1584 results about "Text entry" patented technology

The invention discloses a large-model real-time voice text intention recognition method and system based on text prediction, and the method comprises the steps: obtaining the real-time voice data of a user, carrying out the real-time voice recognition processing through a streaming voice recognition interface, and obtaining a part of transcriptional text; inputting the partial transcription text into a mask language model for text prediction, and generating a plurality of high-credibility complete sentence candidates; based on the complete sentence candidates, the complete sentence candidates are input into a large language model in parallel for intention recognition, a corresponding intention result is obtained, and a mapping relation between the candidate sentences and the intention recognition result is established; and obtaining a sentence completely expressed by the user, calculating the similarity between the complete actual sentence and a plurality of high-credibility complete sentence candidates through a multi-level text similarity algorithm, selecting the candidate sentence with the highest similarity score, and directly obtaining a corresponding final intention recognition result based on the mapping relationship. The objective of the invention is to solve the technical problem of high response delay of an existing voice intention recognition system.

Provided in the embodiments of the present disclosure are a question-answering processing method, and a device, a product and a storage medium. In the question-answering processing method, after a query instruction is acquired, target knowledge information that matches the query instruction can be acquired from among a plurality of pieces of knowledge information in a knowledge base, and the query instruction and content-parsed text that corresponds to the target knowledge information are input into a large language model for question-answering processing, wherein the content-parsed text that corresponds to the target knowledge information is obtained by means of performing content parsing on a target document element that corresponds to the target knowledge information, and when the target document element comprises a document element of a non-text modality, content parsing is performed on the target document element before the target document element is input into the large language model, such that the document element of the non-text modality in the target document element can be understood by the large language model, so as to provide question-answering reference knowledge with a relatively high reliability for the large language model. Therefore, the accuracy of answering of the large language model for the query instruction can be improved.

The invention provides a financial fraud detection method based on a large language model. The method comprises the steps of obtaining a to-be-recognized text; performing word segmentation on the to-be-recognized text through the target word segmentation tool and the financial fraud dictionary, and determining a fraud sensitive word list; calculating the weight of each sensitive word in the sensitive word list according to a target algorithm to obtain a sensitive word weight feature vector; inputting the sensitive word weight feature vector and a to-be-recognized text into a large language model, and determining a context semantic vector of the sensitive word in combination with a word embedding technology; determining the similarity between the context semantic vector of the sensitive word and a preset financial fraud type semantic vector; and determining a financial fraud type according to the similarity. Through the implementation of the method, the generalization ability of the pre-training model is utilized to capture text deep semantics, priori knowledge is injected in combination with a sensitive word weight mechanism, the model is guided to focus high-risk vocabularies, the defect of a traditional method in semantic comprehension is overcome, the financial fraud recognition rate is increased, and the omission ratio is reduced.

The invention, which relates to the technical field of intelligent interaction, discloses an AI-based adaptive scene intelligent interaction system comprising a multi-modal data acquisition module, a modal preprocessing module, a multi-modal embedded coding module, an intention fusion and representation module, a service scene matching module and a service execution and reinforcement learning module. The method comprises the following steps: acquiring multi-modal original data in a user interaction process, including voice signals, text input and user behavior tracks, and synchronously recording an acquisition timestamp; according to the method, through a multi-modal unified embedding and dynamic weighting mechanism, the problem of characteristic dimension imbalance is effectively solved, and the user intention recognition accuracy is improved; meanwhile, reinforcement learning and a multi-factor scoring model are combined, personalized scene matching and dynamic response are achieved, the adaptive capacity and service accuracy of the system in a complex environment are improved, and therefore the stability and user experience of the intelligent interaction system are remarkably optimized.

