Abstract: Methods, systems, and devices for wireless communications are described. A transmitting device may transmit an expiration indication to a receiving device as part of a scheduling message for a transport block. The expiration indication may provide information related to an expiration time for the transport block. If the expiration time is reached prior to successful reception by a receiving device, the receiving device may assume that the transport block has expired and may refrain from transmitting a retransmission grant, or may empty a hybrid automatic repeat request (HARM) buffer associated with the transport block. If the transmitting device fails to successfully receive an indication from the receiving device of a successful reception of the transport block prior to the expiration period, the transmitting device may also assume that the transport block has expired.

Abstract: A method and apparatus for hydrocracking mineralized refuse pyrolysis oil. The method may use the following steps: (a) crushing and pyrolyzing mineralized refuse to obtain arene and alkane precursor biomass oil; (b) hydrogenating the arene and alkane precursor biomass oil obtained in step (a), and separating the obtained hydrocrackate to obtain arene and alkane; and (c) purifying, recovering and optimizing the arene and alkane obtained in step (b), and performing deep processing to produce naphtha, jet fuel, light diesel oil, and heavy diesel oil.

Abstract: The present specification discloses video matching. In a computer-implemented method, a plurality of feature vectors of a target video is obtained. A candidate video similar to the target video is retrieved from a video database based on the plurality of feature vectors of the target video. A time domain similarity matrix feature map is constructed between the target video and the candidate video based on the target video and the candidate video. Using the time domain similarity matrix feature map as an input into a deep learning detection model, a video segment matching the target video in the candidate video and a corresponding similarity is output.

Abstract: A method of forming a three-dimensional (3D) memory device includes: forming, over a substrate, a layer stack having alternating layers of a first conductive material and a first dielectric material; forming trenches extending vertically through the layer stack from an upper surface of the layer stack distal from the substrate to a lower surface of the layer stack facing the substrate; lining sidewalls and bottoms of the trenches with a memory film; forming a channel material over the memory film, the channel material including an amorphous material; filling the trenches with a second dielectric material after forming the channel material; forming memory cell isolation regions in the second dielectric material; forming source lines (SLs) and bit lines (BLs) that extend vertically in the second dielectric material on opposing sides of the memory cell isolation regions; and crystallizing first portions of the channel material after forming the SLs and BLs.

Abstract: Disclosed is a method of wireless communication. The method is performed at a first wireless node. The method comprises: obtaining a sequence of bits to be encoded, selecting a puncturing pattern from a plurality of puncturing patterns, generating, based on the sequence of bits to be encoded, a sequence of parity bits in accordance with a binary linear block coding scheme, puncturing, based on the selected puncturing pattern, at least one of the sequence of bits to be encoded or the sequence of parity bits, generating a plurality of modulation symbols based on remaining bits in the sequence of bits to be encoded and based on remaining bits in the sequence of parity bits, and transmitting, to a second wireless node, the plurality of modulation symbols.

Abstract: A memory device includes: a first conductor extending in parallel with a first axis; a first selector material comprising a first portion that extends along a first sidewall of the first conductor; a second selector material comprising a first portion that extends along the first sidewall of the first conductor; a first variable resistive material comprising a portion that extends along the first sidewall of the first conductor; and a second conductor extending in parallel with a second axis substantially perpendicular to the first axis, wherein the first portion of the first selector material, the first portion of the second selector material, and the portion of the first variable resistive material are arranged along a first direction in parallel with a third axis substantially perpendicular to the first axis and second axis.

Abstract: Methods, systems, and devices for wireless communication are described. In one example, a base station may dynamically configure a user equipment (UE) to monitor or avoid monitoring for a preemption indication. Accordingly, the UE may have a lower chance of monitoring for a preemption indication when it is unlikely that an uplink or downlink transmission will be preempted. In another example, a base station may transmit a control message indicating whether future indications received from the base station are to be interpreted as preemption indications or permission indications. Accordingly, the base station may choose to use either preemption indications or permission indications (e.g., based on the probability of collisions between mobile broadband (MBB) and low latency transmissions) to facilitate MBB and low latency communication multiplexing with limited signaling.

Abstract: In the present disclosure, a method for compressing a feature map is provided, where the feature map is generated by passing a first input through a deep neural network (DNN). A respective optimal index order and a respective optimal unifying method are determined for each of super-blocks that are partitioned from the feature map. A selective structured unification (SSU) layer is subsequently determined based on the respective optimal index order and the respective optimal unifying method for each of the super-blocks. The SSU layer is added to the DNN to form an updated DNN, and is configured to perform unification operations on the feature map. Further, a first estimated output is determined, where the first estimated output is generated by passing the first input through the updated DNN.

Abstract: A method and apparatus for neural network based cross component prediction with low-bit precision during encoding or decoding of an image frame or a video sequence, which may include reconstructing a chroma component based on a received luma component using a pre-trained deep neural network (DNN) cross component prediction (CCP) model for chroma prediction, and updating a set of parameters of the pre-trained DNN CCP model with low-bit precision. The method may also include generating an updated DNN CCP model for chroma prediction with low-bit precision based on at least one video sequence, and using the updated DNN CCP model for cross component prediction of the at least one video sequence at reduced processing time.

