40 results about "Multiple response" patented technology

A method and apparatus for preference-training a language model are provided. The method according to some embodiments may include obtaining a dataset including a plurality of pieces of response data, wherein each of the plurality of pieces of response data includes multiple responses generated by the language model for a query and user preference information corresponding to each of the multiple responses, filtering out some of the plurality of pieces of response data included in the dataset using reward values for each of the plurality of pieces of response data, output from a proxy model that receives the dataset as input and training the language model using other pieces of response data that have not been filtered out.

The invention relates to the technical field of enterprise risk management and intelligent decision, and discloses an enterprise risk prediction and control method and system, and the method comprises the steps: collecting enterprise multi-source data, carrying out the preprocessing, constructing a risk prediction model of fusion modeling, and outputting a risk score; extracting an operation log behavior path based on the score, and generating a risk situation coding vector; similar events are matched in a historical event library, and the events are combined to re-calculate a response path; inputting the plurality of response paths into a strategy evolution sandbox for dynamic environment simulation, and outputting an evolution stability index; and performing path optimization through a multi-criterion decision method, updating an execution result and behavior codes to a historical mapping pool, and constructing a self-learning closed loop. The method has high interpretability, structural adaptability and strategy evolution capability, and improves the intelligent level of enterprise risk control.

The invention discloses a hybrid adaptive enhanced fine tuning method, system and device, relates to the technical field of artificial intelligence, and is particularly suitable for a deep reasoning task of a large language model. The problems of response level length deviation, problem difficulty level deviation, insufficient exploration efficiency, insufficient sample utilization and the like existing in an existing reinforcement learning algorithm are solved. The method comprises the steps of data preprocessing, model parameter and reference strategy initialization, multiple response generation, award calculation, advantage calculation and correction, model strategy updating, sampling probability adjustment and iterative optimization. Wherein all potential useful samples are ensured to be fully utilized by introducing the correction advantages of a length normalization factor and a difficulty normalization factor and combining a hybrid cutting mechanism and an adaptive sampling strategy update model. Deviation is effectively eliminated, the model reasoning ability and training efficiency are improved, the long reasoning task exploration ability is enhanced, and diversified task requirements are met.

The invention discloses a test method and device of a vehicle-mounted voice system, electronic equipment and a readable storage medium. The method comprises the steps that a received initial test instruction is responded, generalization processing is conducted on the initial test instruction, multiple target test instructions are obtained, and the multiple target test instructions are instruction variants which are the same as the initial test instruction in semantics and different from the initial test instruction in expression; inputting the plurality of target test instructions into the vehicle-mounted voice system for natural semantic processing to obtain a plurality of response results; comparing the plurality of response results with corresponding expected response results to obtain a plurality of comparison results; and based on the plurality of comparison results, determining a processing performance index of the vehicle-mounted voice system, the processing performance index being used for representing the natural voice processing capability of the vehicle-mounted voice system. According to the invention, the technical problem that the natural voice processing capability of the vehicle-mounted voice system cannot be comprehensively tested is solved.

The invention relates to a sequential test design method based on multi-response modeling considering response correlation. The method comprises the steps of constructing a response model capable of simultaneously predicting multiple responses based on a multi-response Bayesian support vector machine model, performing iterative sampling by combining a local development criterion based on expanded nearest neighbor gradient projection and a global exploration criterion based on a variance matrix diagonal product, and continuously updating the response model, and until the sequential test design stop criterion is met. By adopting the method, the mapping relationship between a plurality of factors and a plurality of responses in a complex application system can be constructed at the same time, the correlation between the responses is represented, available information is increased through information transmission between the responses, the modeling precision of the response model is improved, and the modeling efficiency is improved. And thus, the effectiveness of sequential test design of a complex application system with multiple responses is improved.

A method includes obtaining training data for a retrieval-augmented generation (RAG) architecture having retriever and generative models. The retriever model is configured to identify information chunks relevant to input queries, and the generative model is configured to generate outputs based on the information chunks and the input queries. The method also includes generating a prompt for the generative model and generating multiple sets of queries for the retriever model. Each query in the multiple sets of queries is configured to cause the retriever model to select a set of information chunks associated with the prompt. The method further includes generating multiple responses to the prompt using the generative model and the sets of information chunks and determining rewards associated with the RAG architecture based on the responses. In addition, the method includes training the generative model based on the training data and the rewards to produce an updated RAG architecture.

