23 results about "Context dynamics" patented technology

The invention discloses a deep learning-driven customer service robot multi-round dialogue intention accurate recognition method, which comprises the steps of receiving a user voice signal in an outbound call scene and converting the user voice signal into text data, and obtaining a real-time reasoning resource state of a customer service robot; constructing a context dynamic association mechanism and an emotion interference quantitative model, calculating a context association degree and an emotion interference coefficient, and generating an intention confidence coefficient through a collaborative decision module in combination with a basic intention recognition probability output by the lightweight deep learning model; determining a final intention recognition result according to the intention confidence and the reasoning response time delay; model parameters and decision weights are corrected through a closed-loop feedback mechanism; according to the method, the problems of multi-round dialogue context fault, unquantized emotion interference and unbalanced instantaneity and precision of the existing outbound robot are solved, the intention recognition accuracy and instantaneity are improved, and high-concurrency outbound scene application is supported.

The invention discloses a context dynamic cutting method for a multi-agent dialogue. The method comprises the following steps: S1, input acquisition and upper and lower text field object standardization: standardizing multi-source heterogeneous data into upper and lower text field objects with state marks and storing the upper and lower text field objects; s2, context storage and clipping trigger judgment: monitoring the state of a storage area to trigger dynamic clipping; s3, context compression: compressing the completed historical context into a structured abstract; s4, context evaluation and dynamic clipping: when clipping is triggered, obtaining active and abstract contexts from a storage area to form a candidate set, and executing dynamic clipping processing of a core; s5, task routing, and then routing the task to the sub-agents based on the capability matrix and the weight; and S6, executing result aggregation by the sub-agents, and generating a final result after aggregating the output of each agent. According to the invention, the problem of unbalanced efficiency and precision in long context management is solved, and intelligent retention and precise routing of key information are realized.

The invention discloses a long dialogue context dynamic management method and system based on semantic chunk aggregation, and relates to the technical field of artificial intelligence and natural language processing, and the method comprises the steps: constructing an offline training semantic classification model; dialogue data streams are obtained online, semantic blocks are dynamically aggregated by calculating the similarity of adjacent statement vectors, and semantic compression is carried out; classifying the semantic blocks by using a classification model, generating embedded vectors and Key-Value Cache corresponding to the large language model for the semantic blocks, and constructing a context database with indexes according to categories; when a new query is processed, mixed retrieval is executed based on the category of the new query, a retrieved semantic block and a latest dialogue record are combined into a context, and after dynamic compression, a response is generated through a large language model in combination with a Key-Value Cache of the context; and finally, refluxing the new data to update the database. And high efficiency and accuracy of the system in processing the dynamic dialogue context are ensured.

The invention relates to the technical field of speech recognition, and discloses a speech recognition method and device, equipment and a storage medium. Aiming at the defects of fixed filter coefficient, static context modeling and difficulty in deep training and deployment of the existing deep feed-forward sequence memory network (DFSMN), the method comprises the following steps of: acquiring a frame-level acoustic feature sequence of input voice, and inputting an acoustic model containing a neural network unit; in the unit, the filter weight of the memory module is generated in real time through a parameter generation network based on the intermediate feature of the current frame; performing weighted aggregation on the historical and future frame features by using the weight to obtain a context enhancement feature; and outputting features based on the feature generation unit, and decoding to obtain an identification result. According to the method, context dynamic self-adaptive modeling is realized, the recognition precision and robustness are improved, the efficient reasoning characteristic of a pure feed-forward network is kept, the calculation overhead is not remarkably increased, and the method is adaptive to a server and terminal equipment and is suitable for multi-scene speech recognition of keywords, command words and the like.

The invention belongs to the crossing field of bioinformatics and computational biology, and relates to a method for identifying succinylation sites based on a language model driven by entity knowledge. According to the method, firstly, amino acid composition (AAC), One-Hot coding and residue distance-based features (DR) are fused to construct a multi-dimensional composite feature set, then an entity knowledge enhanced LUKE model and a robustness optimized RoBERTa model are respectively coupled with a bidirectional long-short term memory network (BiLSTM), and context dynamic modeling is realized by extracting deep features of a pre-training model. And a final prediction model is obtained through model fine tuning and integration strategy fusion. Finally, according to the method, t-SNE feature visualization and SHAP interpretability are combined to analyze and verify biological, physical and chemical reasonability of model decision, and accurate recognition of protein succinylation sites is achieved.

