42 results about "Language semantics" patented technology

The invention relates to the technical field of natural language processing, and discloses a generated text quality processing method and system based on a large language model, and the method comprises the steps: constructing a logic topological graph and a Laplacian matrix of an original generated text, and extracting a feature value sequence; recognizing a logic bearing wall based on the attention gradient and generating an anti-fact contrast text; calculating a logic collapse index by combining the difference between the map characteristic values of the original text and the contrast text and the logic polarity overturning condition; and judging the quality of the generated text according to the logic collapse index, blocking the text which does not pass the judgment, and shaping and resampling the output probability value of the error position by utilizing the spectrum difference information to generate a new text. The method does not need to depend on an external knowledge base, quantifies the stability of text logic through anti-fact interference and spectrum analysis, and identifies a high-risk illusion text; and a wrong logic path is automatically corrected through Logits shaping, so that the logic self-consistency of the generated content is improved on the premise of ensuring the semantic smoothness of the text language.

The invention provides a method and system for pre-training a sign language understanding framework based on semantic enhancement of skeletons, and relates to the technical field of sign language recognizing.The method comprises the steps that sign language video data and skeleton sequences and text data matched with the sign language video data are obtained; extracting a skeleton key point sequence from the sign language video data, modeling to form skeleton features, and performing word segmentation processing on a text; in a pre-training stage, skeleton features and word segmentation texts are input into a fusion network to generate bidirectional enhanced features, and global and local similarities are obtained through double-layer semantic alignment; calculating the comparison loss based on the similarity, and coordinating the weight through the balance parameter to obtain the hierarchical loss; executing the matching task and the language modeling task to obtain corresponding loss, weighting and combining the three types of loss into pre-training total loss, and adjusting parameters to complete training; and finally, part of parameters are optimized in combination with specific task types in a fine tuning stage, and enhanced understanding of sign language semantics is realized. The sign language understanding accuracy is improved.

The invention discloses a school history culture intelligent guide system and method based on a multilingual knowledge graph, and relates to the technical field of knowledge graphs and intelligent guide, and the system comprises a multilingual school history knowledge graph construction module which constructs a multilingual ontology model and generates the multilingual knowledge graph through cross-lingual semantic alignment; the edge collaborative perception interaction module is used for collecting interaction data and realizing low-delay response based on an edge-cloud collaborative strategy; the cross-language semantic retrieval and recommendation module is used for calculating semantic similarity and generating personalized recommendation; the knowledge-enhanced multilingual intelligent question and answer module constructs an enhancement prompt and generates a question and answer result; according to the school history culture international communication system and method, the problems of language barriers and single forms in school history culture international communication are solved, and the tour guide service quality is improved.

The invention provides a dynamic sign language time sequence modeling method and system based on an attention mechanism, belongs to the technical field of sign language recognition and man-machine interaction, and is used for solving the problems of poor sign language semantic recognition scene adaptability, insufficient personalization and inaccurate semantic feature capture in the related technology. According to the method, hand space coordinates and inertial data are synchronously collected and fused to generate a double-domain time sequence tensor, a four-dimensional fusion feature is constructed in combination with rhythm features and scene prior, semantic expression is enhanced through three-level attention weighting, scene adaptation is achieved based on hierarchical meta-parameters, and the accuracy of scene matching is improved. Personalized optimization is achieved through hand feature fingerprints and full-link parameter feedback, and precise sign language semantic modeling, scene self-adaption and individual adaptation are achieved.

Described is a system and method for training a multilingual semantic parser. A method includes receiving, by a multilingual semantic parser, a multilingual training dataset, wherein the multilingual training dataset includes pairs of utterances and meaning representations from at least one high-resource language and at least one low-resource language and wherein the multilingual training dataset is initially a machine-translated dataset, training, the multilingual semantic parser, by translating the utterances in the multilingual training dataset to a target language; and iteratively performing selecting, by an acquisition functions estimator, a subset of the multilingual training dataset for human translation, updating the multilingual training dataset with the human-translated subset of the multilingual training dataset with, and retraining, the multilingual semantic parser, with the updated multilingual training dataset.