In some embodiments, while a keyboard is visible in a three-dimensional environment, the computer system detects a gaze of a user move from a position away from a first key of the keyboard to the first key. In some embodiments, in response to detecting the gaze of the user moving from the position away from the first key to the first key, in accordance with a determination the one or more criteria are satisfied, the computer system initiates a process to select a first character corresponding to the first key for entry. In some embodiments, in response to detecting the gaze of the user moving from the position away from the first key to the first key, in accordance with a determination that the one or more criteria are not satisfied, the computer system forgoes initiating the process.

Methods, systems, and apparatus, including computer-readable media, for managing interactions with multiple artificial intelligence chatbots. In some implementations, a text input from a user is received. The system identifies multiple chatbots that the user is authorized to access, and the system selects a subset of the multiple chatbots based on the text input from the user. The system provides the text input from the user to each of the chatbots in the subset to generate a response to the text input from each of the chatbots in the subset. The system provides an output response to the text input from the user for presentation at the user device, where the response is based on one or more of the responses generated the chatbots in the subset.

The invention relates to a memory enhanced vision-language-action submerged space dynamic fusion automatic driving method. Comprising the following steps: generating a bird's-eye view feature map; extracting scene, agent and map marks from the aerial view feature map, and fusing the mark at the current moment and the previous i historical marks to generate memory enhanced visual marks; the memory enhanced visual mark and the vehicle state information are converted into a submerged space, text input of a driver is marked and unified into the submerged space, the submerged space is represented and fused, and fusion representation is generated; an automatic driving instruction data set is introduced for adjustment, and a large language model adapting to an automatic driving task is obtained; according to the fusion representation, track planning is carried out in an autoregression mode by using a large language model, and path point coordinates are obtained; and designing a transverse controller and a longitudinal controller based on PID (Proportion Integration Differentiation) to track and control the coordinates of the path points. According to the invention, the visual representation capability of end-to-end driving is improved, and the visual-language-action fusion effect is improved.

The invention provides a network security decision-making method and device based on a large language model. The method comprises the following steps: collecting network state data and converting the network state data into natural language description; determining a semantic feature vector corresponding to the natural language description, and retrieving target threat knowledge matched with the semantic feature vector in a knowledge base storing various basic network threat knowledge; generating an analysis prompt text in combination with the target threat knowledge and the natural language description, inputting the analysis prompt text into a large language model, sequentially executing network threat analysis steps indicated in a preset thinking chain according to the analysis prompt text, and determining a network anomaly attribution and a corresponding protection strategy; and issuing the protection strategy to each distributed execution unit deployed in the target network environment, and converting the protection strategy into an equipment configuration command and executing the equipment configuration command. A closed-loop intelligent defense process from information perception to semantic understanding to collaborative response can be realized, and the real-time performance, the accuracy and the expandability of a security decision are improved.

The invention belongs to the technical field of text processing, and particularly relates to a text processing method and system based on a large model, a terminal and a storage medium, and the method comprises the steps: preprocessing an input text, inputting the preprocessed text into a pre-established first large model, and outputting an error list or an error-free prompt; when the first large model outputs an error list, the output error list is detected, and the preprocessed text is corrected according to the detected error list; and inputting the corrected text or the text corresponding to the error-free prompt into a pre-established second large language model, and outputting a sensitive word prompt or a sensitive word-free prompt. According to the method, the sensitive word problem possibly introduced or triggered by the error correction operation is actively recognized and processed through the serial flow of error correction and sensitive word analysis. And the second model performs sensitive word analysis on the basis of the error-corrected text, so that the real-time accuracy and safety of sensitive word recognition are remarkably improved.

Disclosed are a multi-mode emotion recognition method, a system, an electronic device, and a storage medium. The method includes obtaining a spectrogram of a voice to be recognized and a corresponding text and inputting the spectrogram and the text into a multi-mode emotion recognition model to obtain an emotion recognition result output by the multi-mode emotion recognition model. The multi-mode emotion recognition model is trained based on a sample spectrogram, and a corresponding sample text, and a sample emotion recognition result, and is configured to extract a feature from the spectrogram and the text by a self-attention mechanism to obtain the voice features and the text feature, fuse the text feature and voice feature to obtain a multi-mode fusion feature, and make an emotion classification decision to obtain an emotion recognition result based on the text feature, the voice feature, and the multi-mode fusion feature.

