Abstract: An inference operation method and a controlling circuit of a 3D NAND artificial intelligence accelerator are provided. The 3D NAND artificial intelligence accelerator includes a plurality of memory cells, a plurality of bit lines, a plurality of word lines and a plurality of string selecting line groups each of which includes at least one string selecting line. The inference operation method includes the following steps: The patterns are inputted to the bit lines. The word lines are switched to switch the filters. The string selecting line groups are switched to switch the filters. In a word line pioneering scheme and a string selecting line group pioneering scheme, when the patterns inputted to each of the bit lines are switched, any one of the word lines is not switched and any one of the string selecting line groups is not switched.

Abstract: An in-memory computing device including a plurality of memory cell arrays and a plurality of sensing amplifiers are provided. The memory cell arrays respectively receive a plurality of input signals. The input signals are divided into a plurality of groups. The groups respectively have at least one partial input signal. The at least one partial input signal of each of the groups has a same value. Numbers of the at least one partial input signal in the groups sequentially form a geometric sequence with a common ration equal to 2. The memory cell arrays respectively provide a plurality of weightings, and perform multiply-add operations respectively according to the received input signals and the weightings to generate a plurality of computation results. The sensing amplifiers respectively generate a plurality of sensing results according to the computation results.

Abstract: An inference operation method and a controlling circuit of a 3D NAND artificial intelligence accelerator are provided. The 3D NAND artificial intelligence accelerator includes a plurality of memory cells, a plurality of bit lines, a plurality of word lines and a plurality of string selecting line groups each of which includes at least one string selecting line. The inference operation method includes the following steps: The patterns are inputted to the bit lines. The word lines are switched to switch the filters. The string selecting line groups are switched to switch the filters. In a word line pioneering scheme and a string selecting line group pioneering scheme, when the patterns inputted to each of the bit lines are switched, any one of the word lines is not switched and any one of the string selecting line groups is not switched.

Abstract: A memory device is provided. The memory device includes a stacked structure, a tubular element, a conductive pillar and memory cells. The tubular element includes a dummy channel layer and penetrates the stacked structure. The conductive pillar is enclosed by the tubular element and extending beyond a bottom surface of the dummy channel layer. The memory cells are in the stacked structure and electrically connected to the conductive pillar.

Abstract: A memory device comprises a stack of patterned conductor layers, at least a plurality of the layers comprising conductive strips including strips continuous with a pad and other strips isolated from the pad. An array of vertical pillars extends through the stack of patterned conductor layers, wherein memory cells are disposed at cross-points between the vertical pillars and patterned conductor layers. The array has an array boundary proximal to the pad. A first set of isolation blocks extends through the plurality of patterned conductor layers separating the strips continuous with the pad from the other strips isolated from the pad. A second set of isolation blocks inside the array boundary extends through the plurality of patterned conductor layers isolating the other strips from the pad.

Abstract: Provided is an operation method for a memory device, the operation method comprising: performing an erase operation; performing a verify-read operation on a memory cell to generate a cell current, the memory cell including a first transistor and a second transistor; checking whether the cell current is lower than a first cell current threshold; when the cell current is not lower than the first cell current threshold, increasing a memory gate voltage until the cell current is lower than the first cell current threshold, wherein the memory gate voltage is applied to the first transistor; fixing the memory gate voltage and increasing a drain voltage; checking whether the cell current is lower than a second cell current threshold; and if the cell current is not lower than the second cell current threshold, increasing the drain voltage until the cell current is lower than the second cell current threshold.

Abstract: A semiconductor structure includes a stack of memory cells and a CMOS structure. The CMOS structure is located below the stack of memory cells. The CMOS structure includes a source line transistor and a bit line transistor.

