100 results about "Problem set" patented technology

The invention provides a multi-agent query enhancement generation method, device and equipment and a storage medium, and the method comprises the steps: carrying out the sub-problem decomposition processing of an original query problem through a problem rewriting agent, and obtaining a group of sub-query problem sets; performing retrieval method selection processing on each sub-query problem through the retrieval agent according to the entity type contained in each sub-query problem in the sub-query problem set to obtain a corresponding retrieval method; performing document retrieval processing on the knowledge base by using a retrieval method to obtain a group of candidate documents, and performing correlation screening processing on the candidate documents through a document selection agent to obtain related target documents; and inputting the target document and the original query question into an answer generation agent for answer generation processing to obtain an answer result. According to the method, a multi-agent cooperation framework is adopted, joint optimization of all agents is achieved through end-to-end reinforcement learning training, and the retrieval and answer quality of complex questions is effectively improved.

A framework for machine learning modeling of structured data that includes one or more artificial intelligence-based agents. These artificial intelligence-based agents are configured to create and execute chains of repeatable actions to perform user-driven and user-defined workflows with a given problem set and identified outcomes. Structured data that has been processed is fed by the artificial intelligence-based agents to language models to formulate actions operate as tools for analyzing a problem set that can be chained together to address a given workflow, in one or more prompts for constructing and delivering the identified outcomes. Chains of repeatable actions for saved and utilized for additional workflows having similar problem sets, and executed based on pre-identified triggers.

The invention relates to the fields of artificial intelligence, finance and medical treatment, and discloses an intelligent question and answer accuracy improving method and device, computer equipment and a storage medium, and the method comprises the steps: converting structured supervision fine tuning training data into a high-dimensional vector; constructing a multi-dimensional joint index for the high-dimensional vector, and storing the indexed high-dimensional vector in a distributed vector database; receiving a current dialogue text input by a user, and retrieving an associated question set from the distributed vector database through a similarity matching algorithm; dynamically splicing the retrieved associated questions, the corresponding answers and preset template instructions to generate enhanced cue words containing knowledge contexts; inputting the enhanced cue word into a large language model, and generating N candidate answers in parallel; and carrying out quantitative evaluation on the N candidate answers by adopting a multi-dimensional reward model, and selecting the answer with the highest score and returning the answer to the user. According to the scheme, the answering accuracy of the dialogue robot can be improved, so that the user experience is improved.

The invention provides a question and answer model training method and device and an intelligent question and answer method and device for a specific field, and relates to technologies such as large models, model training and deep learning in the field of artificial intelligence and the technical fields such as intelligent search and intelligent question and answer. According to the specific implementation scheme, corpus data of a specific field are obtained; meta-knowledge question and answer data are constructed based on the corpus data through a first model, and the meta-knowledge question and answer data comprise meta-questions and answers corresponding to the meta-questions; based on the meta-knowledge question-answer data, domain knowledge training is carried out on the to-be-trained question-answer model; on the basis of the meta-knowledge question and answer data, a semantic-related question set is constructed for the meta-questions, and structured reasoning training data is generated through a second model on the basis of the question set of the meta-questions; and based on the structured reasoning training data, performing reasoning ability enhancement training on the question and answer model to be trained after domain knowledge training. The question and answer performance of the model in the vertical field can be remarkably improved.

The invention provides a vertical field data construction method based on a large model, which belongs to the technical field of data processing and artificial intelligence, and comprises the following steps: converting a vertical field source document into an intermediate format text, and segmenting the intermediate format text into a plurality of text blocks; inputting the text blocks into a pre-trained generative language model, guiding the pre-trained generative language model according to pre-designed cue words to generate a plurality of candidate questions according to the content of the text blocks, and performing preliminary screening and fine screening on each candidate question to obtain a question set; pre-defining a mode of a knowledge graph according to field characteristics of the vertical field, processing all text blocks based on an information extraction model, and constructing a field knowledge graph; and performing local context retrieval on each final question in the question set based on the text block of the question source, performing global knowledge retrieval based on the domain knowledge graph, and generating a final answer and a final thinking chain. The method is suitable for different vertical fields, the data quality can be effectively improved, and the problem generation accuracy is guaranteed.