The present application relates to a large model Lao translation method and system based on agent cooperation and cross-language semantic alignment, belonging to the field of natural language processing and machine translation. In the independent multilingual translation generation stage, the target Lao translation corresponding to the source sentence, the reference language translation and the linguistically similar language translation are generated respectively; cross-language semantic alignment and correction are carried out, the English translation is taken as the semantic alignment reference, the Thai translation provides morphological and syntactic guidance, and through the iterative information interaction between agents, the Lao translation is continuously evaluated and corrected; in the semantic clarification and strengthening stage, the key semantic fragments of the source sentence are extracted by analyzing the agent, the overall structure is simplified, and the Lao agent is assisted to generate the final high-fidelity translation result. The present application significantly improves the semantic accuracy and structural rationality of the Lao translation result; effectively solves the semantic illusion problem easily produced by the large language model in the Lao machine translation scene due to data scarcity.

The application relates to the technical field of speech recognition, and discloses a cross-modal attention fusion method and device based on voiceprint features and language semantics, which comprises the following steps: performing a feature extraction operation on speech data to obtain voiceprint features; based on a sentiment perception mask mechanism and a context sentiment memory unit, fusing sentiment features into text data to obtain semantic features; constructing a cross-modal relationship graph of the voiceprint features and the semantic features; determining a time sequence dependency relationship between the voiceprint features and the semantic features based on the edge weights generated by the cross-modal relationship graph; projecting the voiceprint features and the semantic features into the same semantic space based on the time sequence dependency relationship, and performing a feature reconstruction fusion operation based on a reconstruction loss function to obtain voiceprint and semantic fusion features; and adjusting a dialogue strategy determined based on the voiceprint and semantic fusion features; and using the adjusted dialogue strategy to generate interactive response data with historical interactive data, so that a high-accuracy and personalized speech interaction experience is realized.

The invention discloses a big language model security test method and system based on a small language cross-language attack. The method comprises the steps of constructing a small language selection strategy library, designing a cross-language semantic conversion engine, constructing a multi-dimensional language barrier, implementing a self-adaptive language switching mechanism and establishing a cross-language security evaluation system. According to the method, through dual language barriers of minority language input and Chinese output, a traditional single-language security detection system is effectively bypassed, attack content is presented in a minority language form, but a target model is forced to reply with Chinese, and a language understanding gap is established between input and output; the dynamic language switching strategy realizes intelligent cross-language attack, and the attack method can be adjusted in real time according to the defense response of the target model. According to the method, the adaptability, the concealment and the overall test effect of attacks are remarkably improved, and a systematized and efficient technical means is provided for large model security assessment in a multi-language environment.

The application discloses a building short-term load prediction method based on a lightweight language model, and relates to the field of building short-term load prediction.The application comprises the following steps: performing data preprocessing on historical multivariate time series data, dividing the processed historical multivariate time series data into multiple non-overlapping time segments, processing each time segment, outputting text embedding and numerical embedding, splicing and linearly aligning semantic embedding vectors and numerical embedding vectors, generating multivariate fusion embedding through a multivariate block feature fusion mechanism, and inputting the multivariate fusion embedding sequence into a frozen pre-trained large language model.The application effectively bridges the representation gap between the numerical time series mode and the language semantic space by converting timestamp information and statistical features into structured natural language descriptions and introducing a pre-trained language model for coding, so that the model can better understand the periodicity, trend and context dependence in the load data.

The application relates to the technical field of artificial intelligence, affective computing and computer-assisted teaching, in particular to a sign language emotion recognition and teaching feedback method based on machine learning, which collects learner sign language videos and extracts face, hand and posture holographic key points; action semantic features and emotion expression features are obtained in parallel by using a double-branch time sequence coding network; the sign language content and the emotion state containing continuous values of valence-arousal-dominance and discrete categories are obtained by a semantic recognition subnetwork and an emotion recognition subnetwork respectively; the recognition result is compared with standard semantics and emotion labels, and emotion intensity, naturalness, semantic matching degree scores and comprehensive quality scores are calculated; emotion correction instructions, reinforcement learning training sequence planning and key point trajectory visualization feedback are generated based on the scores; and teaching strategies are iteratively improved through closed-loop optimization and emotion resonance models. The application realizes synchronous recognition and quantitative evaluation of sign language semantics and emotions, and provides an adaptive feedback means for sign language emotion expression ability training.