Abstract: Provided is an operation method for a memory device, the operation method comprising: performing an erase operation; performing a verify-read operation on a memory cell to generate a cell current, the memory cell including a first transistor and a second transistor; checking whether the cell current is lower than a first cell current threshold; when the cell current is not lower than the first cell current threshold, increasing a memory gate voltage until the cell current is lower than the first cell current threshold, wherein the memory gate voltage is applied to the first transistor; fixing the memory gate voltage and increasing a drain voltage; checking whether the cell current is lower than a second cell current threshold; and if the cell current is not lower than the second cell current threshold, increasing the drain voltage until the cell current is lower than the second cell current threshold.

Abstract: An in-memory computation device including a memory array, p×q analog to digital converters (ADCs) and a ladder adder is provided. The memory array is divided into p×q memory tiles, where p and q are positive integers larger than 1. Each of the memory tiles has a plurality local bit lines coupled to a global bit line respectively through a plurality of bit line selection switches. The bit line selection switches are turned on or cur off according to a plurality of control signals. The memory array receives a plurality of input signals. The ADCs are respectively coupled to a plurality of global bit lines of the memory tiles. The ADCs respectively convert electrical signals on the global bit lines to generate a plurality of sub-output signals. The ladder adder is coupled to the ADCs, and performs an addition operation on the sub-output signals to generate a calculation result.

Abstract: A memory device and a manufacturing method for the same are provided. The memory device comprises a stack structure and a channel structure. The stack structure is on a substrate and comprises gate electrodes and insulating films stacked alternately. The channel structure is electrically coupled to the gate electrodes, and is on sidewall surfaces of the gate electrodes. The channel structure comprises a first channel structure and a second channel structure. The second channel structure is on an upper surface of the first channel structure. The first channel structure and/or the second channel structure has a ring shape.

Abstract: A semiconductor structure and a method for manufacturing the same. The semiconductor structure includes a first staircase structure, a second staircase structure, a conductive pillar, and a contact pillar. The first staircase structure includes conductive stair layers. The conductive pillar passes through the second staircase structure. The conductive pillar has an upper conductive pillar end and a lower conductive pillar end opposing to the upper conductive pillar end. The contact pillar is electrically connected on one conductive stair layer of the conductive stair layers. The contact pillar has an upper contact end and a lower contact end opposing to the upper contact end. The upper contact end and the lower contact end are respectively electrically connected to the upper conductive pillar end of the conductive pillar and the one conductive stair layer.

Abstract: A memory apparatus and compensation method for a computation result thereof are provided. The memory apparatus includes a memory sub-block, a reference memory sub-block and a control circuit. During a computation phase, the memory sub-block receives an input signal, and generates a computation result by a multiply-accumulate operation according to the input signal. The reference memory sub-block includes a plurality of memory cells pre-programmed with a reference weight value. The reference memory sub-block receives a reference input signal during a calibration phase, and generates a reference computation value by a multiply-accumulate operation according to the reference input signal and the reference weight value. The control circuit generates an adjustment value according to the reference computation value and a standard computation value, and during the computation phase, adjusts the computation result according to the adjustment value to generate an adjusted computation result.

Abstract: A three-dimensional memory device includes a plurality of conductive layers and insulating layers alternately formed to define a multi-layer stacked structure. The multi-layer stacked structure includes a stair region and an non-stair region, the stair region includes a plurality of steps, each step includes an immediately-adjacent pair of the conductive layers and insulating layers. A plurality of memory structures are located in the non-stair region, and each memory structure passes through the conductive layers and the insulating layers. A fishbone dielectric structure includes a main bone and a plurality of side bones extending from the main bone in the stair region, wherein the main bone crosses the memory structures in the non-stair region.

Abstract: A memory device comprises a stack of patterned conductor layers, at least a plurality of the layers comprising conductive strips including strips continuous with a pad and other strips isolated from the pad. An array of vertical pillars extends through the stack of patterned conductor layers, wherein memory cells are disposed at cross-points between the vertical pillars and patterned conductor layers. The array has an array boundary proximal to the pad. A first set of isolation blocks extends through the plurality of patterned conductor layers separating the strips continuous with the pad from the other strips isolated from the pad. A second set of isolation blocks inside the array boundary extends through the plurality of patterned conductor layers isolating the other strips from the pad.

