endpoint values of the data range; and during inference by using the floating-point model, performing, by using the quantization parameter of the input activation value of each network layer, quantization processing on the input activation value generated during inference of each network layer.

Abstract: Provided are peptide biomarkers for diagnosis of allergy, monitoring development of clinical tolerance in an allergic individual, and predicting whether an allergic subject is likely to develop clinical or natural tolerance over time. The invention also relates to diagnostic methods and diagnostic kits employing the peptide biomarkers.

Abstract: A language-driven system that integrates the capabilities of specialized methods into a unified framework. ChatHuman functions as an assistant proficient in utilizing, analyzing, and interacting with tools specific to 3D human tasks, adeptly discussing and resolving related challenges. Built on a Large Language Model (LLM) framework, ChatHuman is trained to autonomously select, apply, and interpret a diverse set of tools in response to user inputs. Our approach overcomes significant hurdles in adapting LLMs to 3D human tasks, including the need for domain-specific knowledge and the ability to interpret complex 3D outputs. The innovations of ChatHuman include leveraging academic publications to instruct the LLM on tool usage, employing a retrieval-augmented generation model to create in-context learning examples for managing new tools, and effectively discriminating between and integrating tool results by transforming specialized 3D outputs into comprehensible formats.

Abstract: A distance detection device, a distance detection method, and a processing device are provided. The processing device includes a processing head configured to process a workpiece and the distance detection device detechably connected to the processing head. The distance detection device is configured to detect a height of a processing surface of the workpiece. The processing head is configured to process the processing surface of the workpiece according to the height of the processing surface of the workpiece. When external forces in the same direction are applied to a probe shaft of the distance detection device, the probe shaft is extendable and retractable. In this way, no specific reset position is needed, and the probe shaft is extendable and retractable at different positions. The probe shaft is allowed to directly reset along a Z-axis direction without resetting in an X-axis and a Y-axis.

Abstract: A variable transmission system for a hybrid electric vehicle includes an input shaft connected to an engine and sleeved with first and second gears, and a first synchronizer connected to the input shaft. Third and fourth gears sleeve an output shaft, and a second synchronizer is connected to the output shaft. A shaft of a first motor is connected to a fifth gear and a sixth gear, the fifth, first, and third gears drive each other, the sixth, second, and fourth gears drive each other, and a shaft of a second motor is connected to the output shaft via a transmission mechanism.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. For example, a user equipment (UE) may determine a probability of each beam in an initial set of candidate beams being a target beam (for example, a best beam given current channel conditions) according to a current source beam and one or more parameters that relate to a propagation path from a transmitter to the UE. Accordingly, the UE may select, among the beams in the initial set of candidate beams, a set of candidate beams to measure in a beam management procedure according to the respective probabilities of each candidate beam being the target beam, which may reduce power consumption at the UE, reduce a beam selection or beam switching latency, and/or improve performance associated with beamformed communications.

Abstract: Techniques and apparatus for beam selection in compliance with radio frequency (RF) exposure limits. A method that may be performed by a wireless device includes receiving, from another wireless device, a plurality of signals via a plurality of beams; measuring a received signal power for each of the beams, based on the received plurality of signals; determining a transmit power limit for each of the beams based at least in part on an RF exposure limit; determining an estimation of a signal quality at the other wireless device for each of the beams based at least in part the received signal powers and the transmit power limits; selecting one of the beams based on the estimations of the signal qualities for the beams; and transmitting a signal via the selected beam.

Abstract: A multi-position variable speed transmission system for a hybrid electric vehicle includes an engine, a first motor, a second motor, an input shaft, an output shaft, a first intermediate shaft, a second intermediate shaft, and first, second and third synchronizers. The input shaft is connected to the engine. The intermediate shafts are respectively connected to the first and second motors. The input shaft is connected to the first synchronizer. The output shaft is connected to the second and third synchronizers. The first intermediate shaft is connected with a first transmission gear. The second intermediate shaft is connected with second transmission gears. The first or second transmission gear respectively meshes with one of gears on the output shaft.