The invention relates to the technical field of contract risk analysis based on cross-language semanteme, and discloses a contract risk analysis method and device based on cross-language semanteme, equipment and a medium. The method comprises the steps that contract terms of different languages are converted into standardized legal semantic vectors through a multi-language semantic model, according to the method, analysis deviation caused by language and culture differences is overcome, the understanding accuracy of the small language contract is improved, the knowledge graph is constructed by utilizing the law domain legal provision vectors associated with the target law domain, the timeliness of the legal basis of a matching link is ensured, and the conflict risk between contract terms and laws and regulations can be found. The method has the beneficial effects that the comprehensiveness and objectivity of risk identification are improved, the qualitative grading and quantitative evaluation of the risk are also realized, a direct risk evaluation result is provided for a decision maker, the efficiency of legal affair examination and the risk control capability are improved, and the automation, precision and real-time performance of contract risk analysis are realized.

The application provides a sign language generation method and device, electronic equipment and a storage medium. The method can obtain to-be-converted information, wherein the to-be-converted information includes text information and / or voice information. The semantic content of the to-be-converted information is determined, and the semantic content is converted into sign language semantic content with a semantic expression mode conforming to a sign language expression mode. Based on the sign language semantic content, a sign language action image is generated. In this way, the sign language action image can be automatically generated based on the to-be-converted information, and it is ensured that the hearing-impaired person can obtain effective external information.

The invention relates to the technical field of multi-language processing, and discloses an intelligence mining platform for multi-language data processing, which comprises a multi-source heterogeneous data access module for converting multi-field original data into intermediate format data; the multi-language adaptive analysis module performs language detection, semantic alignment and language analysis on text data, audio and video data and structured data in the intermediate format data to generate analysis results containing semantic vectors; a multi-format data processing module extracts key information with a cross-language semantic identifier from an analysis result; the heterogeneous data fusion module associates and fuses the key information, and constructs a knowledge graph and an event chain to form unified knowledge representation; the knowledge extraction and generation module extracts intelligence elements from the knowledge spectrogram or the event chain. Through multi-module cooperative processing, unified access, analysis, key information extraction, fusion and intelligence generation of multi-language and multi-format data are realized, and structured and high-value intelligence support is provided for related fields.

This invention relates to the field of natural language processing technology, and discloses a method and system for improving the quality of generated text based on a large language model. The method includes: constructing a logical topology graph and Laplacian matrix of the original generated text and extracting a sequence of feature values; identifying logical load-bearing walls based on attention gradients and generating counterfactual comparison text; calculating a logic collapse index by combining the differences in spectral feature values ​​and logical polarity reversals between the original and comparison texts; determining the quality of the generated text based on the logic collapse index; blocking texts that fail the determination; and using spectral difference information to reshape and resample the output probability values ​​at erroneous locations to generate new text. This invention does not rely on an external knowledge base, quantifies the stability of text logic through counterfactual interference and spectral analysis, identifies texts with high-risk illusions, and automatically corrects erroneous logical paths through Logits reshaping, improving the logical consistency of the generated content while ensuring the semantic fluency of the text.

The invention relates to the technical field of artificial intelligence, and discloses a natural language analysis-based cross-border logistics process automatic generation method and device, and the method comprises the steps: carrying out the cross-language semantic analysis and task structuring of a natural language instruction, generating structured task description including logistics operation logic, target area information and equipment interaction points; based on the structured task description, driving a large language model to perform logical reasoning and process construction, and generating an executable logistics process comprising an equipment control node and a compliance document generation node; performing protocol dynamic mapping on the equipment control node based on an equipment function vector space alignment mechanism to generate an equipment driving strategy; generating nodes for the compliance document, and performing compliance content dynamic generation processing based on the multi-modal rule base and the real-time policy information to obtain a formatted compliance document; and finally, integrating to form a logistics process execution package and deploying the logistics process execution package. According to the method, the cross-border logistics process customization efficiency, adaptability and compliance can be effectively improved.

