Abstract: A semiconductor package and the method of forming the same are provided. The semiconductor package may include a substrate, an integrated circuit package component having a semiconductor die bonded to the substrate, and a ring structure on the substrate, wherein the ring structure may encircle the integrated circuit package component in a top-down view. The ring structure may comprise a first attached segment, a second attached segment attached to the substrate by an adhesive, and a first suspended segment between the first attached segment and the second attached segment. The first suspended segment may be suspended over the substrate. The first attached segment and the second attached segment may be spaced apart from the package component by a first distance and a second distance, respectively. The first suspended segment may be spaced apart from the package component by a third distance different from the first distance and the second distance.

Abstract: A semiconductor package and a manufacturing method thereof are provided. The semiconductor package includes at least one semiconductor die, an interposer, an encapsulant, a protection layer and connectors. The interposer has a first surface, a second surface opposite to the first surface and sidewalls connecting the first and second surfaces. The semiconductor die is disposed on the first surface of interposer and electrically connected with the interposer. The encapsulant is disposed over the interposer and laterally encapsulating the at least one semiconductor die. The connectors are disposed on the second surface of the interposer and electrically connected with the at least one semiconductor die through the interposer. The protection layer is disposed on the second surface of the interposer and surrounding the connectors. The sidewalls of the interposer include slanted sidewalls connected to the second surface, and the protection layer is in contact with the slant sidewalls of the interposer.

Abstract: A semiconductor package module includes a package, a conductive layer, and a heat dissipating module. The package includes a semiconductor die. The conductive layer is disposed over the package. The heat dissipating module is disposed over the conductive layer, and the package and the heat dissipating module prop against two opposite sides of the conductive layer, where the heat dissipating module is thermally coupled to and electrically isolated from the package through the conductive layer.

Abstract: This application is directed to decoding video data including a plurality of luma components of a plurality of pixels in a video frame. The plurality of pixels belong to a coding block, and include a boundary pixel inside the coding block wherein the boundary pixel is immediately adjacent to a boundary of the coding block. One or more neighboring pixels of the boundary pixel are outside of the coding block and determined as being not available. A respective luma component corresponding to the boundary pixel is assigned to a luma component corresponding to each of the one or more neighboring pixels. A boundary luma component is determined based, at least, on the luma components of the one or more neighboring pixels and the boundary pixel according to a predefined luma interpolation scheme. A boundary chroma component is converted from the boundary luma component according to a linear mapping model.

Abstract: An electronic apparatus performs a method of decoding video data. The method comprises: receiving, from bitstream, video data corresponding to a transform-skip mode coded block; decoding, from the video data, a first codeword, a second codeword and a first group of codewords for a pixel within the transform-skip mode coded block; deriving an initial level value from the first group of codewords; converting the first codeword into a remainder of the pixel in accordance with a predefined mapping relationship that is generated using a constant Rice parameter; converting the second codeword into a sign value of the remainder; and deriving a quantified residual of the pixel from the remainder, the sign value and the initial level value.

Abstract: An electronic apparatus performs a method of encoding and decoding video data. The method comprises: decoding a first coding block and a second coding block that shares a common edge on a first picture, wherein decoding the first coding block and the second coding block includes reconstructing a first residual block for the first coding block and a second residual block for the second coding block; determining that the first picture has a first resolution, a first reference picture corresponding to the first coding block has a second resolution, and a second reference picture corresponding to the second coding block has a third resolution; deriving a deblocking strength (bS) value based, at least in part, on the first resolution, second resolution, and the third resolution; and performing in-loop filtering on the reconstructed first residual block and the reconstructed second residual block using a deblocking filter in accordance with the derived bS value.

Abstract: This application is directed to processing video data that includes a plurality of luma samples and a plurality of chroma samples corresponding to a plurality of pixel groups of a video frame. For each pixel group, an electronic device identifies a respective chroma sample and a set of luma samples and determines an anchor luma sample from the set of luma samples. A chroma refinement value is generated based on the set of luma samples by differencing a respective luminance value of each luma sample in the set by an anchor luminance value of the anchor luma sample and applying a cross component filter to the difference luminance values of the set of luma samples. The electronic device then updates the chroma sample using the chroma refinement value for each pixel group and stores the updated respective chroma sample of each pixel group in association with the video frame.

Abstract: A stacked via structure disposed on a conductive pillar of a semiconductor die is provided. The stacked via structure includes a first dielectric layer, a first conductive via, a first redistribution wiring, a second dielectric layer, a second conductive via, and a second redistribution wiring. The first dielectric layer covers the semiconductor die. The first conductive via is embedded in the first dielectric layer and electrically connected to the conductive pillar. The first redistribution wiring covers the first conductive via and the first dielectric layer. The second dielectric layer covers the first dielectric layer and the first redistribution wiring. The second conductive via is embedded in the second dielectric layer and landed on the first redistribution wiring. The second redistribution wiring covers the second conductive via and the second dielectric layer. A lateral dimension of the first conductive via is greater than a lateral dimension of the second conductive via.

Abstract: This application is directed to coding video data including a plurality of transform blocks according to a transform skip mode. Two or more control flags are obtained with the video data to indicate signaling of a Rice parameter for the transform blocks on different hierarchical levels. In accordance with a determination that a first control flag is enabled on a first hierarchical level and a second control flag is enabled on a second hierarchical level lower than the first hierarchical level, entropy coding is performed for a subset of the transform blocks based on a Rice parameter obtained with the video data. In accordance with a determination that the first control flag is enabled and that the second control flag is disabled on the second hierarchical level, entropy coding is performed for the subset of the transform blocks based on a predetermined Rice parameter at the second hierarchical level.