The invention provides an intention-based task cooperation method and device, relates to the technical field of computers, and mainly aims to solve the problem that the interaction effect is affected due to the fact that complex or fuzzy instructions of a user cannot be fed back during intention-based task cooperation in the prior art. According to the main technical scheme, a user intention is determined based on input information of a user; the user intention is analyzed, a plurality of response tasks with an execution sequence are obtained, and different response tasks have different operation categories; an intelligent agent corresponding to the operation category is used for executing each response task to obtain response information, and the intelligent agent is used for querying the response information matched with the response task in a preset database; and processing the response information according to the execution sequence to obtain a response result corresponding to the user intention, and feeding back the response result to the user through the browser page.

A computer-implemented method for determining an adaptive threshold for classifying a response signal to a signal emitted by a radar unit as a single scatterer response or a non-single scatterer response. The method includes obtaining multiple unit-independent thresholds based on multiple response signals with different signal-to-noise ratios. The method includes obtaining multiple unit-dependent thresholds based on multiple signals emitted by the radar unit with multiple configurations of azimuth and elevation angles. The method includes determining, based on the unit-independent thresholds and the unit-dependent thresholds, the adaptive threshold.

The application discloses a kind of microservice performance anomaly detection methods based on Transformer, comprising: obtaining the span log of microservice system, the span with same tracking ID in span log is combined together, to obtain multiple different call chains, using depth-first search algorithm searches all paths of the call chain from root node to leaf node, to obtain multiple service call sequences in the call chain and its corresponding multiple response time sequences, each service call sequence is input into the detection model trained in advance, to obtain the probability distribution of each feature value in the path vector set corresponding to the service call sequence, according to the probability distribution of service feature value output by detection model, the probability of service actual feature value in path vector is obtained as the abnormal score of service, according to the abnormal score of service, whether service is the service that appears performance anomaly in call chain is judged.The application can solve the technical problems of poor convergence speed and detection effect of existing method model.

We provide the system. [Solution] A means for generating multiple response scenarios using generative artificial intelligence, A means for analyzing response data received from users and creating appropriate responses using generative artificial intelligence, A means for conducting training sessions for users and evaluating the results of those sessions, A means of providing feedback to users based on the evaluation results, A system that includes means for retaining and managing training history and evaluation information.

The embodiment of the invention relates to a method, device and equipment for training a generative model and a computer readable storage medium. The method provided by the invention comprises the following steps: constructing a training prompt item; and executing multiple rounds of iterative training based on the training prompt item, wherein each round of iterative training comprises: obtaining multiple response contents generated by the generative model based on the training prompt item; based on the evaluation information of the multiple response contents, first response contents and second response contents are determined from the multiple response contents, and the evaluation of the first response contents is superior to that of the second response contents; and adjusting parameters of the generative model to increase a first probability of outputting the first response content and reduce a second probability of outputting the second response content.

Concepts and technologies are disclosed herein for an infringement detection system that obtains images of products and images of proprietary objects, and analyzes the images to make coarse matches. An image comparison engine may transform or augment the data for comparison for coarse and refined matching. The outputs of the image comparison engine are initial infringement predictions that are further evaluated using refined matching including shape fitting. The detection system outputs refined infringement predictions, which may be optionally confirmed as counterfeit based on various considerations (e.g., known inauthentic products, suspect sales history, text and image anomalies, etc.). Upon a refined match or confirmation, the detection system records potential infringements and associated metadata into a database to initiate an optional response action. Multiple response actions are possible for transmission over a computer network to one or more receiving electronic addresses, computer servers, network system gateways or mobile / fixed devices.

The application discloses a kind of question and answer recommendation method, system, equipment, storage medium and program product, the method includes: real-time acquisition first information, first information contains first inquiry information and multiple second inquiry information, the inquiry time corresponding to second inquiry information is earlier than the inquiry time corresponding to first inquiry information;First information is carried out semantic analysis, determine first demand feature, first demand feature is used to describe first information in from at least one of the following characterization dimensions: first characterization dimension, second characterization dimension and third characterization dimension;First demand feature is searched in first vector knowledge base based on, determine multiple response result data for first inquiry information;Wherein, the similarity between response result data and first demand feature is greater than similarity threshold, and first vector knowledge base is formed by the response vector data corresponding to the response data of multiple different modalities.The present scheme can realize automatic and fast accurate question and answer recommendation.

