263 results about "Long-term memory" patented technology

The invention discloses an agent long-term memory modeling method based on a memory network, and relates to the field of agents, and the method comprises the steps: storing vector data to a vector database, and storing structured metadata to a relational database; receiving an input request of a user, and executing vector similarity retrieval in the vector database to obtain semantic similar fragments; executing structured data query in the relational database to obtain structured metadata; the memory abstract is retrieved; forming a candidate data set; taking a filtered result as memory information, inputting the memory information into a large language model through a cue word project, and generating reply content; the input request of the user and the generated reply content are combined to form a new interaction record; processing the new interaction record and historical interaction records stored in a vector database and a relational database through a large language model to generate a memory abstract; aiming at the insufficient long-term memory emotion interaction coherence of the intelligent agent, the emotion interaction coherence is improved.

The invention relates to the field of natural language understanding in an artificial intelligence technology man-machine dialogue system, in particular to a psychological accompanying dialogue method based on multi-agent collaboration and a storage medium. The psychological accompanying dialogue with individuation, high naturalness and accurate emotion recognition is provided, the user experience is effectively improved, and the psychological support requirement of the user is met. According to the main scheme, the method comprises the steps of performing multi-mode emotion evaluation on user input to obtain user emotion state description; retrieving related dialogue states and external knowledge from a shared memory module according to user questions and user emotional state description; according to the retrieved dialogue state and external knowledge, the strategy planning module generates a reply strategy to obtain a strategy group; the dialogue content generation module calls a corresponding vertical domain agent according to the generated reply strategy to generate specific reply content; and storing the user questions and model replies in the current round into the short-term memory of the shared memory module, and updating the long-term memory. The method is used for AI dialogues.

The invention relates to the technical field of AI interaction, and discloses an AI multi-modal emotion interaction memory terminal, which realizes microsecond-level synchronization of voice, facial expression and text data through a multi-thread acquisition engine, dynamically allocates each modal weight by adopting a multi-head cross attention mechanism, and adaptively adjusts modal importance based on a conversation context hidden state; when the cross-modal confidence difference exceeds a threshold value, a gating LSTM conflict resolution module is activated, and the multi-source data collaboration problem is solved; the emotional memory modeling constructs an emotional state transition topology based on a graph convolutional network, protects user privacy in combination with a differential privacy mechanism, and realizes associated event storage of millisecond backtracking of short-term memory and long-term memory. The technology integrates multi-modal dynamic perception, privacy security calculation and adaptive learning ability, significantly improves the real-time performance and personification degree of emotion interaction, and can be applied to the fields of intelligent customer service, emotion accompanying, health monitoring and the like.

The invention discloses an intelligent optimization method based on lightweight language training data, which comprises the following steps of: constructing a self-attention mechanism by using a Transform architecture, giving an embedding matrix of an input sequence, n being the sequence length, d being the embedding dimension, representing Query, Key and Value matrixes, and controlling the product magnitude by using a scaling factor; a low-rank adapter is added beside an original weight matrix, a parameter update quantity is decomposed into a product of two small matrixes, and the original weight matrix is a sum of two low-rank matrixes introduced by LoRA. According to the method, an end-to-end personalized dialogue memory system is constructed based on a MiniMind framework, dynamic internalization and accurate recall of user habit information are realized on the premise that a local vector knowledge base does not need to be constructed through organic integration of a supervision fine tuning strategy and a LoRA technology, a dialogue memory coding mechanism is designed, and the memory efficiency is improved. User interaction data is directly mapped into an implicit memory unit of a model parameter space, a lightweight memory trigger-retrieval protocol is developed, key features are extracted through dialogue context, and long-term memory response is activated.

The invention relates to the technical field of long and short term memory and natural language processing, and discloses a session processing method and device based on long and short term memory, equipment and a medium. Using the updated large language model to respond to the input verbal skill based on the stored and downloaded long-term memory data to form interactive session information, and storing the interactive session information as short-term memory data in real time; wherein the large language model is updated through federal learning according to the short-term memory data; determining a theme according to the short-term memory data, screening important memory data in the short-term memory data based on a preset priority, and incrementally storing the important memory data as long-term memory data of the corresponding theme; and releasing the long-term memory data subjected to incremental storage to the message queue middleware according to the theme, so that other service line servers subscribing to the theme download the long-term memory data. The problems of memory decline, poor memory consistency and poor integration of a session system are solved.