Abstract: Methods and apparatuses for video coding are provided. A method includes signaling a first syntax element for determining whether information of one or more reference picture lists is present in a picture header (PH) associated with a picture; and signaling the information of the one or more reference picture lists in the PH in a case that the first syntax element has a first value, wherein in a case that the information of the one or more reference picture lists is signaled in the PH and one or more slices associated with the picture are determined, from the information of the one or more reference picture lists, as being not bi-predictive, one or more second syntax elements in the PH are not parsed.

Abstract: Methods and apparatuses for video coding are provided. The method includes that a decoder determines whether a disable flag is present in a picture header (PH) associated with a picture, wherein the disable flag specifies whether a coding tool is disabled in one or more slices associated with the PH. Additionally, the method includes that the decoder infers value of the disable flag according to one or more enable flags signaled in a sequence parameter set (SPS) of the picture in response to determining that the disable flag is not present in the PH, inferring, by the decoder.

Abstract: A spin orbit torque magnetoresistive random access memory (SOT MRAM) includes at least a spin current source alloy layer, a ferromagnetic free layer, and an insulation layer. The spin current source alloy layer is a nickel-tungsten alloy layer. The ferromagnetic free layer is located on the spin current source alloy layer. The insulation layer is located on the ferromagnetic free layer. Since the nickel-tungsten alloy layer has favorable perpendicular magnetic anisotropic and can maintain a high spin Hall angle, it is suitable as a spin current source for the SOT MRAM.

Abstract: An electronic apparatus performs a method of encoding video data. The method comprises: determining a first syntax element for a first set of palette mode coding units at a first coding level; in accordance with a determination that the first syntax element has a first predefined value: choosing a first binarization process for samples of the first set of palette mode coding units; encoding, into bitstream, the samples of the first set of palette mode coding units using the first binarization process; in accordance with a determination that the first syntax element does not have the first predefined value: choosing a second binarization process for the samples of the first set of palette mode coding units, the second binarization process being different from the first binarization process; and encoding, into the bitstream, the samples of the first set of palette mode coding units using the second binarization process.

Abstract: Methods and devices are provided for reducing the decoding latency introduced by LMCS. In one method, during encoding of a coding unit (CU), a plurality of reconstructed luma samples is selected from a first pre-determined region neighboring to a second pre-determined region where the CU is located, an average of the plurality of reconstructed luma samples is calculated, and the average of the plurality of reconstructed luma samples is used directly, without any clipping, in deriving a chroma residual scaling factor, a bitstream comprising luma mapping with chroma scaling (LMCS) related information is formed.

Abstract: An electronic apparatus performs a method of coding video data. The method comprises: receiving, from the bitstream, the video data corresponding to a coding unit, wherein the coding unit is coded by intra-prediction mode or inter-prediction mode; receiving a first syntax element from the video data, wherein the first syntax element indicates whether the coding unit has been coded using adaptive color-space transform (ACT); processing the video data to generate residuals of the coding unit; in accordance with a determination based on the first syntax element that the coding unit has been coded using the ACT, performing a clipping operation that confines a dynamic range of the residuals of the coding unit within a predefined range; and applying an inverse ACT to the residuals of the coding unit after the clipping operation.

Abstract: A semiconductor structure includes a circuit substrate, a semiconductor die, and a cover. The semiconductor die is disposed on the circuit substrate. The cover is disposed over the semiconductor die and over the circuit substrate. The cover comprises a lid portion and a support portion. The structure includes a first adhesive bonding the support portion to the circuit substrate and a second adhesive bonding the support portion and the lid portion.

Abstract: Methods, apparatuses, and non-transitory computer-readable storage mediums are provided for encoding a video signal. An encoder determines a reference picture associated with a video block within the video signal and obtains reference samples of the video block from the reference picture. The encoder determines a luma interpolation filter for the video block coded in an affine motion mode based on a scaling ratio derived from resolutions of the reference picture and a current picture. The encoder further obtains luma inter prediction samples of the video block by applying the luma interpolation filter to the reference samples.

Abstract: A wireless device includes a satellite receiver to receive data from multiple satellites. The wireless device also includes processing circuitry and memory. The memory stores one or more neural network models. The processing circuitry is operative to identify a neural network model that has been trained to adapt to a region in which the wireless device operates, classify satellite raw measurements from each satellite at a given time into a corresponding quality level using the neural network model, and identify satellite raw measurements with a quality level higher than a threshold. The location of the wireless device is calculated using the identified satellite raw measurements.

Abstract: Aspects of the invention include techniques for improving the accuracy of access-limited neural network inference in low-voltage regimes. A non-limiting example method includes training a first machine learning model to perform input transformation for reducing low-voltage bit errors for a deep neural network operating in a low-voltage regime. The training includes inputting training data into the first machine learning model such that, in response, the first machine learning model produces transformed training data; inputting the transformed training data into a clean machine learning model and into perturbed machine learning models, the perturbed machine learning models being generated by applying random bit errors to the clean machine learning model; and optimizing the first machine learning model based on a comparison of output of the clean machine learning model and of the perturbed machine learning models compared to groundtruth labels for the training data.

Abstract: A semiconductor package includes a circuit board structure, a first redistribution layer structure and first bonding elements. The circuit board structure includes outermost first conductive patterns and a first mask layer adjacent to the outermost first conductive patterns. The first redistribution layer structure is disposed over the circuit board structure. The first bonding elements are disposed between and electrically connected to the first redistribution layer structure and the outermost first conductive patterns of the circuit board structure. In some embodiments, at least one of the first bonding elements covers a top and a sidewall of the corresponding outermost first conductive pattern.