23 results about "Teaching assistant" patented technology

The invention provides an intelligent teaching-assistant system base LLM training method for a drilling simulator, and belongs to the technical field of petroleum drilling large language model training, and the method comprises the steps: obtaining the simulation data of the drilling simulator based on an intelligent teaching-assistant system base large language model, and carrying out the field self-adaptive pre-training; performing fine adjustment on the original weight by utilizing low-rank self-adaption, and performing operation steps and fault diagnosis generation; the method comprises the following steps: constructing an enterprise private knowledge base, training a large language model by utilizing retrieval enhancement generation, and obtaining a trained verification header through explicit supervision; a direct preference optimization loss function is defined, a direct preference optimization loss function of retrieval perception is obtained in combination with the verification head, the large language model is optimized, and large language model training is completed; according to the invention, an external professional knowledge base can be fused, an equipment operation mechanism can be understood, an intelligent assistant architecture with teaching guidance and error diagnosis capabilities is provided, and the intelligent level of the drilling simulator is improved.

The invention relates to the technical field of intelligent education and Internet of Things fusion, in particular to a multi-modal interaction control method and system of a multifunctional teaching-assistant robot and the robot. According to the system, a tablet AI processor runs an Android system as a core, a display module, a man-machine interaction module, an audio input module, an audio output module, an image acquisition module, a temperature and humidity sensor module, a WIFI or Bluetooth module, an Ethernet module and an Internet of Things module are connected, and the Internet of Things module supports RS485 wired access and Zigbee wireless access at the same time; the tablet AI processor executes voice interaction, video call, face recognition, environment monitoring, network interaction and equipment linkage, and provides a face recognition alignment and depth feature matching algorithm and an annular microphone array beam forming and sound source direction estimation algorithm, so that multi-modal interaction and multi-equipment linkage in a teaching scene are realized. The convenience and the safety are improved; and the equipment access cost is reduced.

The application relates to the technical field of teaching, and particularly discloses a nursing intelligent teaching auxiliary system and method, which comprises a cloud server, a plurality of teacher terminals and a plurality of student terminals; the cloud server comprises a knowledge base construction module, which is used for acquiring teaching documents of adult nursing science, and establishing structured course knowledge graph data based on the teaching documents; a model construction module, which is used for acquiring a general large language model, and performing field adaptability training on the general model by using the course knowledge graph data, and outputting a special model fused with adult nursing knowledge; a teaching assistant module, which is used for accepting a lesson preparation request of the teacher terminal, inputting the special model, and generating lesson preparation information; and a student assistant module, which is used for receiving an auxiliary request of the student terminal, calling the special model according to the content input by the student, and generating auxiliary teaching content. The technical scheme of the application can meet the learning needs of different students and effectively improve the teaching quality.

The invention discloses an edge-end collaborative continuous learning method and system based on multi-step knowledge distillation. The method is executed on the basis of an edge-end collaborative architecture comprising an end side (a preset student model) and an edge side (a preset teaching assistant and teacher model): the end side collects a data frame, part of the data frame is transmitted to the edge side according to a dynamic frame unloading rate, and the end side and the edge side are reasoned through corresponding models respectively; the edge side monitors the difference between the inference results of the assistant model and the teacher model, and retraining is started if the difference is lower than a preset threshold value; through hierarchical knowledge distillation, the scale gap between teacher and student models is filled up, and the training and reasoning effects of the student models are effectively improved. And then the incidence relation between the reasoning precision and the GPU resources is determined through mathematical modeling, an optimal resource allocation scheme is solved by means of a dynamic programming algorithm, and the precision benefit is maximized. According to the continuous learning scheme, computing resources can be flexibly allocated in a dynamic environment, the demand of reasoning precision maximization is accurately met, and the continuous learning scheme is suitable for scenes with double requirements for resource efficiency and reasoning performance.

1. Name of the product in this design: English Teaching Assistant. 2. Purpose of this design: For English learning. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: 3D view 1.