The invention provides a large language model long-term memory method based on human cognition inspiration, which comprises the following steps of: calculating semantic similarity distribution of historical interaction text vectors according to a sliding window, and calculating local information entropy of the window; determining an event boundary based on the local information entropy difference of the adjacent windows; taking each discrete event as the memory of the model, and performing partition management on the memory; when the event memory partition is full, the memory event with the low retention rate is transferred to a long-term memory area; when the interactive content relates to historical information, retrieving related historical events in the event memory partition by using a contextualized memory retrieval method; and the large language model generates content for the user in combination with the current context and the retrieved related historical events. According to the method, memory coding and storage, multi-level dynamic memory management and event-level situational memory retrieval methods based on the event cutting theory are adopted, and therefore the problem that an existing large language model long-term memory method lacks dynamic memory and situational memory is solved.

The invention discloses an interaction method, device and equipment based on a memory system, a storage medium and a vehicle. The method comprises the following steps: acquiring an input query statement; target memory associated with the query statement is determined through the language big model, the target memory comprises at least one of long-term memory, short-term memory and memory summary, and the memory summary is content obtained by the language big model for the long-term memory and the short-term memory summary; fusing the target memory and the query statement to obtain a cue word; and inputting the cue word into the trained language large model to obtain a reply statement output by the language large model. According to the embodiment of the invention, the accuracy of the reply statement output by the language large model can be improved.

The invention discloses a self-learning intelligent question answering system and method with long and short term memory, and the system comprises a self-learning module, a large model reply module, a short term memory module, a long term memory module, and a safety guarantee module. The self-learning module realizes real-time learning and updating, so that the system can be continuously improved according to latest user interaction; the long-short term memory module processes short-term information and long-term information respectively, so that the system can flexibly meet various different types of user requirements, and accurate and timely answers are provided. According to the technical scheme, the technical problem that an existing system lacks self-learning ability and long-short-term memory design is solved, the intelligent level and user experience of the system are remarkably improved, more intelligent and efficient services are provided for a user, and the advantages and the positive effects are beneficial to popularization and application. Therefore, the system has strong competitiveness and wide application prospect in the market.

The invention discloses a self-adaptive memory enhancement imitation learning method for a robot control task, which comprises the following steps of: performing standardization processing and time sequence feature extraction on original multi-modal data, and constructing hierarchical feature representation; establishing a layered structure including short-term memory, working memory and long-term memory, and evaluating memory importance and dynamically updating memory content based on use frequency, task correlation and novelty; retrieving related memories according to the current state and carrying out multi-modal feature fusion; a shaping reward is calculated based on neighbor sample search; and optimizing strategy network parameters through a multi-loss function combination. Through adaptive memory management and a multi-modal feature fusion mechanism, the learning efficiency and execution performance of the robot control system are improved.

The invention provides a method and a system for hierarchical audio classification. For getting high accuracy prediction with high resolution and low predictor complexity, the disclosed method uses a hierarchical classification approach with stateful prediction per frame aided by parallel AI transient detector for resetting the states of all stages at class transitions. To improve accuracy perfectly tagged database by innovative techniques of labeling are utilized. Further data augmentation is also done using signal processing techniques like audio mixing, blending of different type of data. The disclosed method applies short term audio normalization on database for normalized training and prediction of AI based Long Short-Term Memory (LSTM) networks. The disclosed method then uses a novel hierarchical classification approach with stateful LSTM prediction per frame aided by a parallel transient detector for resetting the states of all stages of hierarchical LSTM classifiers at class transitions.