The invention relates to the technical field of artificial intelligence, and provides a domain relation extraction method and system based on a large language model. The method comprises the following steps: identifying entity information in a field text set, and obtaining a labeled entity set with type labels; on the basis of entity pairs in the labeled entity set and predefined relation types, a question set is constructed by utilizing a judgment question generation algorithm, and a domain relation judgment question set is obtained; reconstructing the structured domain data into training data through a dialogue format conversion algorithm, and training a large language model based on the training data by adopting a QLoRA quantization fine tuning algorithm to obtain a domain fine tuning model; a double-layer retrieval algorithm is applied to retrieve and obtain related information from the knowledge graph, the domain relation judgment question set and the related information are combined and input into the domain fine tuning model for reasoning, and a domain relation triple is obtained. According to the method, high-precision and interpretable domain relation extraction is realized, and the accuracy and robustness of the model in the vertical domain are improved.

The invention discloses a dynamic credibility evaluation method and device suitable for a large language model, and belongs to the technical field of artificial intelligence. According to the method, dynamic credibility evaluation is carried out on a large language model through a multi-round induction mechanism, self-adaptive termination control and a multi-dimensional evaluation system, and the method comprises a multi-dimensional reference problem set module; the multi-round dynamic induction module is used for generating a new question by adopting a '1 + n' strategy (one rewriting strategy is matched with n persuasion strategies), and entering a next round of dialogue after effectiveness screening; the self-adaptive termination control module is used for answering and scoring the model, terminating and outputting a result if a rule is not met, continuously generating a question if the rule is met, and judging that the question meets the rule if the upper limit of an attack round is not broken; and the model credibility measurement module is used for summarizing credibility scores of all dimensions to obtain a final evaluation result. According to the method, efficient, systematic and dynamic evaluation of the credibility of the large language model can be realized, expandability and model independence are realized, and a basis can be provided for optimization and supervision of the large language model.

Embodiments of the present description provide an answer generation method and apparatus, a storage medium, and an electronic device. The answer generation method comprises: acquiring an original question inputted by a user, and acquiring, in a knowledge base, a plurality of similar questions associated with the original question; determining a standard question set on the basis of the plurality of similar questions, wherein the standard question set comprises standard questions respectively corresponding to the plurality of similar questions; using a plurality of models of different structures to respectively perform sorting on the standard questions in the standard question set in respect to similarity with the original question to obtain a sorting result corresponding to the plurality of models of different structures; and on the basis of the sorting result corresponding to the plurality of models of different structures, selecting, from the standard question set, a target standard question having the highest similarity with the original question, and taking a standard answer of the target standard question as an answer of the original question.

The invention provides an automatic long text fine-tuning instruction set construction method based on a large language model, which comprises the following steps: acquiring an input text, and segmenting the input text by adopting a recursive character segmentation method to generate a paragraph set; aiming at the generated paragraph set, the large language model generates a question set and an answer set by adopting a self-guidance learning method through a preset question type set and a prompt template according to the task type; and generating an instruction set based on the generated question set and answer set, evaluating the quality of the instruction set in multiple dimensions, and optimizing the instruction set according to an evaluation result to obtain an optimized instruction set. According to the method, the high-quality long text fine tuning instruction set can be automatically generated, so that the performance of a large language model is improved, and meanwhile, the challenge of long context processing is solved. By automatically constructing the long text fine-tuning instruction set, the requirement of manual annotation is reduced, the cost is reduced, and meanwhile, the efficiency of the fine-tuning process and the long text processing capacity of the model are improved.

The invention discloses a control method and device of a customer relationship management system and a storage medium, and belongs to the technical field of management systems. The method comprises the steps of generating an association problem set according to service scene information of at least one predefined service scene, constructing graph structure data based on the association problem set, forming a domain knowledge graph, distributing priority weights for conflict problems in the domain knowledge graph, and integrating the domain knowledge graph into an intention recognition module, and in response to a target service scene triggered by a user, outputting at least one guide problem matched with the target service scene through an intention recognition module on the basis of the domain knowledge graph and the priority weight. According to weight distribution and a context association mechanism of the knowledge graph, the accuracy of a recommendation system in a complex service scene is improved.

The invention provides an orbit game system and method combining reinforcement learning and optimal control, and the system comprises an initialization problem setting module, an optimization solving module, a reinforcement learning module, a virtual environment module and a loop execution module. And solving a spacecraft orbit control quantity based on a CW equation by using an optimization solving module, and generating a final control strategy in combination with a TD3 algorithm in the reinforcement learning module. The system provided by the invention considers J2 perturbation influence in a virtual environment, continuously iterates and optimizes a control strategy, realizes an efficient and reliable orbit pursuit game under incomplete information and nonlinear disturbance, and is suitable for a multi-spacecraft autonomous decision-making task in a complex space environment.