The invention relates to the technical field of intelligent teaching assistance, and discloses a teaching assistance decision intelligent auxiliary system based on RAG, comprising: a resource index module for converging course resources and teaching materials, constructing a course knowledge base index, and maintaining a historical sample library divided according to a networking search path, a tool path and a retrieval enhancement generation path; the relevance evaluation module is used for obtaining tool relevance items through semantic matching of input questions and tool names or tool descriptions; an intelligent auxiliary decision-making module; executing the integration module; checking a combined scale block; and loading the feedback persistence module. Historical items, tool correlation items and knowledge base matching items are subjected to linear combination according to configurable weights, tool generation, RAG and networking three-path scoring, path offset and cross suppression are set, under the constraint of a difference threshold value, a comparison threshold value and a minimum confidence threshold value, only a single path is output, a scoring result is written back to a historical sample library and is subjected to periodic persistence, and the result is obtained. And online adaptive decision making is realized.

1. Name of the product in this design: Teaching Assistant Robot (K12). 2. Purpose of this design: A robot used to assist in teaching. 3. The key design elements of this product are the combination of shape, pattern, and color. 4. The image or photograph that best illustrates the design's key points: 3D view 1. 5. The design for which protection is sought includes color. 6. The bottom view of this product is for parts that are not easily seen or cannot be seen during use, so the bottom view is omitted.

The application provides an intelligent teaching assistant system, a learning path optimization method, equipment and a medium. Multi-modal sensors are used to collect multi-channel learning behavior data of students when they interact with an intelligent teaching assistant terminal for personalized learning. The sequence dependency relationship between learning content units is obtained based on a preset teaching knowledge graph. For each teaching stage, learning behavior features are extracted from the multi-channel learning behavior data, a structured feature vector is constructed based on the inherent attributes of the learning content units, and a joint representation vector is generated through multi-modal fusion. The joint representation vectors are aggregated to generate a teaching stage representation vector. The teaching effectiveness is evaluated based on the teaching stage representation vector to obtain a learning state deviation. The learning state deviation is used to filter teaching stages with substandard teaching effectiveness, and the learning path is optimized and adjusted based on the sequence dependency relationship. The application can realize closed-loop dynamic optimization of the intelligent learning path.

The application discloses a nursing education traceable, interpretable and verifiable large model intelligent teaching assistant system and method, which comprises an input capture and semantic analysis module, a multi-source knowledge retrieval and versioned traceability module, an executable reference and piece-by-piece binding module, a causal evidence graph and structured explanation generation module, an evidence consistency and counterfactual stability verification module, and a replayable decoding and evidence storage module; the input analysis module generates a standardized problem template and risk level information; the knowledge retrieval and traceability module outputs a candidate evidence set with source segment fingerprints; the information binding module binds answer units and evidence fingerprints to generate reference mapping; the causal evidence graph module constructs a structured explanation graph; the multi-dimensional verification module performs logical consistency and counterfactual stability tests and outputs verification results; and the decoding and evidence storage module stores replayable decoding information. The application realizes answer segment level traceability, structured machine verifiable explanation and deterministic result replay through the construction of a modular system.

The application discloses a student model training method, an image forgery detection method and computer equipment, and belongs to the image detection field. The image forgery detection method comprises the following steps: determining a target image to be detected; calling a trained student model based on the target image to generate a detection result; wherein the student model is trained by fusing a synthetic forgery feature, training images and intermediate features; the training images are provided by more than two domains; the intermediate features are generated by calling a corresponding teacher model based on the training images; the synthetic forgery feature is generated by calling a teaching assistant model based on all the intermediate features; and the synthetic forgery feature refers to a feature corresponding to common information in all the intermediate features. The method can improve the accuracy of identifying whether an image is a forged image.

The application provides a drilling simulator-oriented intelligent teaching assistant system base LLM training method, and belongs to the technical field of large language model training of oil drilling, and the method comprises the following steps: based on the intelligent teaching assistant system base large language model, drilling simulator simulation data is acquired, and field self-adaption pre-training is carried out; low-rank self-adaption is utilized to fine-tune the original weight, and operation steps and fault diagnosis generation are carried out; by constructing an enterprise private knowledge base, a large language model is trained by using retrieval enhancement, and a trained verification head is obtained through explicit supervision; a direct preference optimization loss function is defined, a retrieval-aware direct preference optimization loss function is obtained by combining the verification head, a large language model is optimized, and the large language model training is completed; the intelligent assistant framework capable of fusing external professional knowledge base, understanding equipment operation mechanism and having teaching guidance and error diagnosis capability is obtained, and the intelligent level of the drilling simulator is improved.