The invention discloses a Mongolian-Chinese neural machine translation method fusing a pre-training model and pseudo data enhancement, the model adopts a shared encoder and bidirectional decoder architecture, the shared encoder is composed of a fusion encoding layer and a structure sensing layer, and the bidirectional decoders respectively correspond to Mongolian-Chinese and Chinese-Mongolian translation directions; the fusion coding layer is used for extracting semantic features by adopting MongolBERT and XLM-R-Mongolian respectively, and performing linear weighted fusion through a learnable weight so as to generate cross-language semantic representation; a lightweight syntactic information integration mechanism is introduced into the structure perception layer, syntactic distance is embedded into position coding, and the structure perception capability is enhanced in combination with a relative position attention mechanism. In the training stage, a staged multi-task learning strategy is adopted, and Mongolian-Chinese bilingual data generated by a practical pseudo-parallel corpus enhancement method is used for model optimization. Complementation is formed in the aspects of model structure, data expansion, multi-task optimization and the like, and an extensible, efficient and robust solution is provided for neural machine translation under the low-resource condition.

This invention discloses a semantic matching model and method for constructing a multilingual search system. The semantic matching model includes a multilingual adapter network, a first fully connected network classifier, a unified language semantic matching network, a second fully connected network classifier, a fusion layer, and a third fully connected network classifier. The multilingual adapter network includes a multilingual encoder and an adapter group. The unified language semantic matching network includes an English encoder and a translator. The fusion layer and the third fully connected network classifier constitute a fusion network. The method involves: establishing a relationship between a target label and a first predicted label; establishing a relationship between a target label and a second predicted label; and establishing a relationship between a target label and a final predicted category. As the semantic matching result adopted by the search system, this invention can fully explore the commonalities among multiple languages ​​while taking into account the differences between them, ensuring the accuracy of semantic search.

This application relates to the field of artificial intelligence technology and discloses a method and apparatus for automatically generating cross-border logistics processes based on natural language parsing. The method includes: performing cross-language semantic parsing and task structuring on natural language instructions to generate a structured task description including logistics operation logic, target area information, and equipment interaction points; driving a large language model to perform logical reasoning and process construction based on the structured task description to generate an executable logistics process including equipment control nodes and compliance document generation nodes; dynamically mapping protocols for equipment control nodes based on an equipment function vector space alignment mechanism to generate equipment driving strategies; dynamically generating compliance content for compliance document generation nodes based on a multimodal rule base and real-time policy information to obtain formatted compliance documents; and finally integrating these into a logistics process execution package for deployment. This application can effectively improve the efficiency, adaptability, and compliance of cross-border logistics process customization.

The invention provides an infrared point target recognition method and system based on sensing and cognition fusion. The method comprises the steps that word vectors of language description, high-level semantic features and target semantic categories containing continuous frame image samples in a data set and space-frequency characteristic data of the samples after time dimension frequency domain analysis are obtained; obtaining language high-level semantic feature codes; obtaining a visual advanced semantic feature code; obtaining the similarity between the language high-level semantic feature code and the visual high-level semantic feature code, and obtaining the distance loss between the language high-level semantic feature code and the visual high-level semantic feature code based on the similarity; performing sensory cognition fusion interactive training on the language semantic coding model and the visual semantic coding model to obtain an updated visual language model; and determining the category of a to-be-detected infrared point target according to an identification demand, and inputting the to-be-detected infrared point target and the category of the to-be-detected infrared point target into the updated visual language model to realize sensing and cognition fusion type infrared point target identification.