The application discloses a kind of text context dynamic reconstruction methods, comprising: to original document is step length overlap segmentation processing, generate semantic fragment and record physical offset index information;Build document context manager storage original document complete content;Perform vector retrieval and obtain the physical offset index of matching fragment;Real-time analysis and determine the physical position boundary of fragment in original document;From original document, real-time extraction predecessor and successor text block;Splice form complete logical context window.The application establishes the accurate position anchor point of semantic fragment and original document by physical offset index, realizes zero-cost backtracking original;Adopt retrieval time dynamic reconstruction, support according to the context expansion range of adaptive adjustment of retrieval confidence;Provide three-level context mode to meet different retrieval scene demand;Intelligent deduplication eliminates redundant content.Compared with traditional RAG system, retrieval accuracy is improved by about 18%, context completeness is improved by about 31%, and Token consumption is reduced by about 30%.

The application provides a text branch-based AI conversation structure graph and a context dynamic assembly method. The method allows the user to open a new conversation unit as a text branch by designating the selected content as a context anchor from any conversation unit, thereby dynamically organizing the conversation into a visual structure graph with each unit as a node. The context dynamic assembly algorithm is based on the context anchor and the graph, intelligently evaluating the relevance of each historical conversation and the current input from the dimensions of semantic association, structural proximity, logical context, etc., and calculating the comprehensive weight and sorting for the most relevant content. Finally, the algorithm intelligently balances the fidelity and abstractness of the information at each level according to the sorting and the budget of the smallest semantic unit (Token), constructs a context load with optimal information density and dynamic fidelity, and provides it to the AI, significantly improving its understanding and response ability to the user's real intention in complex multi-threaded conversations.

The invention discloses a traffic influence evaluation method and system considering context state transition, relates to the technical field of traffic influence evaluation, and solves the problems of poor adaptability and weak interactivity to traffic influence evaluation in the face of complex and changeable traffic environments, personalized evaluation requirements and high-frequency task scenes in the prior art. And thus, the accuracy of traffic influence evaluation is not high. According to the method, a multi-dimensional state vector is constructed through context information of a historical evaluation task, a Markov decision process is constructed according to the multi-dimensional state vector, the Markov decision process is trained through a Q-learning algorithm, and a traffic influence evaluation model is obtained; the traffic influence evaluation model can dynamically update the state information according to the task context, cut the action space and realize accurate strategy matching, thereby being beneficial to improving the accuracy of traffic influence evaluation.

The invention discloses a context dynamic tracking and course learning method oriented to multi-round dialogue perception. The method comprises the following steps: S1, performing utterance feature extraction in an utterance feature extraction module: extracting a semantic feature vector of each utterance in a dialogue through a first pre-training language model; the first pre-training language model is a RoBERTa model; s2, performing common sense feature extraction in a common sense feature extraction module: extracting common sense feature vectors related to emotion dynamic from the semantic feature vectors through a second pre-training knowledge generation model; the second pre-training knowledge generation model is a COMET model; s3, performing dynamic tracking modeling through a context dynamic tracker, and performing speaker information modeling through a speaker information module; and S4, course learning training: introducing a dynamic weight distribution mechanism and a weight loss function to carry out adaptive training on the dialogue sample. According to the method, dynamic contexts, multi-dimensional speaker features and common knowledge can be deeply fused, and the understanding of complex dialogue emotions is improved.

The application discloses a kind of text context dynamic reconstruction methods, comprising: to original document is step length overlap segmentation processing, generate semantic fragment and record physical offset index information;Build document context manager storage original document complete content;Perform vector retrieval and obtain the physical offset index of matching fragment;Real-time analysis and determine the physical position boundary of fragment in original document;From original document, real-time extraction predecessor and successor text block;Splice form complete logical context window.The application establishes the accurate position anchor point of semantic fragment and original document by physical offset index, realizes zero-cost backtracking original;Adopt retrieval time dynamic reconstruction, support according to the context expansion range of adaptive adjustment of retrieval confidence;Provide three-level context mode to meet different retrieval scene demand;Intelligent deduplication eliminates redundant content.Compared with traditional RAG system, retrieval accuracy is improved by about 18%, context completeness is improved by about 31%, and Token consumption is reduced by about 30%.

The invention discloses a dialogue type recommendation method and system based on multi-source collaborative enhancement, and relates to the technical field of artificial intelligence and recommendation systems, and the method comprises the following steps: S1, multi-source feedback collection and matrix construction: extracting hidden feedback data from a dialogue context and extracting dominant feedback data from an external platform, a user and project interaction matrix is constructed based on implicit feedback, a user scoring matrix is constructed based on dominant feedback, and normalization processing is performed on the dominant feedback to improve cross-user comparability; s2, multi-source collaborative weight learning: respectively taking the user and item interaction matrix and the user scoring matrix as collaborative data input, and executing EASE learning in parallel to obtain an implicit weight matrix and an explicit weight matrix to represent an item similarity structure; according to the method, multi-source collaborative enhancement of the dialogue type recommendation candidate set is realized through EASE parallel learning of correlation weights of multi-source items, density statistics-based adaptive weighted fusion and combination of dialogue context dynamic popularity adjustment.

