Abstract: A semiconductor structure is provided. The semiconductor structure comprises a substrate, a via, a liner layer, a barrier layer, and a conductor. The via penetrates through the substrate. The liner layer is formed on a sidewall of the via. The barrier layer is formed on the liner layer. The barrier layer comprises a conductive 2D material. The conductor fills a remaining space of the via.

Abstract: A memory device and a method for manufacturing the memory device are provided. The memory device includes a stack and a plurality of memory strings. The stack is disposed on the substrate, and the stack includes a plurality of conductive layers and a plurality of insulating layers alternately stacked. The memory strings pass through the stack along a first direction, wherein a first memory string in the memory strings includes a first conductive pillar and a second conductive pillar, a channel layer, and a memory structure. The first conductive pillar and the second conductive pillar respectively extend along the first direction and are separated from each other. The channel layer is disposed between the first conductive pillar and the second conductive pillar. The memory structure surrounds the second conductive pillar, and the memory structure includes a resistive memory material.

Abstract: A semiconductor element and a method for manufacturing the same are provided. The semiconductor element includes a plug and a via on the plug and electrically connected to the plug. The plug includes a tungsten plug and a conductive layer on the tungsten plug. The tungsten plug and the conductive layer include different materials. The tungsten plug has a first width in a lateral direction. The conductive layer has a second width in the lateral direction. The second width is greater than or equal to the first width. The conductive layer is between the via and the tungsten plug.

Abstract: The application provides a memory device and an operation method thereof. The memory device includes: a memory array, for processing model computation having a plurality of input values and a plurality of interact coefficients; and at least one calculation unit. In receiving the input values, a first part and a second part of the memory cells generate a first part and a second part of the common source currents, respectively. The first part of the memory cells is electrically isolated from the second part of the memory cells based on a diagonal of the memory array. The at least one calculation unit calculates a first part and a second part of a local field energy of the model computation based on the first part and the second part of the common source currents.

Abstract: A memory device includes a stack and a plurality of memory strings. The stack is disposed on the substrate, and the stack includes a plurality of conductive layers and a plurality of insulating layers alternately stacked. The memory strings pass through the stack along a first direction, wherein a first memory string in the memory strings includes a first conductive pillar and a second conductive pillar, a channel layer, and a memory structure. The first conductive pillar and the second conductive pillar respectively extend along the first direction and are separated from each other. The channel layer is disposed between the first conductive pillar and the second conductive pillar. The memory structure surrounds the second conductive pillar, and the memory structure includes a resistive memory material.

Abstract: Provided is an operation method for a memory device, the memory device being used for implementing an Artificial Neural Network (ANN). The operation method includes: reading from the memory device a weight matrix of a current layer of a plurality of layers of the ANN to extract a plurality of neuro values; determining whether to perform calibration; when it is determined to perform calibration, recalculating and updating a mean value and a variance value of the neuro values; and performing batch normalization based on the mean value and the variance value of the neuro values.

Abstract: A semiconductor structure is provided. The semiconductor structure includes a substrate and a via structure. The via structure is through the substrate. The via structure includes a first conductive portion, a second conductive portion, a first barrier portion, a second barrier portion, and a third barrier portion. The first conductive portion has a ring-shaped cross section. The second conductive portion is disposed at an inner side of the first conductive portion. The second conductive portion has a ring-shaped cross section. The first barrier portion is disposed at an outer side of the first conductive portion. The second barrier portion is disposed between the first conductive portion and the second conductive portion. The third barrier portion is disposed at an inner side of the second conductive portion. At least one of the first barrier portion, the second barrier portion, or the third barrier portion includes an insulating 2D material.

Abstract: A semiconductor structure and a manufacturing method for the same. The semiconductor structure includes a plug element and a via element. The plug element includes a tungsten plug. The plug element has a plug size in a lateral direction. The via element is electrically connected on the plug element. The via element is non-symmetrical with respect a center line of the plug element extending along a longitudinal direction. The via element has a via size in the lateral direction. The plug size is bigger than the via size.

Abstract: A semiconductor structure is provided. The semiconductor structure comprises a substrate, a via, a liner layer, a barrier layer, and a conductor. The via penetrates through the substrate. The liner layer is formed on a sidewall of the via. The barrier layer is formed on the liner layer. The barrier layer comprises a conductive 2D material. The conductor fills a remaining space of the via.

Abstract: A memory device includes a stack and a plurality of memory strings. The stack is disposed on the substrate, and the stack includes a plurality of conductive layers and a plurality of insulating layers alternately stacked. The memory strings pass through the stack along a first direction, wherein a first memory string in the memory strings includes a first conductive pillar and a second conductive pillar, a channel layer, and a memory structure. The first conductive pillar and the second conductive pillar respectively extend along the first direction and are separated from each other. The channel layer is disposed between the first conductive pillar and the second conductive pillar. The memory structure surrounds the second conductive pillar, and the memory structure includes a resistive memory material.

