Shenzhen University

Shubin Cai

Zhong Ming

Xiaogang Feng

Ningsheng Yang

Huaifeng Zhou

A pre-training language model-based summarization generation method, including: acquiring text information of a summarization to be generated, and performing multi-feature weighted language pre-training processing on the text information to obtain a candidate summarization; inputting the candidate summarization into a pre-training language model to obtain output data of the pre-training language model; and inputting the output data of the pre-training language model into a decoder model to obtain a target summarization. The pre-training processing is performed on the text information of the summarization to be generated, and then the target summarization can be generated after processing performed by an encoder and a decoder, improving the quality and readability of the summarization.

Pre-training language model-based summarization generation method

Disclosed is a sharding block chain generation method based on a tree structure. Method comprises: dividing a server system into at least one group, and respectively selecting a leader of each group; packaging to generate at least one block by leader according to an own local database state, and broadcasting at least one block to all members of group in which leader is located; verifying consistency of at least one block, executing block according to a verification result of consistency, and synchronizing block to other groups. According to method, by means of tree grouping architecture, efficiency of generating block by a block chain framework is improved, block is generated according to own local database state of the leader, a resource waste phenomenon when block is generated in system is reduced, and delay when the block is generated is reduced.

Growing community of inventors

Shenzhen, China

Shubin Cai