The invention discloses a Multi-agent-based psychological counseling platform and method. The counseling platform comprises an evaluation agent used for dynamically extracting psychological features based on dialogue behaviors of a visitor agent and updating results to a traits library module; the planning agent is used for constructing a personalized psychological counseling scheme based on the traits library module and dynamically adjusting a counseling process and a strategy according to counseling feedback; the consultant agent is used for carrying out dialogue interaction with the visitor agent according to the personalized consultation scheme and implementing emotion pacification and psychological intervention; the visitor agent is used for expressing psychological troubles, feeding back psychological state changes and promoting the consultation process; the historical dialogue long and short-term memory module is used for storing multiple rounds of psychological counseling interaction contents and comprises historical dialogue short-term memory and historical dialogue long-term memory; and the dynamic probability memory retrieval module is used for generating probability distribution based on the current consultation context and dynamically calling memory nodes from the historical dialogue long and short term memory module or the traits library module.

The invention provides a user dialogue generation method and system based on memory fusion, and relates to the technical field of artificial intelligence, and the method comprises the steps: building a user memory knowledge base based on a preset knowledge graph tool; based on the user memory knowledge base, reading current text data input by the current dialogue user and historical text data of the current dialogue user; obtaining short-term memory data, long-term memory data and scene memory data based on the user memory knowledge base, a preset large language model, a preset cue word engineering algorithm, the current text data and the historical text data; and generating a user dialogue based on the preset large language model, the preset cue word engineering algorithm, the short-term memory data, the long-term memory data and the scene memory data. According to the invention, the memory range of the user from recent interaction to long-term important information is fully covered, and the reliability of responding to the user is improved.

The invention discloses a pre-inquiry method based on a long and short term memory management large model. The pre-inquiry method comprises the following steps: S100, acquiring identity information and complaint symptom information of a patient; s200, establishing long-term memory and short-term memory corresponding to the patient; all information generated in the pre-inquiry dialogue process is stored in the short-term memory; the long-term memory and the short-term memory are updated and maintained based on a memory storage mechanism; s300, based on the long-term memory and the short-term memory, using a large model to construct multiple rounds of pre-inquiry dialogues, and obtaining patient condition information; and S400, summarizing the patient identity information, the complaint symptom information, the long-term memory and the short-term memory to generate a pre-inquiry report. According to the pre-inquiry method based on the long-short-term memory management large model, the long-short-term memory management is separated from the large model to carry out multiple rounds of pre-inquiry dialogues, the method does not need to depend on the memory ability of the large model, and the memory loss phenomenon in multiple rounds of dialogues commonly seen in an existing large model can be effectively avoided.

The invention discloses a universal agent evaluation method, device and equipment, a medium and a product, and belongs to the technical field of agent evaluation, and the method comprises the steps: obtaining basic parameters of an agent; based on the basic parameters, the agents are evaluated in five dimensions of long-term memory accuracy, task arrangement rationality, human setting consistency, tool calling effectiveness and content compliance, so that evaluation of the agents is achieved; wherein in the process of evaluating the intelligent agent in any one of the five dimensions, a corresponding evaluation set is constructed based on the basic parameters, the large language model is adopted to score the output of the intelligent agent on the evaluation set, and the score of the intelligent agent in the dimension is obtained. According to the method, the intelligent agent can be evaluated universally and accurately.

The invention relates to the field of artificial intelligence, natural language processing and knowledge graph modeling, in particular to a digital human continuous dialogue and memory enhancement method based on a knowledge graph. Comprising the following steps: receiving a dialogue text and a historical dialogue context input by a user in the current round, and processing through a multi-semantic channel to obtain a channel semantic representation vector and a semantic vector of a dialogue in the current round; performing language multi-dimensional scoring on each node in the global knowledge graph to obtain a multi-dimensional score of the node, and forming a knowledge graph sub-graph used by the current round of dialogue; for each activated node, calculating memory intensity; and performing weighted fusion on the node embedded representation vector, generating a semantic enhancement vector fusing the current context information and the historical memory content, and generating a natural language reply conforming to the current context. The technical problems that an existing digital human dialogue lacks long-term memory and context consistency in multiple rounds of dialogues, and a knowledge graph is difficult to dynamically update are solved.