The invention discloses an implementation method, device and equipment of a teaching assistant, a medium and a product. The method comprises the following steps: receiving a calling request for a synthetic API generated by a teaching large model according to a student assistance request; calling a preset API (Application Program Interface) of virtual simulation training software and / or PID (Proportion Integration Differentiation) setting software according to the calling request; and feeding back a calling result to the teaching large model, wherein the calling result is used for indicating the teaching large model to generate assistance feedback according to the calling result. According to the embodiment of the invention, the training efficiency of virtual simulation training can be improved.

The application discloses an implementation method, device, equipment, medium and product of a teaching assistant. The method comprises the following steps: receiving a calling request of a synthetic API generated by a large teaching model according to a student assistance request; calling a preset API of a virtual simulation training software and / or a PID setting software according to the calling request; feeding back a calling result to the large teaching model, so as to instruct the large teaching model to generate an assistance feedback according to the calling result. The embodiment of the application can improve the training efficiency of virtual simulation training.

The system according to this embodiment aims to provide individually customized learning plans based on the learner's progress and level of understanding. [Solution] The system according to the embodiment comprises an analysis unit, a provision unit, a content unit, an assistant unit, and a reporting unit. The analysis unit analyzes the learner's progress and level of understanding. The provision unit provides an individually customized learning plan based on the data analyzed by the analysis unit. The content unit provides interactive content. The assistant unit provides an AI teaching assistant. The reporting unit visualizes the learning progress and provides a report.

This invention discloses an intelligent test question analysis and feedback method, storage medium, and device. By introducing visual-semantic joint segmentation, dynamic structure tree generation, cross-modal answer matching, and real-time error correction feedback mechanisms, it achieves high-precision analysis and intelligent grading of complex test questions. This invention can automatically identify and distinguish key elements such as text, formulas, and answer areas without relying on manual annotation or fixed templates, improving segmentation accuracy. It enhances the understanding of the logical hierarchy of questions by constructing a semantic structure tree through graph neural networks. It effectively supports the identification of open-ended and non-unique answers by employing an answer matching strategy that integrates embedded representation and symbolic reasoning. It generates targeted feedback by combining an error cause knowledge base and a large language model, significantly improving the system's teaching interactivity and practicality. This invention possesses good robustness, scalability, and feasibility for implementation, and is suitable for various application scenarios such as online education, intelligent homework systems, and AI teaching assistants.

A progressive tracking model compression method, device and medium based on knowledge distillation, through knowledge distillation, multi-level distillation of a teacher model, a teaching assistant model and a student model, and through a dynamic decay coefficient in the distillation, gradually weakening and then removing the layers that need to be removed, a progressive compression single-stream tracking model is realized, and a lightweight tracker FOST is obtained, the single-stream tracking model is based on a Transformer, and feature extraction and information fusion are simultaneously performed on an input template and a search picture. Through the progressive removal of the number of layers and the teaching assistant network, the feature mismatch and discontinuity problems in the depth compression process of the Transformer model are overcome, the teacher model information is effectively transmitted, the tracker proposed in the application can have high precision and high speed at the same time, and the single-stream tracker can be deployed on a CPU for the first time.

A computer-implemented method of providing artificial intelligence (AI)-driven personalized adaptive assessment question generation based on instructor-specific course material, student-specific query-response data and student-specific assessment scores, the method including receiving course material for a particular course from a knowledge database; receiving student query-response data between an individual student and an interactive natural language-based teaching assistant and associated with the particular course; receiving student scores of the individual student associated with assessments corresponding to the particular course; identifying a knowledge concept associated with a learning difficulty of the individual student based on the course material, the student query-response data, and the student scores; generating prompts including the course material and the particular knowledge concept; initiating large-language models (LLMs) to generate questions for the individual student based on the prompts; receiving the questions from the LLMs; and outputting the questions to a user device associated with the individual student.