The invention relates to the technical field of virtual reality, and discloses a stage acousto-optic interaction method and system based on virtual reality, and the method comprises the steps: firstly carrying out the multi-modal context vectorization of a real-time audio stream, performance time axis data and historical user data, so as to obtain a context vector set; meanwhile, user behavior feature extraction is carried out on the original sensor data flow of the user to obtain a user behavior vector; context attention fusion is then performed on the set of context vectors to generate a modulation parameter pair reflecting the current comprehensive context. Then, based on the modulation parameter pair, performing feature-level modulation on the user behavior vector, and decoding to generate an effect parameter vector; and finally, analyzing the effect parameter vector and converting the effect parameter vector into a rendering instruction which can be executed by a rendering engine. Therefore, the acousto-optic feedback on the user behavior can be dynamically adjusted according to the performance context, the accurate matching of the interaction effect and the stage art atmosphere is realized, and the immersion of the user and the art expressive force of the interaction experience are remarkably improved.

The application discloses an interactive robot control method and a control system, comprising: dynamically regulating a semantic understanding process based on a dialogue context dynamic model; performing weight self-adaptive optimization adjustment on a multi-modal input signal; constructing a feedback correction mechanism for cross-modal information fusion; and generating a continuous and consistent response to maintain dialogue coherence. The application has the beneficial effects that the complex changes of the dialogue context are dynamically regulated to solve the problems of semantic misunderstanding and discontinuity, and the weight distribution of the multi-modal input is optimized and adjusted to solve the problem of inaccurate information fusion.

The invention relates to the technical field of artificial intelligence, in particular to a large language model calling method and system based on a context dynamic clipping tool set. The method comprises the following steps: pre-registering tools available for a system into a tool meta-database, analyzing context information of a request, and retrieving the tool meta-database as a pre-screening query condition; integrating and retrieving pre-screened and matched tools, and constructing a dynamic candidate tool set; generating a prompt containing a user task request and a candidate tool set definition, sending the prompt to a large language model, and only selecting a tool needing to be called from the dynamic candidate tool set by the large language model; a tool calling instruction is obtained, a corresponding tool is executed, an execution result is returned to the large language model, and the large language model generates a final response to the user task request. The LLM tool selection efficiency, accuracy and safety can be improved through application of the LLM tool selection method, and the LLM tool selection method can be applied to enterprise-level intelligent office, intelligent customer service and other scenes needing LLM and multiple tool integration.

The invention discloses a software component hot upgrade method and system based on execution context dynamic switching, and relates to the technical field of thread switching. The software component hot upgrade method and system based on execution context dynamic switching comprises the following steps: S1, collecting context switching monitoring data, and executing preprocessing on the context switching monitoring data; s2, constructing a thread-level data sequence, marking switchable threads, evaluating the current switching risk of each switchable thread, and outputting a switchable window sequence; s3, comprehensively evaluating the switching safety in the switchable window, and determining a safe switching opportunity point; and S4, executing thread switching, evaluating the stability of the switching process, judging whether to release the scheduling queue freezing or not, and completing closed-loop control of the hot upgrade of the component. The problem that in the prior art, the safe switching opportunity of the thread in the I / O blocking and lock waiting state cannot be recognized, and consequently the new code segment address is illegal and the segment error occurs after the thread is switched is solved.

A computer system implements a unified framework integrating an adaptive elastic funnel (AEF) with a convergent intelligence fabric (CIF) for multi-agent AI collaboration. The system provides a multi-layer key-value subsystem for sharing computations across GPU partitions, applying hybrid placement strategies for dynamic memory management. It unifies physical GPU sub-allocation and virtual GPU time-slicing through a common abstraction layer while maintaining isolation through policy-based multi-tenancy. A predictive resource orchestration system forecasts computational needs, while speculative data scheduling proactively manages memory access patterns. The architecture includes risk-based scheduling for uncertain workloads, test-time compute scaling, and federated learning for edge collaboration. This framework delivers computational efficiency, security, and adaptive intelligence in high-dimensional environments while supporting incremental adoption through modular interfaces.

The application discloses a kind of context dynamic pruning methods for multi-agent dialogue.The method comprises the following steps: S1: input acquisition and context field object standardization, standardize multi-source heterogeneous data into context field object with state marker and store;S2: context storage and pruning trigger judgment, monitor the state of storage area to trigger dynamic pruning;S3: context compression, the completed history context is compressed into structured abstract;S4: context evaluation and dynamic pruning, when pruning triggers, obtain active and abstract context from storage area to form candidate set, and execute core dynamic pruning processing;S5: task routing, then route the task to sub-agent based on capability matrix and weight;S6: sub-agent execution and result aggregation, generate final result after aggregating each agent output.The application solves the problem of efficiency and accuracy imbalance in long context management, and realizes intelligent retention and accurate routing of key information.