The embodiment of the invention discloses an information recommendation method and device and electronic equipment. The method comprises the following steps: acquiring to-be-reasoned information; obtaining a target memory based on the to-be-reasoned information, a memory bank and a large language model; and obtaining recommendation information corresponding to the intention based on the to-be-reasoned information, the target memory and the large language model. Through the above mode, the target memory related to the intention can be screened out from the memory library based on the to-be-reasoned information representing the user intention, the memory library and the large language model, and then the recommendation information corresponding to the intention is obtained based on the to-be-reasoned information, the target memory, the service list and the large language model. Therefore, the recommendation information better conforming to the intention and personality of the user can be obtained based on the long-term memory and / or the short-term memory related to the intention, and the accuracy of the recommendation information is improved.

This invention provides a method and system for hierarchical audio classification, aimed at achieving high accuracy in prediction with enhanced decision time resolution. The disclosed method uses a perfectly tagged database which utilizes innovative techniques of labelling. Further data augmentation is also done using signal processing techniques like audio mixing and blending of different types of data. The disclosed method applies short term audio normalization on database for normalized training and prediction using AI based Long Short-Term Memory (LSTM) networks. The method and system employ the LSTM networks in a hierarchical structure to classify audio into desired 3 or more audio classes which include at least a background noise audio class. Decision time accuracy is improved by running the LSTM predictors over time overlapped slices and by using a separate transition detection neural network.

The invention relates to a man-machine interaction method and device, computer equipment, a storage medium and a program product. The method comprises the following steps: acquiring interaction content input by a user; scene recognition is carried out on the interaction content through a large language model, and a scene label of the interaction is determined; determining a target long-term memory partition corresponding to the scene tag from a plurality of long-term memory partitions preset for the user; inputting target long-term memory content in the target long-term memory subarea and short-term memory content stored in a short-term memory area preset for the user into an intelligent agent; the intelligent agent is used for generating reply content for the interaction content according to the target long-term memory content and the short-term memory content. By adopting the method, effective association and collaborative calling of different scene memories can be realized, and the dynamic requirement of a user on coherent interaction in multi-scene switching is met.

An end-side-cloud three-in-one active chat robot system for high-emotional merchants comprises an end side, a local edge server and a cloud end, and the end side is used for collecting multi-modal data of a user and performing local lightweight real-time processing and response execution; the local edge server is used for receiving and fusing the multi-modal features and the context information from the end side, and carrying out sentiment calculation, dialogue management and active trigger decision making with medium complexity; the cloud end is used for operating a super-large-scale model and providing global knowledge management, long-term user portrait storage and model training optimization; and the end side, the local edge server and the cloud end carry out cooperative communication through an encrypted channel to form a distributed intelligent processing architecture. According to the method, global optimization is realized by integrating end-side lightweight sensing, edge multi-modal fusion and cloud long-term memory. A composite finite state machine (FSM) active questioning mechanism is combined with sentiment calculation, and is different from traditional rule type triggering. Off-line and on-line fusion scheduling and multi-agent role playing are combined to be applied to a high-emotional-quotient interaction scene, and the simulation is improved. Multi-modal emotion perception circulation is introduced, and the problem of single text emotion misjudgment is solved.

The invention discloses a rainstorm torrential flood disaster-causing element prediction method and system based on modal decomposition and deep learning, and solves the problems that a traditional model is insufficient in non-linear time sequence feature capture, and a physical model depends on complex data and is weak in generalization ability. Comprising the steps of collecting flow, flow velocity and water level data of an upstream site as input, and taking downstream disaster point data as output; preprocessing the data; a frost ice optimization algorithm is adopted to optimize variational mode decomposition parameters; a Fourier transform high and low frequency feature enhanced attention network is constructed, low-frequency and high-frequency components are divided, trend features are extracted through a fluctuation enhancement module, dynamic changes are captured through a multi-path difference calculation unit, self-perception attention is introduced to achieve feature weighted fusion and long-term memory, and a downstream hydrological state prediction result is output. By optimizing a modal decomposition and deep learning cooperation mechanism, the precision, robustness and generalization ability of sudden mountain torrent prediction are significantly improved, and the method is suitable for disaster emergency management in complex scenes of small and medium watersheds.

