Abstract: According to an embodiment of the present disclosure, a method for video moment retrieval performed by a computing device may include obtaining a pair of video and query to perform the video moment retrieval, determining based on meaning of the query whether a retrieval bias regarding the query has a positive effect on the video moment retrieval, and selectively removing the retrieval bias from a result of the video moment retrieval according to the determination result to generate a final retrieval result.

Abstract: Described apparatuses and methods facilitate sharing redundant memory portions at a controller-level to enable memory repair between two or more memory blocks. Each memory die of multiple memory dies can include, for instance, multiple spare rows for use if a row of a memory array has a faulty bit. If a memory die has more faults than spare rows, the memory die cannot repair the additional faults. This document describes a controller that can inventory unrepaired faults and available spare rows across multiple memory dies. The controller can then “borrow” a spare row from a second memory die that has an available one and “share” the spare row with a first memory die that has a fault than it cannot repair. The controller can remap a memory access request targeting the row with the unrepaired fault in the first memory die to a spare row in the second memory die.

Abstract: A display driving IC includes first channel block to Nth channel block each including M source amplifiers, N and M being an integer, source driving pads each connected to the M source amplifiers, a multiplexer configured to alternate data outputs of the source amplifiers or selectively provide a test path so that a probe test is performed on a plurality of source amplifiers for each of the first to Nth channel blocks through a test pad selected from the source driving pads, and a control unit configured to control driving of the source amplifiers and multiplexer.

Abstract: Disclosed is a semiconductor device comprising a substrate including an active pattern, a channel pattern on the active pattern and including semiconductor patterns spaced apart from and vertically stacked on each other, a source/drain pattern connected to the semiconductor patterns having a p-type, a gate electrode on the semiconductor patterns and including inner electrodes between neighboring semiconductor patterns and an outer electrode on an uppermost semiconductor pattern, and a gate dielectric layer between the gate electrode and the semiconductor patterns and including an inner gate dielectric layer adjacent to the inner electrode and an outer gate dielectric layer that extends from bottom to lateral surfaces of the outer electrode. The outer electrode and the outer gate dielectric layer have an inverted T shape.

Abstract: A neural network device for learning dependency of feature data includes: a memory in which at least one program is stored; and a processor that performs a calculation by executing the at least one program, in which the processor is configured to acquire graph information including a data node for a human body; extract feature data corresponding to a plurality of joints constituting the human body from the graph information; acquire a self-attention output corresponding to the feature data based on a self-attention mechanism; and generate result data for a motion of the human body based on the self-attention output, and the self-attention output includes position information acquired based on positional encoding of the feature data and structural information acquired based on geodesic encoding of the feature data.

Abstract: An apparatus for generating text from an image may comprise: a memory configured to store at least one instruction; and a processor configured to execute the at least one instruction, wherein the processor is further configured to generate encoding information for an image based on the image and extract text information related to content of the image based on a degree of association with the encoding information.

Abstract: The present disclosure relates to a polyester film and a preparation method of the same. Since the polyester film includes a resin layer formed from a polyester resin including a first diol moiety derived from isosorbide and a second diol moiety derived from cyclohexanedimethanol in a controlled ratio on at least one surface of a base layer, it is possible to exhibit excellent heat resistance and adhesion.

Abstract: Described herein is a nucleic acid molecule including an asymmetrically modified inverted terminal repeat (ITR). An AAV vector including the nucleic acid molecule has advantages of increased productivity and expression efficiency of a transgene, and decreased genotoxicity, by having an asymmetric ITR in which any one of two ITRs is modified. Also, described herein is are compositions and vectors.

Abstract: Provided is a method and system for verifying a virtual code based on a sphere. A method for verifying a virtual code based on a sphere includes: a virtual code reception operation of receiving, by a virtual code verifying means, a virtual code from a virtual code generating means, wherein the virtual code includes a fixed code and a plurality of detailed codes; a storage location search algorithm selection operation of determining, by the virtual code verifying means, a group, to which the virtual code generating means belongs, through the fixed code and selecting a storage location search algorithm for the group; a detailed code extraction operation of extracting, by the virtual code verifying means, the plurality of detailed codes included in the virtual code.