The invention relates to the technical field of personnel examination scoring, in particular to an end-to-end-based multi-agent personnel examination scoring method, which comprises the steps of generating fusion features based on image data of an examinee answer sheet, inputting the fusion features into a heterogeneous OCR engine for processing, and outputting an examinee answer text; establishing a knowledge base, analyzing a question requirement text, and screening out a reference answer text from the knowledge base; constructing a plurality of dimension scoring agents, determining an original state based on the question requirement text and the reference answer text, training each dimension scoring agent through the original state, inputting the examinee answer text into the dimension scoring agents, outputting dimension scoring scores, and forming dimension scoring vectors; carrying out self-consistency analysis by integrating the dimension score vector and the original state, and outputting a final score; the whole-process closed-loop processing from paper answer sheet scanning to high-precision, automatic and interpretable scoring is realized, and the marking efficiency, fairness and credibility are improved.

The invention relates to a medical visual question and answer method and device based on a knowledge graph and a large language model. The method comprises the following steps: firstly, training a medical visual question and answer classification model by using a medical image, a knowledge graph and a question text; then, inputting the medical image and the question text into the medical visual question and answer classification model to obtain a predicted answer; inputting the knowledge graph and the question text into a knowledge graph retrieval enhancement generation module, and retrieving the knowledge graph to obtain a structured cue word; and integrating the predicted answer, the structured cue word and a basic prompt containing role positioning and question and answer requirements by adopting LLM to obtain a final medical answer. According to the method, the response content is richer and more professional, the diagnosis logic is better, and a medical visual question-answering system better meets the real clinical requirements.

The invention relates to the technical field of multi-modal large language model training, in particular to a prompt construction method and system for a multi-modal large language model, computer equipment and a medium. The method comprises the following steps: extracting a key frame set from an input video stream; executing a motion reconstruction process on the video stream to generate motion track information; and performing visualization processing on the motion track information to generate a track visualization graph. Performing space-time correlation coding on the key frame set and the motion track information to generate an enhanced key frame; a multi-modal prompt is constructed in a mode of integrating visual input and text input, and the multi-modal prompt is input into a preset multi-modal large language model for spatial reasoning. Through the mode, the technical problem that an existing prompting method is difficult to give consideration to the spatial reasoning precision and the calculation efficiency is solved, efficient and accurate spatial reasoning of the multi-modal large language model is achieved, and the calculation efficiency, the reasoning precision and the environmental adaptability of the model are improved.

The embodiment of the invention discloses a short text classification method based on knowledge enhancement prompt learning and a related device, and the method comprises the steps: carrying out the extension of short text information through a large language model, obtaining a rich context information text, enabling an extended information text to be the same as the short text in semantics, and enabling the extended information text to comprise the context information text; performing concept retrieval by utilizing the extended information text and a preset open knowledge graph to obtain a tag word set; based on the tag word set and a preset category tag, constructing a tag word mapper, the tag word mapper comprising a mapping relationship between the category tag and the tag word; and inputting the label word mapper and the short text into a pre-training language model, and carrying out text classification through a prompt learning method to obtain a prediction category label of the short text. Through the method, the problems of high ambiguity and feature sparsity of the short text classification task are reduced, and the short text classification accuracy is improved.

The invention discloses a real-time emotion perception and voice interaction system for an intelligent cabin. The system comprises a multi-modal data acquisition module used for synchronously acquiring a facial image, a voice signal and text input of a driver; the visual feature enhancement unit is used for carrying out restoration and emotion distribution extraction on the low-quality image; the audio noise reduction and feature extraction unit is used for extracting voice emotion features; the text emotion coding unit is used for fusing relative position coding and context semantic information; the cross-modal fusion module is used for outputting an emotion classification result by integrating visual, audio and text features; the personalized emotion database is used for storing historical emotion data of the user and performing emotion trend prediction and early warning judgment; the large language model feedback module is used for generating structured cue words according to the emotion recognition result and the driving situation and generating natural language feedback; the voice synthesis and output module is used for adjusting voice parameters and performing feedback output through a vehicle-mounted multi-channel; according to the invention, the emotion expression of human-vehicle interaction is enhanced.

