2results about How to "Correct answer" patented technology

The application discloses a kind of based on the continuous named entity recognition method of generative paradigm, it is related to artificial intelligence algorithm technical field.The application includes constructing generative task framework, using Instruction learning technology, according to different NER application scene and demand custom task instruction, according to general entity category and specific field requirement flexible adjustment entity option, obtain input sentence from the text data source of webpage text, document library, social media text, determine output entity format as [(entity type 1: entity mention 1), (entity type 2: entity mention 2)... ], to guide generative model to complete NER task.By using Instruction learning technology to construct generative task framework, can customize task instruction according to different NER application scene, adjust entity option, guide T5 language model to convert traditional task into generative task, enhance the flexibility and practicality of model processing different tasks, so that it can better adapt to various text scenes, improve the processing capacity to complex and various text.

This application discloses a question-answering interaction method, apparatus, computer device, and storage medium based on multi-source fusion, relating to the field of Internet technology. It can significantly improve the quality and reliability of answers to complex questions, and enhances the overall efficiency and intelligence level of question answering through parallel retrieval and intelligent fusion. The method includes: acquiring a natural language question input by a user; standardizing the natural language question to obtain structured intent information; determining multiple candidate retrieval sources based on the question characteristics of the natural language question and the interaction with the user; calling the multiple candidate retrieval sources to retrieve the structured intent information, and fusing the multiple retrieval results from the multiple candidate retrieval sources to generate a candidate answer set; summarizing, reasoning, and optimizing the candidate answer set to obtain a target answer; and providing feedback of the target answer to the user.