Machine generated text detection method and system based on integrated multiple models

The invention discloses a machine generated text detection method and system based on multiple integrated models, and relates to the technical field of natural language processing, and the method comprises the steps: inputting an obtained to-be-detected text into a trained detection model based on multiple integrated models, and outputting a more accurate text detection result. Wherein the training of the detection model comprises the following steps: acquiring text data generated by a large language model and text data created by human beings, and constructing a training set; setting a plurality of pre-training models including a plurality of baseline models, a RoBERTa model and a DeBERTa model and training parameters of the pre-training models, and performing training by using the training set; the model is finely adjusted during training, and in the fine adjustment process, a LoRa adapter is introduced into the DeBERTa model, and a multi-scale-based forward no-label training framework is introduced into th