The invention discloses an Agent-based education evaluation method and system, and relates to the technical field of artificial intelligence and education evaluation, and the method comprises the steps: obtaining historical evaluation data and knowledge graph labeling data of students, constructing a student knowledge state matrix, calculating a mastering probability value and a forgetting attenuation coefficient of each knowledge point, and generating an evaluation question recommendation sequence; inputting the question text and the student answering text into a multi-Agent collaborative evaluation model to generate a multi-dimensional scoring result; dynamically updating the knowledge point mastering probability value of the student in the continuous answering process, calculating the mastering degree change rate of each knowledge point node, constructing an adaptive difficulty adjustment function, and generating a question difficulty parameter and a knowledge point coverage range parameter of the next round of evaluation; and taking the question difficulty parameter and the knowledge point coverage range parameter as constraint conditions, screening a candidate question set meeting the conditions from a question bank, and generating a self-adaptive evaluation path and a capability diagnosis report. The invention also discloses a method.

Provided herein are technologies for framing and evaluating biological sequence-based analysis tasks in a unified, natural-language-based, text in and text out format. Among other things, methods and systems of the present disclosure provide machine-learning technologies for combining biological sequence data, representing, for example, DNA, RNA, and protein sequences, with natural language, conversational style prompts that set out particular analysis tasks to be performed on the biological sequence data. This approach, for example, allows complex analysis tasks, including, but not limited to, identification of various sequence modifications, genes, and regulatory elements in DNA sequences, and quantification of properties such as degradation propensity of RNA and protein stability, to be input to a machine learning model in a uniform text-based format and for output to be generated in a same, unified, text-based format.

The embodiment of the invention is suitable for the technical field of artificial intelligence, and provides a data privacy protection method, system and device for large language model application, the method is applied to client equipment, and the client equipment is deployed with an input layer and an output layer of a large language model. The method comprises the steps that text input data of the large language model is coded according to a vocabulary, a mark list represented by integers is obtained, and the vocabulary comes from a server; processing the mark list through the input layer to obtain intermediate input data; the intermediate input data is sent to the server; receiving intermediate output data returned by the server; and processing the intermediate output data based on the output layer and the vocabulary to obtain text output data. Through the method, the privacy protection of the input data can be realized while the output data is automatically obtained by using the large language model.

The invention discloses a knowledge destruction attack method and device based on an RAG system and a medium, and relates to the technical field of internet security, and the method comprises the steps: inputting a target question and an error answer into the RAG system, and generating an initial confrontation text; performing multiple rounds of iterative optimization processing on the initial adversarial text to obtain a target adversarial text; inputting the target adversarial text into a knowledge base corresponding to the RAG system; the RAG system responds to a question demand input by a user, and retrieves and outputs a question answer corresponding to the question demand from the knowledge base; and inputting the question demand and the question answer into a large language model, so that the large language model outputs a wrong answer corresponding to the target question. The method and the device are used for solving the problems of low output result precision, poor attack effectiveness and low concealment when knowledge destruction attack is carried out based on an RAG system in the prior art, and the precision of the output result is improved under the condition that the knowledge destruction attack is effectively carried out with high concealment.

A method and a system for generating human speech audio in a conversation using a trained generative AI model are provided. The method includes receiving a text input representing a portion of the conversation, receiving dialog context associated with the conversation, receiving information representing at least one voice and speaking style of at least one speaker in the conversation, generating the at least one voice and speaking style based on the received information, and generating at least one emotional audio response for the at least one speaker using the at least one voice and speaking style and without retraining the trained generative AI model.