The invention provides a software and hardware combined artificial intelligence education learning system, which relates to the technical field of digital systems, and comprises a user layer used for providing corresponding application portals for different user groups; the AI front-end application platform is in communication connection with the user layer and comprises AI exploration learning terminal software which is used for providing an AI exploration type function, an AI learning type function and a special topic type function; the AI teacher teaching assistant platform software is used for assisting the teacher in completing lesson preparation, classroom teaching and after-class learning condition analysis; the AI comprehensive accomplishment evaluation system is used for generating a comprehensive evaluation report by collecting and analyzing student data; and the AI integrated management platform is in communication connection with the AI front-end application platform and is used for realizing a system management function. The interests of students can be stimulated, and the innovation and practical ability can be cultivated; the teaching efficiency and accuracy of teachers are improved; the school competitiveness is enhanced, and course and mode innovation is promoted; future talents can be cultivated for the society, scientific and technological quality of the whole people is improved, and social progress is driven.

This invention discloses a multimodal AI-powered teaching sandbox system and method. The system collects teaching data in real time through a multimodal perception module and relies on three core innovative modules working collaboratively: first, an AI virtual student module based on a large model, capable of dynamically simulating the cognitive behavior of real students; second, an AI teaching assistant diagnostic module, capable of integrating and analyzing multi-dimensional performance in the teaching process, including knowledge delivery, skill achievement, teacher-student interaction, and student attention, generating real-time support strategies and evaluation reports; and third, a three-dimensional sandbox environment module, providing an immersive teaching scenario and recording behavioral data throughout. These three modules together construct a training cycle of "collection—diagnosis—intervention—execution—verification," enabling dynamic access to the teaching process, precise diagnosis, and personalized support. This invention, through multimodal data fusion and intelligent agent collaboration, solves the problems of poor scenario adaptability and limited interaction in traditional teacher training, significantly improving the system's adaptability, immediacy, and scalability.

The invention discloses a voice recognition and model distillation method, related equipment and a program product, and the method comprises the steps: setting at least one stage of teaching assistant model between a teacher model and a student model (the size is between the teacher model and the student model), and carrying out the step-by-step downward training through a knowledge distillation method, and training the student model by using the last level of teaching assistant model to obtain a trained student model. By gradually compressing the model, key knowledge of the teacher model can be better migrated, and a better distillation effect is obtained. In the multi-stage knowledge distillation process, voice recognition performance evaluation is regularly carried out on the currently trained model, and when it is determined that a performance evaluation result does not meet requirements, the number and size of the multi-stage teaching assisting model are dynamically adjusted, so that the knowledge distillation effect is further promoted, and the voice recognition effect of the finally trained student model is improved.

1. The name of the design product: teaching assistant robot. 2. The use of the design product: robot for interactive teaching aid. 3. The design features of the design product: in shape. 4. The picture or photo that best indicates the design features: front view.

We provide technical means to achieve learning outcomes equivalent to those of instruction from instructors or teaching assistants. [Solution] The user terminal 1 displays a learning support screen having a program description field 65 and an execution result display field 75. If an error occurs, it sends an advice request message containing the program written on the line where the error occurred and error information EM to the interactive learning support server 2. The interactive learning support server 2 searches the documents in the knowledge DB 338 for text that is strongly related to the error information EM in the advice request message, and sends a prompt to the external LLM server 8 with the searched text, the program written on the line where the error occurred, and the error information EM embedded in it.

The invention discloses a bamboo weaving process zero-basis student fast weaving training method, and aims to solve the problems of high operation error rate, long learning period, waste of teaching aid resources and difficulty in strengthening of weak links of students in existing bamboo weaving training. According to the method, through four-stage teaching (bamboo cognition and tool adaptation, basic pattern modular practical operation, three-dimensional modeling disassembly and assembly, finished product optimization and process rewinding), core components such as a standardized bamboo sample suite, an adjustable positioning type woven bottom plate and a semi-finished product pre-assembly suite are matched; and in combination with a batch cooperation mechanism (shared teaching aid time division multiplexing, trainee cross-batch demonstration and excellent trainees help teaching), basic trainees can master basic bamboo weaving finished product manufacturing within 12-14 days. The early-stage operation error rate of students can be reduced to be below 10%, the failure rate of three-dimensional finished products can be reduced to be below 15%, the utilization rate of teaching aids is improved by 60%, and popularization and inheritance of a traditional bamboo weaving process are efficiently assisted.