Abstract: A refrigerator includes a main body having a storage compartment formed therein, and a cooling module forming a machine compartment in which a compressor is disposed and detachably mounted to the main body. The cooling module includes a module body forming the machine compartment, a condenser of an L shape disposed on one side of the compressor and including a first part corresponding to a portion of a rear surface of the machine compartment and a second part corresponding to one side of the machine compartment. The condenser is vertically mounted on a base plate disposed below the module body. A module cover is provided to cover the rear of the machine compartment, and a suction flow path includes first and second suction flow paths.

Abstract: A method for assembling a secondary battery cell module by using an assembling jig including a plurality of guide rods disposed on a jig plate includes: mounting a lower frame on the jig plate while the guide rods are inserted into a plurality of arranging through-holes of the lower frame; disposing a plurality of battery cells on the lower frame; mounting an upper frame on the battery cells while the guide rods are inserted into a plurality of arranging through-holes of the upper frame; fastening the upper frame and the lower frame together; and separating the assembling jig from the upper frame and the lower frame.

Abstract: Disclosed is a transmitter which includes an encoder and a transmission interface circuit. The encoder receives data bits and generates conversion bits, a number of is the conversion bits being more than a number of the data bits, based on the number of the data bits. The encoder detects a risk pattern of the conversion bits to generate detection data and converts the risk pattern into a replacement pattern based on the detection data to generate code bits, a number of is the code bits being equal to the number of the conversion bits.

Abstract: Embodiments disclosed herein include stacked forksheet transistor devices, and methods of fabricating stacked forksheet transistor devices. In an example, an integrated circuit structure includes a backbone. A first transistor device includes a first vertical stack of semiconductor channels adjacent to an edge of the backbone. A second transistor device includes a second vertical stack of semiconductor channels adjacent to the edge of the backbone. The second transistor device is stacked on the first transistor device.

Abstract: Various embodiments include methods for reducing peak to average power ratio of wireless transmission waveforms, performed in transmitter circuitry of a wireless communication. Various embodiments may include receiving frequency domain data tones, transforming the frequency domain data tones to time domain data signals, generating, using a set of peak reduction tone (PRT) neural networks, time domain PRTs using the time domain data signals in which the set of PRT neural networks have been trained in conjunction with an augmentation neural network, and a receiver neural network, generating an output of the augmentation neural network based on an input of final combined time domain signals including the time domain PRTs combined with previous combined time domain signals, and generating time domain wireless transmission waveforms that include the output of the augmentation neural network combined with the final combined time domain signals.

Abstract: A request for a page is received. A processor is used to provide an optimized version of the page at least in part by determining dynamically an optimized content element for the page based at least in part on a user attribute associated with the request and content performance data generated based at least in part on testing of alternative versions of the page with respect to other users having the user attribute. The provided optimized version of the page is selected from the alternative versions of the page based on the content performance data indicating that the selected version of the page is optimized for the other users having the user attribute.

Abstract: The present disclosure relates to a method and apparatus for providing a camera service by an electronic device. The electronic device may include a camera, a memory, and at least one processor, the processor configured to control the electronic device to: receive a camera shooting request; display a preview image of an external object obtained through the camera using first area information based on a first shooting mode in response to the shooting request; receive an input corresponding to a second shooting mode in the first shooting mode; and display the preview using second area information corresponding to the second shooting mode in response to the input.

Abstract: The present invention provides a metal-plastic composite material having excellent adhesive strength and formability and a method for manufacturing the same, and specifically, the present invention provides a metal-plastic composite material and a method for manufacturing the same comprising: a metal layer; and a plastic layer on at least one surface of the metal layer, wherein a thin film layer formed of a silane coupling agent is provided between the metal layer and the plastic layer, and the metal layer and the plastic layer are bonded by covalent bonding with the silane coupling agent.

Abstract: Provided are an organic electric element, a display panel and a display device including the organic electric element which include a charge generating layer including a first layer comprising a first compound and a second compound and a second layer comprising a third compound so that they can have excellent efficiency or lifespan.

Abstract: A method of wireless communication by a user equipment (UE) includes receiving an uplink grant from a network device. The method also includes determining, by a lower analog media access control (MAC-A) layer of an analog data communications stack, a quantity of analog neural network gradients to transmit in a data packet using resources allocated by the uplink grant. The determining is based on an analog physical (PHY-A) layer encoding scheme and the resources allocated by the uplink grant. The method further includes segmenting the analog neural network gradients into the data packet. The method still further includes transferring the data packet to the PHY-A layer for transmission to the network device across a wireless network.