The invention relates to the field of artificial intelligence, and provides an easily-confused criminal name prediction method and device based on a big language model, and a terminal. The method comprises the following steps: obtaining a fact description of a target case, and determining a plurality of legal provisions related to the target case; generating a crime name prediction task instruction based on the fact description and the plurality of legal provisions, and inputting the crime name prediction task instruction into the large language model to obtain a predicted crime name of the target case output by the large language model; inputting a target element question set corresponding to the predicted crime name into a large language model under the condition that the predicted crime name is determined to be the easily confused crime name, and obtaining a target element answer set of the target element question set output by the large language model; and adjusting the predicted crime name based on the target element answer set, and determining a final predicted crime name of the target case. According to the big language model-based easily-confused criminal name prediction method provided by the invention, easily-confused criminal names can be distinguished, and the accuracy, reliability and interpretability of criminal name prediction are improved.

The invention discloses a data question and answer method and device based on a large model, electronic equipment and a storage medium. The method comprises the following steps: acquiring a question input by a user and a question and answer knowledge base; wherein the question and answer knowledge base comprises a plurality of question and answer data pairs, and each question and answer data pair comprises a preset question and a corresponding answer; constructing an input prompt word according to the user input question; inputting the input prompt word and the question and answer knowledge base into a question and answer large model for processing, so that the question and answer large model recognizes at least one preset question matched with the similarity of the question input by the user from a preset question set of the question and answer knowledge base, and the at least one preset question serves as a target question set; and responding to the question input by the user according to an answer set corresponding to the target question set identified by the question and answer large model. Based on the technical scheme of the invention, the responded answer can be closer to the question, and the accuracy of question response is improved.

A system makes the abstract, step-by-step logic of math problems visible via the use of digitized mathematical reasoning blocks, which can be used to construct an interactive teaching program that allows a student to select problem sets from an index of problems, allows the student to view a brief instructional video pertaining to that skill if desired, allows the student to choose the mode of problem presentation (i.e., various learning or test modes), allows the student to work randomly generated problems from within the chosen mode and set by dragging and dropping inputs to mathematical reasoning blocks that represent the various necessary / possible sub-steps in route to the solution of that problem.

The invention discloses a problem dynamic recommendation method, device and equipment based on an HITS algorithm. The method comprises the steps of obtaining a questioned question set and a to-be-recommended question set; initializing the hub degree of each asked question; initializing the authority of each question to be recommended; for each to-be-recommended question, iteratively updating the authority of the to-be-recommended question based on the initial value of the authority of the to-be-recommended question and the latest value of the hub degree of each questioned question; for each questioned question, iteratively updating the hub degree of the questioned question based on the initial value of the hub degree of the questioned question and the latest value of the authority degree of each question to be recommended; and after iteration is finished, generating a question recommendation result according to the authority of each question to be recommended. Questioned questions are abstracted as hub pages in the HITS algorithm, to-be-recommended questions are abstracted as authoritative pages in the HITS algorithm, and dynamic recommendation of the questions is realized through iterative updating of hub degrees and authoritative degrees.

This invention belongs to the field of artificial intelligence technology. The embodiments of this invention provide a fully automated interview assistance method, system, and device based on a large language model. First, it acquires recruitment requirements, initial interview questions, and multimodal data of job seekers from the recruitment end to construct a multimodal dynamic knowledge base. This multimodal dynamic knowledge base includes content dynamic mechanisms, structure dynamic mechanisms, and iterative dynamic mechanisms, used to update the knowledge base content and sub-question sequences, update the structure of the knowledge graph in the multimodal dynamic knowledge base, and optimize sub-question generation and retrieval strategies, respectively. Then, based on the multimodal dynamic knowledge base, it determines the job seeker's structured question set, and subsequently determines the job seeker's interview assistance strategy. This invention, through three update mechanisms, makes knowledge application more closely aligned with actual interview scenarios, overcoming the shortcomings of traditional static and rigid knowledge bases.