The invention discloses a material synthesis data extraction method and system based on a knowledge enhancement large model, and relates to the related field of artificial intelligence, and the method comprises the steps: retrieving literature data, and constructing a material synthesis knowledge text by executing data denoising and OCR text conversion; performing LoRA fine tuning on the basic large model, introducing a field instruction data set to perform fine tuning learning, and determining an extraction large model; performing semantic partitioning and vectorization on the material synthesis knowledge text, constructing a multi-level retrieval framework, performing retrieval enhancement in combination with a material science knowledge base, and determining an enhanced knowledge text; and constructing a data extraction prompt, combining the enhanced knowledge text with a material to synthesize a knowledge text, inputting the knowledge text into an extraction large model, and executing knowledge extraction processing. The problem that the accuracy of data extraction is insufficient in the prior art is solved, and the effect of improving the accuracy of data extraction is achieved. Meanwhile, manual work can be replaced to complete literature analysis extraction and domain knowledge association, and support is provided for material synthesis process recommendation.

Devices and techniques are generally described for unlearning information from large language models (LLMs). In various examples, a first language model (LM) trained on a first training corpus D may be determined. First data F that is a subset of D may be determined. A first auxiliary LM may be trained using the first training corpus D and a second auxiliary LM may be trained using a second training corpus D / F, where the second training corpus D / F represents the first training corpus D without the first data F. A first text input may be determined. The first LM may be updated based at least in part on a first prediction difference between predictions the first LM and the second auxiliary LM for a first set of inputs and a second prediction difference between the predictions of the first LM and the first auxiliary LM for the first set of inputs.

The invention relates to the technical field of intelligent teaching, and discloses an English writing intelligent correcting and improving method and system based on big data analysis, and the method comprises the steps: inputting a to-be-corrected English writing text into a multi-dimensional evaluation model, and obtaining an English writing evaluation result, a real-time diagnosis report containing logic fault positioning, defect type labeling and vocabulary replacement suggestions is generated according to the English writing evaluation result; according to the real-time diagnosis report, associating high-frequency error modes in the historical writing data of the student, and generating a personalized learning path; and based on the writing task type of the English writing text to be corrected, determining a weight priority according to the writing genuine characteristics, screening a target strategy combination from the personalized learning path according to the weight priority, and generating a lifting scheme matched with the writing task type. According to the invention, the defects of a traditional correction mode are overcome, targeted improvement guidance is provided for students, and the quality and efficiency of English writing teaching and learning are improved.

The invention relates to a KV cache optimization method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: calculating a key vector and a value vector corresponding to each element in a text input sequence input into a large language model; through a multi-head potential attention mechanism, performing low-rank joint compression on the key vector and the value vector to obtain a potential vector, and storing the potential vector in a KV cache space; based on a scaling law, determining an optimal compression dimension, regenerating an adaptive potential vector and updating a KV cache space; for the same text input sequence, generating corresponding query vectors, and grouping the query vectors according to a preset grouping rule; calculating a semantic association weight between each group and the correspondingly called potential vector, and taking the semantic association weight as a group attention calculation result; in the reasoning process, potential vectors and grouping attention calculation results are calculated to calculate attention weights. By adopting the method, the storage requirement of the KV cache can be further reduced.

The invention discloses an artificial intelligence customer service system based on natural language processing, and particularly relates to the technical field of natural language processing, the artificial intelligence customer service system comprises an input processing module, an intention collaboration module, an intention optimization module and an intention dynamic decision module, the intent collaborative module is used for calculating a context entropy value based on historical N rounds of dialogue feature distribution and a time decay weight, the intent collaborative module is used for outputting an emotion intensity signal through a multi-mode emotion recognition model according to input information and generating an intent correction vector in combination with a preset intent-emotion mapping matrix, and the intent optimization module is used for constructing and optimizing a collaborative loss function so as to improve the intent correction vector. And the intention dynamic decision-making module is used for performing intention classification on the input of the user based on the optimized model parameters and performing dynamic adjustment according to context entropy and emotion analysis, and the intention dynamic decision-making module is used for adjusting and making a decision on the response strategy of the system based on the output of the input processing, intention collaboration and optimization module and according to the context of the current dialogue.

