Sequence model processing method and apparatus

A computing device, comprising a processor and a memory;

• the memory being configured to store a plurality of instructions;
• and
• the processor being configured to execute the instructions in the memory, to perform a plurality of operations including:
• inputting a first sequence of words having a first language into an encoder side of a sequence model, the encoder side comprising a self-attention encoder and a temporal encoder, wherein the temporal encoder comprises an encoding layer having a time series information network and a first feed-forward network;
• encoding the first sequence of words having the first language using the temporal encoder to obtain a first encoding result, the first encoding result comprising time series information obtained by performing time series modeling on the first sequence of words having the first language, the encoding further including:
• performing time series modeling on the first sequence of words having the first language by using the time series information network, to obtain the time series information;
• and
• unidirectionally propagating the time series information by using the first feed-forward network, and outputting the first encoding result;
• while encoding the first sequence of words having the first language using the temporal encoder, encoding, in parallel, the first sequence of words having the first language using the self-attention encoder, to obtain a second encoding result;
• inputting a target sequence, the first encoding result, and the second encoding result into a decoder side of the sequence model;
• and
• decoding the target sequence, the first encoding result, and the second encoding result using the decoder side to obtain a second sequence of words having a second language that is different from the first language, and outputting the second sequence of words in the second language.

A non-transitory computer-readable storage medium storing a plurality of instructions that, when executed by a processor of a computing device, cause the computing device to perform a plurality of operations including:

• inputting a first sequence of words having a first language into an encoder side of a sequence model, the encoder side comprising a self-attention encoder and a temporal encoder, wherein the temporal encoder comprises an encoding layer having a time series information network and a first feed-forward network;
• encoding the first sequence of words having the first language using the temporal encoder to obtain a first encoding result, the first encoding result comprising time series information obtained by performing time series modeling on the first sequence of words having the first language, the encoding further including:
• performing time series modeling on the first sequence of words having the first language by using the time series information network, to obtain the time series information;
• and
• unidirectionally propagating the time series information by using the first feed-forward network, and outputting the first encoding result;
• while encoding the first sequence of words having the first language using the temporal encoder, encoding, in parallel, the first sequence of words having the first language using the self-attention encoder, to obtain a second encoding result;
• inputting a target sequence, the first encoding result, and the second encoding result into a decoder side of the sequence model;
• and
• decoding the target sequence, the first encoding result, and the second encoding result using the decoder side to obtain a second sequence of words having a second language that is different from the first language, and outputting the second sequence of words in the second language.