12 results about "Dependency structure" patented technology

The invention provides a bid invitation document automatic generation and verification method based on a large model, and belongs to the technical field of bid invitation document intelligent processing. The method comprises the following steps: firstly, carrying out structural analysis and semantic analysis on a bid invitation file text, bid invitation standard terms and historical bid invitation examples, constructing a chapter and section structure tree and a term element table, and on the basis, establishing a formalized constraint model comprising structural constraints, logic constraints and numerical value consistency constraints; and then performing constraint compiling on the formalized constraint model to generate a constraint solving model and a constraint dependence structure. And under constraint limitation, calling the large model to generate candidate bid invitation documents. According to the method, large model generation is combined with formalized constraint verification, multi-agent confrontation verification and strategy optimization repair, so that automatic generation, accurate verification and minimum modification repair of the bid invitation document are realized, the internal inconsistency and logic error risks of the bid invitation document are effectively reduced, and the reliability and availability of bid invitation document generation are improved.

The invention relates to the technical field of natural language processing and artificial intelligence, and discloses a large-model complex instruction-oriented fine-grained intention recognition method, which comprises the following steps of: constructing a candidate analysis forest of an input instruction, and executing entity alignment of each node and a domain knowledge graph; performing semantic compatibility verification on the candidate dependency trees by utilizing entity relationships in the atlas, and screening an optimal analytic tree in combination with syntactic probability and knowledge consistency scores; if the analytic tree meeting the threshold value does not exist, a semantic conflict edge is positioned, and candidate mounting points are searched by using a map neighborhood relation to reconstruct a dependency structure; and finally, generating standardized data containing a structured analytic path based on the optimal analytic tree, and performing fine adjustment on the large model. According to the method, logic constraint and dynamic repair are carried out on the syntactic structure by introducing the knowledge graph, analysis errors caused by multiple modification or structural ambiguity in a complex instruction are effectively solved, and the accuracy and robustness of intention recognition of a large model in the vertical field are improved.

The invention relates to the technical field of multi-agent collaboration, in particular to a task execution method and system based on multi-agent collaboration, and the method comprises the steps: carrying out the semantic analysis and multi-granularity decomposition of a user task instruction, generating a subtask dependence structure, and storing the subtask dependence structure into a dynamic task pool; performing collaborative planning and rehearsal execution on the to-be-executed subtasks, generating context snapshots, executing Push, and writing the Push into a historical state stack; retrieving an evidence chain based on the domain knowledge base and extracting a fact triple set, performing consistency comparison based on structured evidence triples on the candidate reasoning paths, and outputting conflict confidence and consistency scores; when the consistency score is lower than a threshold value, triggering a projectile stack, physically clearing wrong branch intermediate data, recovering a historical context, injecting a negative feedback prompt to generate a new inference branch, and returning to continue rehearsal execution; if the verification is passed, outputting a result and entering a next subtask; therefore, the controllability and auditing performance of multi-agent cooperative task execution are improved.

The invention relates to the technical field of brand sentiment analysis, in particular to an artificial intelligence-based brand monitoring system, which comprises a brand word path extraction module, a verb target positioning module, a continuous chain construction module, a stand turning detection module and an intelligent fusion judgment module. According to the method, the dependency structure and the syntactic relationship are introduced into the semantic path, so that the semantic association between the brand word and the verb can be identified and modeled in a more targeted manner, the judgment of the verb target is guided to be more semantic hierarchy and context dependency, and the time sequence connection of brand behaviors is effectively realized; the recognition accuracy of continuous brand behavior nodes in a text is improved, dynamic tracking of emotional standing is achieved by recognizing standing turning features in sentences, and standing fusion phenomena in user emotion expression are captured through sequential change of verb and emotion combinations in a path structure and switching of pointing objects, so that the emotion expression efficiency is improved. The information insight effect of emotional expression steering and behavior continuity in the brand monitoring process can be improved.

This invention belongs to the field of computer vision technology, specifically a tree-structured video description generation method based on perceptual grammar knowledge. This invention explicitly utilizes semantic information inherent in language, using dependency structure analysis tools to transform sequential sentences into a syntax tree structure. By analyzing the connections between parent and child nodes in the tree, the dependency grammar structure within the sentence is explicitly modeled. A perceptual context attention network is used to model the contextual information generated along different paths during the generation process. Simultaneously, reinforcement learning and iterative generation training methods are introduced during the training phase to further improve model performance. Qualitative and quantitative experiments demonstrate that the model has the ability to generate more accurate and semantically richer descriptions.

