Method and apparatus for training few-shot event detection model based on multilingual prompt learning

An apparatus for training a few-shot event detection model based on multilingual prompt learning, applied to an electronic device, a processor, a memory, an input/output interface and a communication interface in the electronic device implementing communication connection with each other in the electronic device through a bus, wherein the apparatus for training the few-shot event detection model comprises:

• a data collector configured for extracting a training data set comprising a plurality of instances from a preset network database, each of the plurality of instances comprising event text, a real trigger tag and a real event tag;
• a triggering recognition module configured for taking any instance in the training data set as an input instance, and performing event triggering recognition by applying a multilingual prompt model based on the event text of the input instance to obtain a predicted probability distribution of a trigger tag and obtain a first loss;
• a joint characterization module configured for generating a contrastive instance and a bilingual instance corresponding to the input instance according to the input instance, and performing multilingual prompt and cross-lingual encoding according to the input instance and the bilingual instance by applying the multilingual prompt model to obtain joint event characterization;
• a tag prediction module configured for performing event tag prediction on the joint event characterization by applying a two-level hierarchical prototype network model and calculating a second loss of the event tag prediction;
• a quadruple contrastive learning module for performing quaternary contrastive learning according to the input instance, the contrastive instance and the joint event characterization, and obtaining a third loss;
• and
• a model training and detection module configured for determining a total loss of a few-shot event detection model according to the first loss, the second loss and the third loss, the few-shot event detection model comprising the multilingual prompt model, the two-level hierarchical prototype network model and the quaternary contrastive learning module, and performing model training optimization on the few-shot event detection model based on the total loss to obtain the few-shot event detection model:
• inputting event text to be detected by applying the input/output interface, the processor being configured for performing language prompt learning on the event text to be detected based on a first prompt template in the multilingual prompt model to obtain a corresponding language prompt, and then performing cross-lingual encoding by applying a cross-lingual encoding model in the multilingual prompt model, which is implemented by using the encoder xlm-roberta-large, to obtain a trigger tag and event embedding:
• calculating a child tag probability distribution for which the event embedding is classified into a respective child tag in a child prototype network by applying a two-level hierarchical prototype network model, and selecting a maximum value of child tag probability as a finally recognized event tag, so as to perform few-shot event management according to the detected trigger tag and the event tag, the two-level hierarchical prototype network model is a prototype network, and the memory being configured for storing the event text to be detected, the trigger tag and the event tag, wherein the encoder xlm-roberta-large is a scaled cross lingual sentence encoder, and is pre-trained on 2.5 T of data across 100 languages data filtered from Common Crawl, and achieves state-of-the-arts results on multiple cross lingual benchmarks, and the two-level hierarchical prototype network is a hierarchical neural network and involves multiple layers where each successive laver captures increasingly abstract features from the input data.