Abstract: The disclosure relates to compositions and methods for the preparation, use, and/or formulation of adeno-associated virus capsid protein variants.

Abstract: An automated dry-heating bedbug eradication device for pest eradication is provided herewith. By the use of a specifically fabricated graphene-coated copper heating film, the dry-heating device of the present invention can readily achieve a temperature that is higher than the well-established temperature of 45° C., which is proven lethal to all life stages of bedbugs, and also effective in ant eradication. The device is carefully designed to achieve high portability and maneuverability, and is customizable to accommodate the needs to sanitize mattresses of different sizes in different settings, ranging from home use to commercial hotel or hostel uses. The device of the present invention is also equipped with wireless connectability to mobile applications, and infrared sensor and PID systems for remote real-time monitoring and control.

Abstract: Provided are peptide biomarkers for diagnosis of allergy, monitoring development of clinical tolerance in an allergic individual, and predicting whether an allergic subject is likely to develop clinical or natural tolerance over time. The invention also relates to diagnostic methods and diagnostic kits employing the peptide biomarkers.

Abstract: A pharmaceutical dosage form comprises a capsule shell and a fill composition, wherein the fill composition comprises an emulsion of (a) simethicone, (b) loperamide or a pharmaceutically acceptable salt thereof, and (c) at least two surfactants, wherein the emulsion has a hydrophilic lipophilic balance ranging from about 8 to about 12. Also described are methods of making the pharmaceutical dosage forms and methods for treating humans suffering from gastrointestinal distress.

Abstract: A method and a protective apparatus for protecting a laser device are provided. By connecting the laser device to a protective housing to safeguard laser engraving operations, safety during a laser engraving process is enhanced. Furthermore, a support plate of the protective apparatus is capable of vertically moving, thereby enabling the protective housing to adapt to objects of varying heights and different types of the laser device to consistently achieve laser engraving within an optimal engraving focal range, an applicability scope of the protective apparatus is expanded, and user experience is further improved. Additionally, two guide structures are respectively disposed a left side of the support plate and a right side of the support plate to drive the support plate to vertically move, thereby ensuring stable vertical movement during the laser engraving process, yielding superior engraving results with enhanced precision.

Abstract: The disclosure relates to compositions and methods for the preparation, use, and/or formulation of adeno-associated virus capsid protein variants.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. For example, a user equipment (UE) may determine a probability of each beam in an initial set of candidate beams being a target beam (for example, a best beam given current channel conditions) according to a current source beam and one or more parameters that relate to a propagation path from a transmitter to the UE. Accordingly, the UE may select, among the beams in the initial set of candidate beams, a set of candidate beams to measure in a beam management procedure according to the respective probabilities of each candidate beam being the target beam, which may reduce power consumption at the UE, reduce a beam selection or beam switching latency, and/or improve performance associated with beamformed communications.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may identify an available energy for uplink transmission of a plurality of communication links of the UE. The UE may configure, from the available energy, a first set of energy allocations for the plurality of communication links. The UE may configure a second set of energy allocations for one or more communication links of the plurality of communication links, wherein the second set of energy allocations are allocated from a remainder of the available energy after the first set of energy allocations are allocated. The UE may transmit based at least in part on at least one of the first set of energy allocations or the second set of energy allocations. Numerous other aspects are described.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may allocate a first amount of power to one or more radios for a time window based on radio frequency (RF) exposure information and one or more other criteria. The UE may allocate a second amount of power to a selected channel or communication utilized by at least one radio of the one or more radios for one or more time frames within the time window based on the first amount of power allocated to the at least one radio for the time window. The UE may transmit the selected channel or communication based on the first amount of power or the second amount of power. Numerous other aspects are described.

Abstract: The disclosure relates to compositions and methods for the preparation, use, and/or formulation of adeno-associated virus capsid protein variants.