This invention relates to the field of media censorship and discloses a media content compliance review system and method based on legal rules. One method involves segmenting the media content to be reviewed into nodes at the paragraph and syntactic levels, marking subject, behavior, and contextual elements with uncertain areas of context to form a set of legal elements; performing cross-segment dependency checks and multi-level matching on the legal element set to establish a dependency structure for content compliance behavior; filling in constraints on missing, ambiguous, or logically incomplete legal components in the text based on the dependency structure to generate standardized compliance reference text fragments; performing hierarchical matching and weighted comparison between the text to be reviewed and the standardized compliance reference text fragments to form a compliance score; and generating a tiered compliance review result based on the compliance score and applicable rule conditions. This invention has the advantage of improving the ability to identify compliance at the legal semantic level.

The invention discloses a complex teaching task-oriented mixed intention recognition and dynamic knowledge retrieval method, which comprises the following steps of: inputting a natural language of a user into a mixed intention recognition engine fused with a prompt template and a large language model, and executing an intention deconstruction and priority sorting algorithm by a system based on analyzed complex intention data, so as to obtain an intention recognition result; the system starts a dynamic knowledge retrieval strategy according to the sequenced structured intentions and keywords, dynamically evaluates all retrieved knowledge fragments, and sends the knowledge fragments to an adaptive cue word construction framework; and step-by-step solving is carried out, and the results of the sub-tasks are finally integrated to form a complete final answer. According to the method, a division and treatment planning thought is used, mixed intention semantics input by a user, a logic dependence structure among sub-intentions and multi-source external knowledge are innovatively interwoven into a whole, and deep mining and complementation of the advantages of the mixed intention semantics, the sub-intentions and the multi-source external knowledge are achieved through an elaborately-designed brand-new framework, complex-to-simple intention analysis and dynamic planning.

The application provides an anomaly detection data enhancement method for cloud-native micro-service system monitoring. The method can maintain the consistency of the dependency structure between multivariate indexes and anomaly semantics, effectively solve the structure damage and semantic distortion problems caused by existing enhancement methods, and significantly improve the accuracy, robustness and generalization ability of the anomaly detection model.

The present application relates to the technical field of multi-agent collaboration, and particularly relates to a task execution method and system based on multi-agent collaboration, which comprises the following steps: performing semantic analysis and multi-granularity decomposition on user task instructions, generating a subtask dependency structure and storing it in a dynamic task pool; performing collaborative planning and pre-execution on the subtasks to be executed, generating a context snapshot and performing Push writing to a history state stack; retrieving an evidence chain and extracting a set of fact triplets based on a domain knowledge base, performing consistency comparison on a candidate reasoning path based on the structured evidence triplets, and outputting a conflict confidence and a consistency score; triggering a pop stack and physically clearing intermediate data in the error branch when the consistency score is lower than a threshold, restoring the historical context and injecting negative feedback to generate a new reasoning branch and return to continue pre-execution; outputting the result and entering the next subtask when the verification is passed; thereby improving the controllability and auditability of multi-agent collaborative task execution.

This invention discloses a dataset construction method for specialized inference models, specifically relating to the field of dataset construction technology. It involves parsing domain task text information, rule constraint information, and causal dependency information to construct a dependency structure graph for the target inference task. Based on the dependency structure graph, topological path expansion and counterfactual topological path backtracking analysis are performed according to semantic relationship direction and constraint inverse relationship to form forward path topological feature data and reverse topological verification data. This establishes a forward and reverse dependency consistency matrix for topological conflict identification and outputs consistency correction results. Finally, based on the consistency correction results, topological labeling, topological equivalence determination, and multi-path admission screening are performed to obtain a topological fidelity sample set. Dependency signature annotation and topological label encoding are then performed to generate a training dataset for a specialized inference model of the target inference task. This improves the reliability of the inference path and the structuring level of the dataset for the specialized inference model.

Computer-aided data processing system, comprehensive an input unit for receiving text data, Optionally, an extraction module for generating structured facts from the text data using a machine language model, or an interface for receiving externally generated structured facts. a persistent data storage for storing the generated structured facts as state data, a processing unit with a stored directed, acyclic dependency structure with a plurality of processing nodes and a selection rule for determining a node-specific subset of the state data, where the processing unit is set up, to deterministically activate the processing nodes along the dependency structure, to automatically generate a subset of the state data as a node-specific context for each processing node by applying the selection rule, to call the machine language model with the node-specific context, to store a result generated by the machine language model as additional state data in the data storage and to make it available to subsequent processing nodes, where the order of activation of the processing nodes is completely determined by the dependency structure and is independent of outputs of the machine language model, and an independent context is created for each processing node, independent of previous model calls.

The invention discloses an operating system dependency structure cross-version evolution detection method and device, and belongs to the technical field of Web operating systems. According to the method, after metadata of a multi-version software package is preprocessed, a cross-version isomorphic dependency graph is constructed, and node source and target embedding vectors are obtained through an HOPE algorithm; and in combination with source vector function semantic similarity and neighborhood Jaccard similarity, an intermediate layer candidate packet is screened, and then evolution types such as a reconstruction type and a migration type are judged through a target vector, so that accurate detection of dependent structure cross-version evolution is realized.