The invention discloses a multi-mode recognition and grading intervention system for cognitive impairment of old people, and the system comprises a voice collection module which is used for collecting original voice data generated when a user carries out cognitive assessment question and answer; the voice coding module is used for coding the clean voice sequence by using a Wav2Vec2.0 voice coder to obtain a high-dimensional voice feature vector aligned with the length of the text token; the semantic extraction module is used for transferring the clean voice sequence into a text through an ASR model, and extracting a language semantic feature vector through a BERT model; the scale retrieval module is used for forming scale feature vectors; and the feature fusion and evaluation module is used for fusing the three feature vectors of the voice, the semantics and the scale, inputting the fused feature vectors into a built-in large language model and outputting a result. According to the multi-modal recognition and hierarchical intervention system for cognitive impairment of old people provided by the invention, cross-modal deep analysis is performed on voice information by introducing a large model and a retrieval enhancement generation technology, so that the intelligence and interpretability of evaluation are improved.

This invention relates to the field of network content security technology and discloses a method for malicious semantic fusion and feature extraction for multilingual social networks, comprising the following steps: S1, multilingual corpus construction and preprocessing stage; S2, cross-language semantic encoding stage; S3, cross-layer malicious semantic fusion stage; S4, cluster feature purification stage; S5, model discrimination and iterative optimization stage. This invention overcomes the limitations of monolingual malicious content detection technology, realizes unified representation and extraction of cross-language malicious semantics in multilingual social network scenarios, effectively eliminates the surface and deep heterogeneity between different language data, can adapt to the high-frequency multilingual mixed text and low-resource language content processing needs in social scenarios, and significantly improves the discriminability and representation ability of cross-language malicious semantic features by directionally strengthening malicious semantic-related features through an adaptive feature fusion mechanism.

This invention discloses an artificial intelligence-based semantic analysis and recognition method, specifically relating to the field of semantic analysis, to address the problem of inaccurate identification and disambiguation of culture-loaded words in existing cross-language text processing, leading to frequent semantic shifts and inconsistent translation results. The method involves decomposing the text structure and parsing cultural metaphors in cross-language comparative corpora to identify culture-loaded words with specific cultural meanings. It further obtains their semantic vector sets in the source and target cultural contexts, calculates the semantic translation distance between them to generate translation paths, and then trains a contextual disambiguation model by combining the translation paths with the context. This enables the model to output the final semantic probability distribution of culture-loaded words in the target culture when the input text is given. Based on this probability distribution, semantic mapping prompts are generated during cross-language semantic conversion, thereby achieving accurate cross-cultural semantic parsing and prompting of culture-loaded words.

The invention discloses a doctor-patient communication co-situation language training scoring system and method based on multi-modal analysis, and relates to the field of artificial intelligence and medical informatization. Comprising the steps that a data collection module collects doctor-patient voice signals, and doctor voice text data and time sequence basic parameters are obtained through a data preprocessing and time alignment module; a common-situation language semantic feature extraction module analyzes the text to generate a common-situation language semantic feature vector; the emotional language semantic scoring module matches the feature vector with a template library to generate an emotional language semantic score value; an emotional language context correlation scoring module generates an emotional language correlation score value according to the patient voice emotion semantic features and the feature vectors; the speech speed analysis and speech speed scoring module generates a speech speed scoring value based on the time sequence basic parameters; the multi-dimensional score fusion module fuses the score values to generate a comprehensive common-situation communication score; and the feedback and training guide module generates an adjustment suggestion according to the comprehensive common situation communication score and each score value to form a training closed loop.

The invention provides a semantic map creating and using method and system, a robot and a medium, belongs to the technical field of artificial intelligence and robots, and aims to solve the problem that an existing semantic map creating method depends on predefined categories and cannot find unknown semantic concepts. The method comprises the following steps: collecting multi-modal data in an environment and adding spatial position information; multi-modal features are extracted; clustering the multi-modal features by adopting an unsupervised clustering algorithm to form a data cluster; generating a natural language semantic tag for the data cluster by using a pre-trained multi-modal large model; constructing a multi-dimensional semantic map; and responding to the natural language query. According to the method, through the core process of first unsupervised clustering and then large model marking, automatic discovery of unknown semantics is realized, and the environment understanding ability of the robot and the naturalness of man-machine interaction are improved.