The invention belongs to the field of artificial intelligence and robot navigation, and discloses a visual language navigation method and system based on a cross-task incremental semantic memory graph, and the method comprises the steps that an intelligent agent executes a zero-sample visual language navigation task in a continuous environment; performing cross-modal alignment on the natural language instruction and environment observation based on a multi-modal large language model, selecting candidate waypoints and updating task progress; a semantic memory graph is constructed and dynamically updated, wherein the semantic memory graph is used for structured storage and cross-task multiplexing of scene semantic information and a spatial topological relation sensed by an intelligent agent in historical tasks; and performing global path planning and local dynamic fine tuning based on the semantic memory graph. According to the method, the problem of task-by-task forgetting in a traditional method is solved, environment understanding and task reasoning capabilities are improved by constructing structured long-term memory, and navigation precision and robustness are optimized through a global-local collaborative strategy.

The invention provides a communication protocol consistency test method and system. Key function points are extracted by analyzing a target protocol document; based on the key function points, a predefined first prompt template is applied to generate a test case prompt, a large language model is guided to generate a standardized test case corresponding to each key function point, and the first prompt template comprises k few-sample examples; dynamically retrieving a related code context from a pre-constructed long-term memory library through a retrieval enhancement generation mechanism, generating a code generation prompt by applying a predefined second prompt template based on the standardized test case and the related code context, and generating an executable test code based on the code generation prompt; and performing iterative optimization on the executable test code through the large language model, and generating a new code generation prompt based on a code execution result of the previous iteration and the retrieved related code context in each iteration. The test case can be automatically generated and dynamically adapted, and the test coverage rate, efficiency and flexibility are improved.

The invention discloses a memory enhancement-based large language model retrieval fine tuning method. The method comprises the following steps of: performing layer-by-layer dense vector representation on text sequences in a text sequence library; performing layer-by-layer alignment analysis to obtain an alignment loss value of each layer; performing layer-by-layer uniformity analysis to obtain a uniformity loss value of each layer; obtaining an optimal alignment layer and an optimal uniform layer according to the alignment loss value of each layer and the uniformity loss value of each layer; inputting the optimal alignment layer and the optimal uniform layer into a retrieval fine tuning module to obtain a final output vector; and substituting the final output vector into the cross entropy training loss function, and obtaining model parameters of the retrieval fine tuning module by minimizing the value of the cross entropy training loss function. According to the method, effective text retrieval is carried out while internal parameters of the large language model are not finely adjusted and the original generation capability of the large language model is kept, model compatibility of text retrieval and text generation is completed, and a technical basis is provided for constructing a long-term memory module of the large language model.

Systems and methods for identifying data of interest are disclosed. The system may retrieve unstructured data from an internet data source via an alert system or RSS feed. The system may input the unstructured data into various models and scoring systems to determine whether the data is of interest. The models and scoring systems may be executed in order or in parallel. For example, the system may input the unstructured data into a Naïve Bayes machine learning model, a long short-term memory (LSTM) machine learning model, a named entity recognition (NER) model, a semantic role labeling (SRL) model, a sentiment scoring algorithm, and / or a gradient boosted regression tree (GBRT) machine learning model. Based on determining that the unstructured data is of interest, a data alert may be generated and transmitted for manual review or as part of an automated decisioning process.

The invention belongs to the field of intelligent psychological counseling, and provides a multi-agent dynamic task integrated allocation method and system, and the method comprises the steps: extracting user historical semantic memory fragments through a long-term memory agent, and constructing a memory node set; obtaining current and historical emotional states of a user by an emotion analysis agent, and constructing an emotion node set; based on the semantic correlation and emotion similarity between the memory node set and the emotion node set, constructing an emotion-memory resonance map; according to the type of a resonance mode in the atlas, activating a corresponding function to drive a marshalling unit; and an intention attraction field is constructed according to the state change of the resonance spectrum in the time dimension, and is used for dynamically guiding the task to flow to the current intention node with the most active degree, and scheduling distribution is carried out on the corresponding intelligent agent in the function-driven marshalling unit based on the attraction field.