Disclosed systems and methods may be provisioned with various one video memory protocol features including, as non-limiting examples, an OEM-custom display layer to support protocols for context-based single video memory map and dynamic video rendering to boost the graphics rendering process for high-performance video / gaming experience when high-resolution displays are connected, context-based single video memory map protocol to create a sharable virtual video memory map, from different memory units, which can be utilized for effective video rendering and providing optimal gaming experience, a runtime adjustable video frame driver, referred to herein as an adjustable video frame driver to dynamically access the virtual video memory map according to the context present on the object interface.

The invention discloses an intelligent call shorthand method and system, a terminal and a medium, and relates to the technical field of voice processing. According to the main technical scheme, entity features are recognized and extracted from an initial call shorthand record; constructing a context dynamic knowledge graph based on the entity features; and when it is detected that the indication feature exists in the initial call shorthand record, analyzing the indication feature according to the context dynamic knowledge graph at the current moment to obtain a target call shorthand record. And constructing a context dynamic knowledge graph by using entity features identified and extracted from the initial call shorthand record, and when detecting that indication features exist in the initial call shorthand record, analyzing the indication features according to the context dynamic knowledge graph at the current moment. In order to expect to achieve the purposes that the generated conversation shorthand record is clear in semantics, coherent in logic and convenient for a reader to understand by digesting indication features and replacing fuzzy pronouns with clear entity names.

The invention relates to the technical field of data verification, in particular to a dynamic data verification method and device based on context awareness. The method comprises the steps of receiving a data verification request; analyzing the data verification request to obtain a to-be-verified data object and verification context information corresponding to the to-be-verified data object; obtaining a verification rule matched with the to-be-verified data object by verifying the context information; and executing the verification logic on the to-be-verified data object by adopting the verification rule to obtain a corresponding data verification result, thereby decoupling the verification rule and the service code, dynamically selecting the verification rule according to the runtime context, enabling the verification strategy to flexibly adapt to different scenes, and reducing the maintenance difficulty while improving the verification flexibility.

The invention discloses a contradictory dispute real-time mediation agent method based on a multi-mode large language model. The method comprises the following steps: firstly, acquiring voice information of two parties of a contradictory dispute and converting the voice information into text information; calculating an emotional state score based on the voice information and the text information, and predicting intentions of both parties of a contradictory dispute; then extracting facts, appeals and dispute focuses of the two parties of contradictory disputes, and quantifying the conflict degree of the dispute focuses at the same time; thirdly, distributing weights for the demands of the two parties with contradictory disputes to generate an optimal compromise point, generating a mediation strategy, and carrying out dynamic optimization by utilizing reinforcement learning; then, mediation terms are generated, and the terms are updated based on the emotional state scores and feedback of the two parties of the contradictory dispute; and finally, generating a protocol hash value according to the updated mediation terms, synchronizing the protocol hash value to the block chain for evidence storage, and automatically executing the updated mediation terms. According to the method, the mediation verbal skill and strategy can be dynamically adjusted according to the real-time context, the mediation pertinence and flexibility are improved, and the method is suitable for diversified and complicated mediation scenes.

The application provides a kind of remote sensing image road segmentation method combining bidirectional multi-level road feature dynamic fusion and double context dynamic extraction, belongs to the field of road segmentation;Solve the problem of current road segmentation network exists and is difficult to distinguish road level;Including the following steps: data set production, model construction, model training and result prediction;The application designs bidirectional multi-level road feature dynamic fusion and double context dynamic extraction module.Bidirectional multi-level road feature dynamic fusion module effectively compensates the problem of spatial information and edge detail loss caused by continuous downsampling through bidirectional interaction transmission of multi-level feature information, enhances the understanding of complex scenes and multi-level road structure by the model.Double context dynamic extraction module realizes the dynamic fusion of long-distance and local multi-scale context information, which helps to suppress background interference, occlusion effect and confusion of similar road structure, and ensures the integrity and continuity of the extracted road.

The invention relates to the technical field of enterprise informatization management training, in particular to an enterprise digital learning training management platform based on AI and big data analysis, which comprises a production data interface module; the context-aware synchronization engine is used for receiving and analyzing the production context data, and dynamically identifying at least one target training theme related to the current or recent production activity according to a preset rule or model; the AI enhancement decision module is used for responding to the recognized target training theme and executing the following operations: based on the target training theme, retrieving and matching at least one associated training resource unit from the learning resource library module; and generating an enabling instruction comprising at least one training resource unit based on a predefined or dynamically calculated push strategy, and a client interaction module. According to the invention, production context can be sensed in real time, personalized learning is dynamically driven, and quantifiable verification and closed-loop optimization of the training effect at the production side are realized.