The application relates to the technical field of retrieval enhancement generation evaluation, and provides a retrieval enhancement evaluation method and device based on multi-round dialogue context perception, which comprises the following steps: a first large language model is constructed; the first large language model is used for question rewriting according to a question and context; historical multi-round question and answer data of real users is extracted; a multi-round evaluation set is generated based on the historical multi-round question and answer data, a knowledge base document, a second large language model and a RAG system; the second large language model is used for summarizing a business classification, a question type and an interaction paradigm of each multi-round question and answer; the first large language model is used for converting the multi-round evaluation set into a single-round question set; and the RAG system is evaluated by using evaluation indexes, the multi-round evaluation set and the single-round question set. The application realizes automatic generation of the multi-round evaluation set, reduces the manual labeling cost, reduces the construction difficulty of the multi-round evaluation set, and realizes evaluation of the multi-round dialogue RAG task.

The present invention discloses a method and system for collaborative detection of static targets based on multi-agent collaboration, comprising the following steps: S1, setting the range constraints of the area to be detected, initializing each agent and the static target; S2, mathematically modeling the collaborative detection problem, and completing the problem setting; S3, setting the task utility of each agent; S4, setting the static target collaborative detection algorithm, and completing the dynamic task allocation; S5, dynamically updating the state quantities of the agent and the static target, and realizing the detection of the static target. The present invention selects the optimal task for a single agent by the utility value of the agent or maximizing the expected utility, and allows the agents within the communication range to share their own state vectors, and then realizes a bidding auction based on the state vectors and the utility of the individual agents, abstracting the optimal task allocation into the problem of solving the bidding vector in the auction, and adopting a greedy decentralized auction algorithm to solve it, realizing the optimal task allocation of multiple agents with dynamic adjustment of the number of detections and detection angles, and guiding the detection of agents.

Methods and systems for generating a frequently asked questions are provided, which include defining, by a computer program executed by a computer, a first large language model (LLM) with a user query and a feedback of the user query; refining, by the computer program, the user query to a question set based on the feedback, the question set comprising one or more sentences; defining, by the computer program, a second LLM to generate a first set of question and answer pairs from a source document; defining, by the computer program, a third LLM to generate a content set from the source document based on a rewriting of the source document; selecting, by the computer program, top questions from the content set to be provided to the second LLM; and generating, by the second LLM, a second set of question and answer pairs based on the top questions.

The present application provides a training method and device for an answer extraction model. The training method for the answer extraction model includes: determining sample texts from the original corpus, and screening at least one query question associated with the sample texts and corresponding answer labels in a pre-constructed question set; inputting any one of the query questions and the sample texts into a pre-trained answer extraction model to determine the answer extraction result of the query question; generating a target loss value for the answer extraction model based on the answer extraction result and the answer label, and optimizing the answer extraction model based on the target loss value to obtain a target answer extraction model.

The embodiment of the invention discloses a scale problem screening method and device, a storage medium and a program product. The method comprises the steps that a candidate scale set is obtained, the candidate scale set comprises a plurality of candidate scales, and each candidate scale comprises a plurality of candidate questions; constructing an exclusive classification model of each candidate scale; and based on the performance of the exclusive classification model, screening the candidate scale and the candidate questions in the candidate scale to obtain a plurality of optimal questions and construct an optimal question set. According to the method, the problems can be screened based on the performance of the scale exclusive classification model, redundant and irrelevant information is accurately removed to obtain the optimal problem set, and the important effects of saving the information collection time and improving the classification efficiency on the basis of guaranteeing the classification accuracy are achieved.

The invention provides an intelligent question answering method and device, equipment, a storage medium and a program product, and relates to the field of financial science and technology. The method comprises the following steps: performing retrieval in a preset bank field document library to obtain a retrieval result list corresponding to each query in a question set generated based on user input; wherein the retrieval result list comprises a plurality of documents sorted according to similarity; according to document sorting and similarity information of each retrieval result list, dynamically optimizing fusion parameters of a fusion sorting algorithm; fusing and reordering documents in each retrieval result list based on a fusion ordering algorithm by adopting the dynamically optimized fusion parameters to obtain a reordered target retrieval result list; and generating a target answer corresponding to the user input based on the target retrieval result list. According to the method, the intelligent question and answer effect can be improved.

The invention provides an interaction method and device, equipment, a storage medium and a product, and relates to the technical field of computers. The method comprises the steps of displaying corresponding question content information in response to a setting operation for question content of a preset interaction event, displaying corresponding creation mode configuration information in response to a configuration operation for a question creation mode, and displaying a question set in response to a question creation operation, the question set comprises a plurality of questions generated by utilizing the first model based on the creation mode configuration information, and the matching degree of each question in the plurality of questions and the question content information meets a preset requirement. By adopting the technical scheme, the problem content of the preset interaction event is set and the problem creation mode is configured, so that a batch of problems which are related to the preset interaction event and are matched with the set problem content can be quickly generated by utilizing the model, and the efficiency of configuring the intelligent object is improved.