The invention discloses a hybrid deep learning sea outlet water quality prediction method based on particle swarm optimization, and the method comprises the steps: firstly obtaining time series data through data preprocessing, carrying out the correlation analysis, screening a water quality index, and reducing the data noise; then constructing a water quality prediction model by applying a mixed deep learning mode and combining the advantages of TCN, LSTM and CBAM networks; then, time sequence data is used for training a model type, TCN is used for effectively capturing a long-time dependency relationship in the sequence data, LSTM is used for enhancing the processing capacity of long-term memory, a complex time dependency mode in water quality data is captured, CBAM is used for introducing an attention mechanism to adaptively select important features, and the accuracy of water quality prediction is optimized; and the prediction precision and stability are improved. The method has higher accuracy in daily monitoring and prediction of conventional water quality indexes and early warning and prediction of water quality sudden change under special events, and can meet diversified practical application requirements.

The invention discloses a VLA model autonomous generalization method, system and device and a medium, and belongs to the technical field of robot control. The method comprises the steps of obtaining visual observation data and natural language task description of a current environment of a robot, generating an action sequence and controlling the robot to execute tasks; track data in the task execution process are collected, and token-level dense rewards and structured reflection information are output; storing the trajectory data, the token-level dense reward and the structured reflection information into a double-memory block; extracting task priori from the long-term memory by using an experience-driven generator, and generating trajectory data containing a sub-target chain; updating the VLA strategy model by using a reinforcement learning algorithm based on the token-level dense reward; and updating the VLA strategy model by using a supervised fine tuning algorithm based on the high-quality track data set. According to the method, the technical problem of autonomous generalization of the VLA model is solved.

The invention specifically discloses a brain-like reinforcement learning method and system based on hierarchical experience playback, and relates to the technical field of reinforcement learning and brain-like computing. The method comprises the steps that S1, observation data are collected and preprocessed; s2, initializing an experience buffer pool, an actor network, a commentator network and a corresponding target network, and performing parameter initialization; s3, exploring noise is initialized, actions are selected from the actor network according to the current state and executed, and obtained experience samples are stored in an experience buffer pool; s4, obtaining a new sample from the experience buffer pool, and updating the short-term memory pool; s5, determining whether partial experience in the short-term memory experience pool is transferred to the long-term memory experience pool or not by using an attention discrimination module; and S6, updating parameters of the actor network, the reviewer network and the corresponding target network. By adopting the method, the experience utilization rate of the intelligent agent is improved, the reinforcement learning performance is improved, and the method has wide application potential in multiple fields.

The embodiment of the invention provides a context information management method and device based on hierarchical memory, equipment and a medium, and the method comprises the steps: analyzing a task request to generate a sub-task node sequence containing a target description and acceptance function, and arranging a plurality of agents to cooperatively execute a task, and after the audit is passed, a complete execution context is stored in a long-term memory, and a structured abstract is generated and stored in a short-term memory, so that the problem that the context management mechanism of the existing multi-agent system is single is effectively solved, stable maintenance of cross-session cognitive continuity is realized, and the efficiency of the multi-agent system is improved. The method improves the execution reliability of a complex long-process task, optimizes the information storage and retrieval efficiency through hierarchical memory, reduces the risk of semantic deviation accumulation and amplification, clarifies the task target and acceptance standard of each stage, enhances the multi-agent cooperation consistency, and improves the efficiency of task execution. And the execution efficiency and the task completion quality of the intelligent agent system in a complex application scene are comprehensively improved.

The invention discloses a large model reasoning method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring long-term memory information required for reasoning to-be-processed information by a large model; sampling the long-term memory information based on an attention mechanism to obtain target information; and based on the target information, utilizing the large model to perform reasoning to obtain output information for the to-be-processed information.