Abstract: A spiking neural networks circuit and an operation method thereof are provided. The spiking neural networks circuit includes a bit-line input synapse array and a neuron circuit. The bit-line input synapse array includes a plurality of page buffers, a plurality of bit line transistors, a plurality of bit lines, a plurality of memory cells, one word line, a plurality of source lines and a plurality of source line transistors. The page buffers provides a plurality of data signals. Each of the bit line transistors is electrically connected to one of the page buffers. Each of the bit lines receives one of the data signals. The source line transistors are connected together. The neuron circuit is for outputting a feedback pulse.

Abstract: An artificial intelligence accelerator receives a binary input data set and a selected layer of layers of overall weight pattern. The artificial intelligence accelerator includes processing tiles and a summation output circuit. Each processing tile receives one of input data subsets of the input data set and performs a convolution operation on weight blocks of each sub weight pattern of the overall weight pattern to obtain weight operation values and then obtains a weight output value expected from a direct convolution operation on the input data subset with the sub weight pattern through performing a multistage shifting and adding operation on the weight operation values. The summation output circuit sums up the plurality of weight output values through a multistage shifting and adding operation, so as to obtain a sum value expected from a direct convolution operation performed on the input data set with the overall weight pattern.

Abstract: A three-dimensional memory device includes a plurality of conductive layers and insulating layers alternately formed to define a multi-layer stacked structure. The multi-layer stacked structure includes a stair region and an non-stair region, the stair region includes a plurality of steps, each step includes an immediately-adjacent pair of the conductive layers and insulating layers. A plurality of memory structures are located in the non-stair region, and each memory structure passes through the conductive layers and the insulating layers. A fishbone dielectric structure includes a main bone and a plurality of side bones extending from the main bone in the stair region, wherein the main bone crosses the memory structures in the non-stair region.

Abstract: An in-memory computing device including a plurality of memory cell arrays and a plurality of sensing amplifiers are provided. The memory cell arrays respectively receive a plurality of input signals. The input signals are divided into a plurality of groups. The groups respectively have at least one partial input signal. The at least one partial input signal of each of the groups has a same value. Numbers of the at least one partial input signal in the groups sequentially form a geometric sequence with a common ration equal to 2. The memory cell arrays respectively provide a plurality of weightings, and perform multiply-add operations respectively according to the received input signals and the weightings to generate a plurality of computation results. The sensing amplifiers respectively generate a plurality of sensing results according to the computation results.

Abstract: The present disclosure relates to a non-volatile memory and operating method thereof. The non-volatile memory includes multiple memory strings, multiple bit switch units, a memory operation circuit and multiple source switch units. The bit switch units are electrically connected to the memory strings. The memory operation circuit is electrically connected to the bit switch units to transmit a write signal to the memory unit strings. The source switch units are electrically connected to the memory string so that the memory strings receive a bias signal via the source switch unit. In a program mode, when a first bit switch unit of the bit switch units is turned on and a first memory strings receives the write signal through the first bit switch unit, the source switch units electrically connected to the other memory strings will be turned on.

Abstract: A three-dimensional memory device includes a substrate, conductive layers and insulating layers, a storage layer, a first channel, a second channel and a first conductive plug. The conductive layers and insulating layers are alternately stacked over the substrate to form a multi-layer stacked structure. The storage layer penetrates through the multi-layer stacked structure, and has a first string portion and a second string portion that are spaced from each other. The first channel is located on a lateral side of the first string portion. The second channel is located on a lateral side of the second string portion. The first channel and the second channel have an upper channel portion and a lower channel portion. The first conductive plug is interconnected between the upper channel portion and the lower channel portion.