The invention discloses an AI medical comment hierarchical classification method and system based on hyperbolic space attention, and relates to the technical field of natural language processing, and the method comprises the steps: obtaining a to-be-classified original AI medical comment text, inputting the preprocessed AI medical comment text into a BERT model, and carrying out the coding and modeling of a multi-layer Transform self-attention mechanism, thereby obtaining an AI medical comment hierarchical classification model. Extracting context-sensitive semantic features; inputting the semantic features into a bidirectional LSTM network, capturing a long dependency relationship and grammatical logic in the semantic features, and extracting semantic enhancement features; based on a hyperbolic space attention module, generating attention weights according to structural distances among the semantic enhancement features in hyperbolic geometry, and calculating to obtain the semantic enhancement features subjected to attention weight weighted aggregation; and performing classification based on the weighted and aggregated features, and outputting a multi-level category label of the text. According to the method, accurate hierarchical classification and semantic aggregation of the AI medical comments can be realized.

The invention discloses a multi-modal English learning interaction system and a vocabulary memory training method, and relates to the technical field of English learning, the system comprises the following components: a data acquisition module, a data analysis module, a strategy adjustment module, a resource push module and an interaction learning module; multi-modal learning behavior data, including text input, voice reading, handwritten notes, video learning behaviors, interactive operation and the like, of learners are collected through the data acquisition module, the learners are subjected to group division by applying a group intelligent algorithm, and the behavior pattern and performance of each group in vocabulary learning are analyzed for each group, so that the learning efficiency of the learners is improved. Based on the analysis, the system can automatically adjust teaching strategies and push customized multi-modal learning resources and training methods, so that personalized requirements of different learners are met, and the learning effect and experience are remarkably improved.

A large language model-based medical examination conclusion generation method includes: obtaining a target manifestation text corresponding to a target medical examination; extracting medical examination inference knowledge that matches the target manifestation text from a medical examination inference knowledge base, where the medical examination inference knowledge includes a manifestation text and a conclusion text corresponding to a medical examination; constructing a sample based on the extracted medical examination inference knowledge, and constructing a prompt text based on the sample and the target manifestation text; and inputting the prompt text into a large language model, and outputting, by using the large language model, a target conclusion text that corresponds to the target medical examination and that is obtained by performing inference based on the target manifestation text and under guidance of the sample.

Exemplary system and methods for processing a natural language query in an edge computing system are disclosed. A processor of the computing system receives a natural language textual input as a query from a user interface and receives one or more containers of documentation over a communication channel. The processor generates a query embedding vector from the textual input. The processor extracts text from the received container and generates text chunks of specified length from the extracted data. Text embeddings are generated from the text chunks and stored in memory for a specified period. The query embeddings are compared with the text embeddings to determine relevant context information. The processor passes the relevant context information and the query through a trained neural network to generate a response. The response generated by the trained neural network is formatted and output to a user interface.

The invention relates to the technical field of image processing, in particular to a directed target detection method and device, electronic equipment and a computer storage medium, and the method comprises the steps: obtaining a target detection task containing a target detection image and a corresponding target annotation text, inputting the target detection image into a preset image classification model, and outputting a target image feature; using the feature pyramid network to construct a target multi-scale feature map based on the features, inputting the target annotation text into a preset language processing model, outputting target text features, performing feature fusion on the target multi-scale feature map and the target text features to obtain target fusion features, and inputting the target fusion features into a preset target detection model to obtain a target detection result. And outputting the position and the category of the target in the target detection task by the model. According to the invention, the accuracy and effectiveness of target detection are improved by fusing the image and text features.