Abstract: A storage device for generating an identity code and an identity code generating method are disclosed. The storage device includes a first storage circuit, a second storage circuit and a reading circuit. The first storage circuit stores a plurality of first data and the first data have a plurality of bits. The second storage circuit stores a plurality of second data and the second data have a plurality of bits. The reading circuit reads the second data from the second storage circuit to form a first sequence, selects a first portion of the first data according to the first sequence, reads the first portion of the first data from the first storage circuit to form a target sequence and outputs the target sequence to serve as an identity code.

Abstract: A memory device and an operation method thereof are provided. The memory device comprises: a memory array including a plurality of memory cells; a plurality of bit lines coupled to the memory array; a plurality of word lines coupled to the memory array; and a plurality of conductance controllable units coupled to the memory array; wherein a memory cell group and at least one conductance controllable unit among the conductance controllable units form a logic operation unit, and a logic operation function of the logic operation unit is determined by an equivalent conductance of the at least one conductance controllable unit.

Abstract: A memory device and an operation method thereof are provided. The memory device comprises: a memory array including a plurality of memory cells; a plurality of bit lines coupled to the memory array; a plurality of word lines coupled to the memory array; and a plurality of conductance controllable units coupled to the memory array; wherein a memory cell group and at least one conductance controllable unit among the conductance controllable units form a logic operation unit, and a logic operation function of the logic operation unit is determined by an equivalent conductance of the at least one conductance controllable unit.

Abstract: A semiconductor structure and a manufacturing method for the same. The semiconductor structure includes a plug element and a via element. The plug element includes a tungsten plug. The plug element has a plug size in a lateral direction. The via element is electrically connected on the plug element. The via element is non-symmetrical with respect a center line of the plug element extending along a longitudinal direction. The via element has a via size in the lateral direction. The plug size is bigger than the via size.

Abstract: A semiconductor structure and a manufacturing method for the same. The semiconductor structure includes a plug element and a via element. The plug element includes a tungsten plug. The plug element has a plug size in a lateral direction. The via element is electrically connected on the plug element. The via element is non-symmetrical with respect a center line of the plug element extending along a longitudinal direction. The via element has a via size in the lateral direction. The plug size is bigger than the via size.

Abstract: An inference engine for a neural network uses a compute-in-memory array storing a kernel coefficients. A clamped input matrix is provided to the compute-in-memory array to produce an output vector representing a function of the clamped input vector and the kernel. A circuit is included receiving an input vector, where elements of the input vector have values in a first range of values. The circuit clamps the values of the elements of the input vector a limit of a second range of values to provide the clamped input vector. The second range of values is more narrow than the first range of values, and set according to the characteristics of the compute-in-memory array. The first range of values can be used in training using digital computation resources, and the second range of values can be used in inference using the compute-in-memory array.

Abstract: A semiconductor structure and a manufacturing method for the same. The semiconductor structure includes a plug element and a via element. The plug element includes a tungsten plug. The plug element has a plug size in a lateral direction. The via element is electrically connected on the plug element. The via element is non-symmetrical with respect a center line of the plug element extending along a longitudinal direction. The via element has a via size in the lateral direction. The plug size is bigger than the via size.

Abstract: Provided is an operation method for a memory device, the memory device being used for implementing an Artificial Neural Network (ANN). The operation method includes: reading from the memory device a weight matrix of a current layer of a plurality of layers of the ANN to extract a plurality of neuro values; determining whether to perform calibration; when it is determined to perform calibration, recalculating and updating a mean value and a variance value of the neuro values; and performing batch normalization based on the mean value and the variance value of the neuro values.

Abstract: Memory devices based on tungsten oxide memory elements are described, along with methods for manufacturing such devices. A memory device includes a plug extending upwardly from a top surface of a substrate through a dielectric layer; a bottom electrode having tungsten on an outside surface, the bottom electrode extending upwardly from a top surface of the plug; an insulating material in contact with the tungsten on the outside surface of, and surrounding, the bottom electrode; a memory element on an upper surface of the bottom electrode, the memory element comprising a tungsten oxide compound and programmable to at least two resistance states; and a top electrode overlying and contacting the memory element. The plug has a first lateral dimension, and the bottom electrode has a lateral dimension parallel with the first lateral dimension of the plug that is less than the first lateral dimension.

Abstract: A resistive memory includes a semiconductor substrate, a dielectric layer, an insulating layer and a metal electrode layer. The semiconductor substrate has a top surface and a recess extending downwards into the semiconductor substrate from the top surface. The dielectric layer is disposed on the semiconductor substrate and has a first through-hole aligning the recess. The insulating layer is disposed in the first through-hole and the recess. The metal electrode layer is disposed on the insulating layer by which the metal electrode layer is isolated from the semiconductor substrate.