Embodiments of this disclosure relate to a method, apparatus, device, and storage medium for information processing. The method proposed herein includes: acquiring an input message; presenting, in an interactive interface, multiple response contents generated by multiple virtual objects based on the input message; and providing evaluation information for the multiple response contents. In this manner, embodiments of this disclosure can provide a user with the output results of multiple virtual objects (e.g., intelligent agents or machine programs) for the same request, thereby improving the user's information acquisition efficiency. Furthermore, by providing evaluation information for the response contents, embodiments of this disclosure can also help users better perceive the differences between the output results of different virtual objects.

This invention belongs to the field of tunnel detection technology and provides a method and system for identifying tunnel structural defects based on multispectral sensing and SLAM. Based on multiple response timestamp data, a residual squared sum objective function is constructed to represent the difference between the theoretically predicted clock vector and the actual measured timestamp matrix. Through iterative optimization, the optimal smooth estimate of the clock drift rate is determined, reducing the combined effects of electromagnetic interference and temperature / humidity variations. A three-dimensional point cloud map is constructed based on the coupling of axial integrals of inertial data and radial constraints of laser data, achieving long-distance, high-precision passive positioning under tunnel corridor effects. Furthermore, a lightweight identification model using a multidimensional knowledge distillation architecture and geometrically guided fraction fusion reduces edge memory consumption. By combining the reduction of electromagnetic interference and temperature / humidity variations, overcoming positioning drift under the dual constraints of corridor effects, and breaking through edge computing power limitations, the tunnel defect identification effect is improved.

This invention relates to a method for real-time online detection of wastewater indicators. Before detection, a low-concentration water body is injected into the detection chamber to perform micro-dose detection. By observing whether there is a nonlinear shift in the initial response behavior, it is determined whether there is residual response on the inner wall of the detection chamber. If so, a local delayed detection and pre-rinsing operation is performed to suppress interference responses caused by the chemical memory of the detection chamber. After completing the pre-response suppression, a slow-varying physical stimulus process is applied to the wastewater to be tested, stimulating the release of masked pollutants in the wastewater. The pollutant response signals are continuously tracked, and fluctuating behavior is identified. Based on the changing trend of the response curve, a stable response interval is identified as the sampling point for final detection. For cases with superimposed responses from multiple pollutants, the rise rate, fall rate, and tailing characteristics of the response profiles of each indicator are analyzed to identify the physical differences of multiple response signals and perform physical-level decoupling, improving the accuracy of identifying multiple pollutants.

This application is applicable to the field of artificial intelligence technology, and provides an intelligent interview method, device and terminal device. The method includes: obtaining the response information of a candidate during an interview; using a preset language model to convert multiple response sentences into corresponding sentence vector information respectively; determining the sentence set vector information corresponding to the response information according to the sentence vector information of the multiple response sentences; using the sentence set vector information to calculate the relationship probability between multiple entities included in the response information; extracting target relationship information from the multiple entities according to the relationship probability; and generating an interview question for the candidate based on the target relationship information. The above method can quickly extract the important parts in the candidate's answer content, facilitate the artificial intelligence interviewer to give necessary and reasonable follow-up questions, and then generate an interview evaluation report. In addition, the interview evaluation report can be uploaded to the blockchain to ensure its security and fairness and transparency.

A method for evaluating content output by a retrieval augmented generation (RAG) system includes obtaining question-answer information for data residing in a source index and prompting a large language model (LLM) to generate answer construct conditions for a first test question included in the question-answer information. Each of the answer construct conditions identifies a condition that is satisfied by a ground truth answer. The method further includes generating a question-specific evaluation metric for the first test question based on the answer construct conditions and prompting multiple differently configured RAG systems to answer the first test question based on information within the source index. Multiple answers to the first test question, generated by the multiple RAG systems, are evaluated by repeatedly assessing the question-specific evaluation metric and presenting, on a user interface, comparative quality data quantifying a relative quality of the multiple responses generated by the multiple RAG systems.

A kind of offshore oil spill risk source fast intelligent identification classification method of high-resolution remote sensing image belongs to the cross technical field of marine remote sensing and artificial intelligence, including: 1) oil platform and ship as the main risk source of offshore oil spill, with the characteristics of pixel value higher than surrounding seawater, according to the method of pixel value gradient and spot detection, a squeezing excitation residual network combining Gaussian Laplacian operator and connected domain controller is designed;2) using prior knowledge such as pixel gradient and target size, suspected offshore oil spill risk source identification is carried out, the center position and size of the target are obtained using connected domain controller on multiple response points around the target, 3) finally, the suspected offshore oil spill risk source identified is classified by deep learning classification network.The present application only classifies the suspected offshore oil spill risk source target picture in the image, reduces the number of pictures, improves the running time, and the training set and the actual prediction picture size are consistent, only small-scale pictures are needed for training without using the whole remote sensing image, which is easier to train and saves time.