The application relates to the technical field of computer software multilingual support, and discloses a method for solving the problem of computer software multilingual support based on machine learning. The method acquires multilingual original corpus data and language type labels, constructs a field term knowledge network, and extracts cross-language correlation. A transfer learning framework is used to generate semantic vector representation with cross-language consistency, and a pseudo-parallel corpus generation technology is used to synthesize low-resource language training data to expand the data set. A dynamic weight adjustment mechanism is used to optimize the attention allocation of a neural translation model, and reinforcement learning and dialogue state tracking are combined to improve semantic consistency in complex contexts. Real-time feedback data is used to dynamically update the term network and the semantic model, and the language support strategy is adaptively adjusted according to user interaction logs, so that localized resource data of software meeting multilingual support is finally output. The method significantly improves the processing capacity of low-resource languages and cross-language semantic consistency.

The invention relates to the technical field of distributed storage, and particularly discloses a multi-language-supported intelligent auxiliary method and system for cross-country dual-teacher lesson preparation. The system comprises a teaching resource unified access and preprocessing module, a multi-mode multi-language knowledge graph construction module, a cross-language semantic understanding and resource matching module, a real-time collaborative lesson preparation work space module and a system performance optimization and resource updating module. Intelligent matching and recommendation of teaching resources are achieved through the multi-language knowledge graph and cross-language semantic coding, real-time collaborative editing and machine translation are combined, the problems of resource isomerism, semantic barrier and low collaboration efficiency in cross-country double-teacher lesson preparation are effectively solved, and lesson preparation efficiency and cross-culture teaching adaptability are improved.

The application provides a method and system for enhancing a pre-training sign language understanding framework based on bone semantics, and relates to the technical field of sign language recognition, wherein the method comprises: obtaining sign language video data, paired bone sequences and text data thereof; extracting bone key point sequences from the sign language video data and modeling to form bone features, and performing word segmentation on the text; in a pre-training stage, inputting the bone features and the segmented text into a fusion network to generate bidirectional enhanced features, and obtaining global and local similarities through double-level semantic alignment; calculating a contrast loss based on the similarities, and obtaining a hierarchical loss by balancing parameters to coordinate weights; performing a matching task and a language modeling task to obtain corresponding losses, weighting and combining the three types of losses into a pre-training total loss, and adjusting parameters to complete training; and finally, in a fine-tuning stage, optimizing part of the parameters in combination with specific task types to realize enhanced understanding of sign language semantics. The application improves the accuracy of sign language understanding.

The application provides a cross-language text clustering method and system based on high-dimensional vector space manifold alignment, and relates to the technical field of natural language processing.The method comprises the following steps: constructing a source language and target language feature matrix through feature extraction, generating a pseudo-anchor point matrix by unsupervised topological matching, performing space centering processing, and executing orthogonal Procrustes analysis; based on the orthogonal transformation matrix, scaling factor and translation vector, performing rigid body transformation on the source language feature matrix, fusing the target language feature matrix, constructing a unified manifold space, performing clustering processing, and obtaining the semantic cluster division result of the cross-language text. Through the application, the technical problem of the prior art that the local topological structure of the source language semantic space is destroyed due to the dependence on large-scale parallel corpus for forced space mapping, which affects the cross-language text clustering accuracy and causes clustering drift can be solved, the topological coincidence of different language texts in a unified space is realized, and the cross-language text clustering accuracy is improved.

The application discloses a world model-based data transmission system and method, belonging to the field of data transmission, and through a first world model of a sending end, semantic decoupling and abstract coding are performed on input original multimedia data to generate a semantic-level state vector and a target identifier, a second world model of a receiving end retrieves and calls corresponding target identification features from a locally preset feature library, and the target identification features are taken as a rendering base to combine the semantic-level state vector and restore multimedia data containing the target object through generative reconstruction. Since the semantic-level state vector only includes one or more of environmental semantics, behavior semantics and language semantics and does not include target identification features, the amount of information sent by the sending end to the receiving end is reduced, high-quality audio and video reconstruction is still guaranteed under a low code rate, and efficient and faithful streaming media transmission is realized under limited channel bandwidth.