This information processing device determines, from among a plurality of elements, an element that satisfies a constraint condition and optimizes an objective function, and comprises: a problem setting unit that selects a candidate element from among the plurality of elements on the basis of an output that can be obtained by giving an input from a user to a large language model; and a solving unit that determines, for the candidate element, an element that satisfies the constraint condition and optimizes the objective function.

The invention discloses a spaceflight control software field-oriented fine tuning data set construction method for a question and answer large model, which comprises the following steps of: firstly, converting all original spaceflight control software documents into structured JSON texts corresponding to chapters, and then carrying out keyword extraction on the structured JSON texts; based on the obtained keywords, generating a question set which is related to the keywords and has clear semantics; next, for the question set, generating a plurality of candidate answers of each question, sorting all the candidate answers of each question, and fusing the candidate answers into a coherent, comprehensive and accurate final answer; and finally, based on each question and the corresponding final answer, constructing a fine tuning data set. The method can achieve the generation of a high-quality question and answer data set in the field of aerospace control software, and supports the fine-tuning training of a large model in the field of aerospace control software.

The invention provides a problem classification method, device and equipment, a medium and a product, and relates to the field of financial science and technology or other related fields. Comprising the steps of obtaining to-be-classified questions, and extracting key information of the to-be-classified questions; according to the key information of the to-be-classified question, based on a structured knowledge graph, determining a sub-classification set to which the to-be-classified question belongs and a department set corresponding to the sub-classification set; the structured knowledge graph comprises each sub-classification set in the historical question set and a department set corresponding to each sub-classification set; calculating similarity scores of the to-be-classified questions and departments in a department set corresponding to the sub-classification set to which the to-be-classified questions belong; and determining a department corresponding to the to-be-classified question according to the similarity score. According to the scheme, the accuracy of question classification is improved.

Methods and systems for generating a frequently asked questions are provided, which include defining, by a computer program executed by a computer, a first large language model (LLM) with a user query and a feedback of the user query; refining, by the computer program, the user query to a question set based on the feedback, the question set comprising one or more sentences; defining, by the computer program, a second LLM to generate a first set of question and answer pairs from a source document; defining, by the computer program, a third LLM to generate a content set from the source document based on a rewriting of the source document; selecting, by the computer program, top questions from the content set to be provided to the second LLM; and generating, by the second LLM, a second set of question and answer pairs based on the top questions.

The invention provides a training data processing method and device, equipment, a storage medium and a product. Comprises: obtaining a question set of a target domain; for any question in the question set, a first response and a second response are generated, the first response is a response that the pre-training language model is not enhanced, and the second response is a response that the pre-training language model is combined with an external knowledge source to be enhanced; obtaining a similarity value of the first response and the second response; and generating a training data set according to the similarity value. According to the method, the human input and cost are reduced, and the training data generation efficiency is improved.

The invention provides a recommendation question determination method and device based on financial question and answer features, which can be used in the artificial intelligence field, the financial field or other fields, and the method comprises the following steps: carrying out word segmentation processing on stock questions by using a word segmentation system, determining a plurality of candidate words and word frequencies of the candidate words, and determining feature values corresponding to the candidate words; sorting the feature values, determining vector space dimensions, performing similarity calculation on each vector, and determining a similar problem set; performing entity recognition on financial preset knowledge points to obtain a domain knowledge graph, determining similarity between the knowledge points, and determining an effective question set according to a similarity calculation result between the knowledge points; and according to the questions input by the user, screening the similar question set and the effective question set to obtain a recommended effective question set. The problem of how to quickly recommend effective questions to the user in the prior art is solved, the provided questions and answers better conform to the expectation of the user, the recommendation efficiency and accuracy are improved, human resources are saved, and the user experience is improved.

An information processing apparatus for solving one or more online optimization problems, including at least one processor and a memory which is configured to store instructions, the at least one processor executing: an obtaining process of obtaining a problem setting; a generating process of generating a metaheuristic solver with reference to the problem setting; a solving process of solving a problem including one or more tasks using the metaheuristic solver so as to provide a solution to the problem including the one or more tasks; and an updating process of updating the metaheuristic solver with reference to a new task so as to provide a solution to the problem including the new task.