Abstract: A processor includes at least one socket and at least one memory. Each socket includes a first die and a second die. The first die receives a boot-enable signal and an internal boot-enable signal to execute a boot procedure, and outputs a boot-completion signal after completing the boot procedure. The second die receives the internal boot-enable signal and the boot-completion signal from the first die to execute the boot procedure. The second die is electrically connected to the first die through a communication bus. The memory is electrically connected to the second die. When the first die executes the boot procedure, the first die accesses the memory through the communication bus and the second die.

Abstract: A method for machine reading comprehension includes: S1, obtaining a character-level indication vector of a question and a character-level indication vector of an article; S2, obtaining an encoded question vector and an encoded article vector; S3, obtaining an output P1 of a bidirectional attention model and an output P2 of a shared attention model; S4, obtaining an aggregated vector P3; S5, obtaining a text encoding vector P4; S6, obtaining global interaction information between words within the article; S7, obtaining a text vector P5 after using the self-attention model; S8, obtaining aggregated data P6 according to the text encoding vector P4 and the text vector P5; S9, obtaining a context vector of the article according to the aggregated data P6 and an unencoded article vector P; and S10, predicting an answer position according to the context vector of the article and the encoded question vector to complete the machine reading comprehension.

Abstract: A method for machine reading comprehension includes: S1, obtaining a character-level indication vector of a question and a character-level indication vector of an article; S2, obtaining an encoded question vector and an encoded article vector; S3, obtaining an output P1 of a bidirectional attention model and an output P2 of a shared attention model; S4, obtaining an aggregated vector P3; S5, obtaining a text encoding vector P4; S6, obtaining global interaction information between words within the article; S7, obtaining a text vector P5 after using the self-attention model; S8, obtaining aggregated data P6 according to the text encoding vector P4 and the text vector P5; S9, obtaining a context vector of the article according to the aggregated data P6 and an unencoded article vector P; and S10, predicting an answer position according to the context vector of the article and the encoded question vector to complete the machine reading comprehension.