Abstract: A fish school detection method and a system thereof, an electronic device and a storage medium are provided, the method includes inputting a to-be-detected fish school image into a fish school detection model; the fish school detection model including a feature extraction layer, a feature fusion layer and a feature recognition layer; extracting feature information of the to-be-detected fish school image based on the feature extraction layer, and determining a fish school feature map and an attention feature map based on an attention mechanism; fusing the fish school feature map and the attention feature map based on the feature fusion layer to determine a target fusion feature map; and determining a target fish school detection result based on the feature recognition layer and the target fusion feature map. Interference from environmental factors on detection results is eliminated, so as to effectively improve accuracy of the fish detection.

Abstract: A method is provided for quantizing a neural network model performed by a processing system. The method comprises determining a scaling factor based on a distribution of weights associated with the neural network model, determining quantized weights based on the scaling factor and the weights associated with the distribution, determining a training loss of the neural network model based on the quantized weights during training of the neural network model, and determining an updated scaling factor for the neural network model based on a gradient of the training loss.

Abstract: Methods, systems, and devices for wireless communication are described. For example, the described techniques provide for a wireless device transmitting an initial transmission associated with a data payload, such as a hybrid automatic repeat request (HARQ) payload. The initial transmission may include a set of bits modulated using a first bit ordering scheme, and the set of bits may include a first bit subset that are transmitted with a lower transmission reliability than a second bit subset of the set of bits. The wireless device may modulate the first bit subset using a second bit ordering scheme that differs from the first bit ordering scheme to generate a modulated bit set. The second bit ordering scheme may be associated with a higher transmission reliability for the first bit subset. The wireless device may transmit the second transmission for the data payload with the modulated bit set.

Abstract: Provided is a semiconductor device includes a gate electrode, a gate dielectric layer, a channel layer, an insulating layer, a first source/drain electrode and a second source/drain electrode, a second dielectric layer, and a stop segment. The gate electrode is located within a first dielectric layer that overlies a substrate. The gate dielectric layer is located over the gate electrode. The channel layer is located on the gate dielectric layer. The insulating layer is located over the channel layer. The first source/drain electrode and the second source/drain electrode are located in the insulating layer, and connected to the channel layer. The second dielectric layer is beside one of the first source/drain electrode and the second source/drain electrode. The stop segment is embedded in the second dielectric layer.

Abstract: Embodiments of the disclosure provide a method, apparatus, device, and storage medium for content capturing. The method includes, in response to a capturing start instruction, presenting a capturing start page which including at least a capturing control; in response to detecting a trigger instruction for the capturing control, switching from the capturing start page to a capturing page for video capturing, the capturing page including at least a stop control and a pause control, and the stop control being presented in a more prominent way than the pause control; and in response to detecting a trigger instruction for the stop control, switching from the capturing page to a video editing page, the video editing page being used to edit a captured video. In this way, it is possible for users to quickly perform video capturing and directly enter the editing page to complete editing of a work.

Abstract: Provided are catechol nanoparticles, catechol protein nanoparticles, and a preparation method and use thereof. The method includes: adding a tannin compound-containing natural herb medicine into water to obtain a mixture, and subjecting the mixture to heating reflux extraction to obtain a herb medicine extract and subjecting the herb medicine extract to fractionation to obtain the catechol nanoparticles.

Abstract: A method performed by at least one processor includes the following steps: generating a layout of an integrated circuit (IC), the layout comprising a cell and a layout context in a vicinity of the cell; receiving from a library a set of context groups and a set of timing tables, wherein each of the context groups is associated with one of the set of timing tables; determining a representative context group for the cell through comparing the layout context of the cell with the set of context groups; and performing a timing analysis on the layout according to a representative timing table associated with the representative context group for the cell.

Abstract: Disclosed in the present disclosure is a double-deck multi-span bridge construction method. According to the double-deck bridge construction method of the present disclosure, construction is carried out by using a method of disassembling a support jig frame in a graded and span-separated mode, an upper chord jig frame and a lower chord jig frame can be used in a recycle manner, and construction costs are reduced. In addition, a construction period of building the support jig frame is shortened, and other construction operations can be synchronously carried out on a span in which the jig frame is disassembled, for example, fire retardant coating construction can be carried out on a mounted bridge deck after the jig frame is disassembled, and the construction period of a double-deck multi-span bridge is effectively shortened.

Abstract: A computer readable storage medium is provided. When contents of the computer readable storage medium are executed by a processor, multi-photon imaging may be performed on a histopathological section containing tumor environment information, and pathological partitioning of a tumor microenvironment may be further performed through image processing. A value of each collagen feature parameters, such as a morphological feature parameter, an energy feature parameter and a texture feature parameter, may be extracted from a tumor tissue region, an invasive margin (IM) region and a normal tissue (N) region. An inter-region difference and a variation may be calculated according to feature parameters of regions. A collagen feature scoring model may be established. A collagen feature score may be calculated with the collagen feature parameters input to the model.

Abstract: A method, computer program, or computer system is provided for compressing a neural network model. One or more blocks are identified from among a superblock corresponding to a multi-dimensional tensor associated with a neural network. A set of weight coefficients associated with the superblock is unified. A model of the neural network is compressed based on the unified set of weight coefficients.

Abstract: Systems, apparatus and methods are provided for managing a removable solid state storage system for data loss prevention. A method may include maintaining a standby mode for a timer of the removable solid state storage system until the removable solid state storage system is disconnected from an external power supply, setting an operation time interval on the timer, using the timer to count how long the removable solid state storage system has been disconnected, sending an interrupt to a storage controller of the removable solid state storage system from the timer when the timer counts to the operation time Interval, and performing data loss prevention operations using a power supplied by a removable battery.