The embodiment of the specification discloses a service response method, device and equipment, the method comprises: receiving the conversation request initiated by the user to the target response party, in response to the conversation request, obtaining the service request information associated with the conversation request, and performing intent recognition on the service request information to obtain the request intent of the user; based on the request intent, determining the target response type for responding to the conversation request, the target response type includes artificial response and intelligent body response; based on the target response type, from the multiple response objects supported by the target response party, determine the target response object adapted to the conversation request, and establish the conversation between the target response object and the user, to provide the conversation service adapted to the request intent for the user through the target response object.

The invention belongs to the field of information security, and discloses a physical unclonable function modeling method based on repetition information, and the method comprises the steps: collecting multiple responses of a PUF to the same excitation, calculating repetition data, and constructing an excitation-repetition data pair; the value of the repetition degree data is converted into a limited category label by adopting a data binning strategy, so that the training complexity is reduced; and training a repetition degree model by using a multi-classification feedforward neural network, and finally obtaining a PUF response through model prediction label mapping. According to the method, response and stability information contained in the repetition degree is fully utilized, efficient modeling of the complex structure PUF is achieved, limitation of traditional reliability attacks is broken through, the method has the advantages of being wide in attack range, high in efficiency, free of knowing the internal structure of the PUF and the like, and the method is suitable for various strong PUF structures and suitable for evaluating the anti-modeling safety of the PUF.

To provide a countermeasure proposal device capable of improving convenience.SOLUTION: The response proposal device includes an acquisition unit configured to acquire an image of a site at which a response-required situation occurs, a generation unit configured to generate a caption based on the image of the response-required situation, a search unit configured to search for a response manual indicating a method of responding to the response-required situation from a plurality of response manuals stored in a storage device based on the caption, and a control unit configured to cause an output device to output the searched response manual.SELECTED DRAWING: Figure 1

An object recognition arrangement includes an enrollment process involving taking a biometric print of an object, which may be a biological object, generating a random seed, deriving challenges from the seed and using the challenges to measure the print resulting in a set of responses. A key is generated and used to encrypt a file. The key is then used to filter the response set, and a filtered subset is retained. Later, during a recognition process, another biometric print is taken of the object, the challenges are generated and applied to the new print, and a full set of responses are measured. The key is then recovered by comparing the full set of responses to the subset. The subset-superset response comparison may be done with multiple response streams, and the best stream is selected based on counting the number of a first binary symbol in the stream.

A key generation method includes: receiving, by a first device, a first pilot signal transmitted by a second device; calculating, by the first device, a channel estimation value based on a first pilot key and the first pilot signal; generating, by the first device, a first key based on multiple first quantization sequences; and transmitting, by the first device, multiple response signals to the second device based on the channel estimation value and the multiple first quantization sequences, where the multiple response signals are used for the second device to generate a second key.

This invention discloses an artificial intelligence-based method and system for monitoring the condition of pile-slab road structures. The method includes: determining the force transmission relationship between monitoring points based on finite element analysis of the structure, and constructing a directed force transmission graph; extracting response feature vectors under healthy conditions and clustering them to obtain multiple response patterns; mapping the response patterns corresponding to multiple historical responses of each monitoring point to the directed force transmission graph, and extracting the response style vector of each node in the directed force transmission graph using a graph convolutional neural network; calculating the response transmission intensity based on the response style vector and response similarity; inferring the response data under real-time load based on the response transmission intensity, and obtaining the condition monitoring results based on the inference bias. This method learns the force transmission path of the pile-slab road structure based on artificial intelligence, continuously monitors the evolution of structural mechanical properties, and achieves a leap from static threshold judgment to dynamic tracking, significantly improving the detection capability of early hidden damage.

The invention relates to the technical field of power maintenance, and discloses a signal injection parameter determination method for a 10kV distribution line grounding fault. The method comprises the following steps: injecting a broadband sweep frequency pulse into a line, adaptively setting a sweep frequency step according to the length of the line, and obtaining no-load spectrum characteristic data; calculating correlation among the candidate frequency points in the available frequency band, and selecting a plurality of characteristic frequency points of which the correlation is lower than a threshold value; injecting energy according to the frequency band where each characteristic frequency point is located, so that the energy is inversely proportional to the attenuation coefficient; collecting a feedback signal and extracting a plurality of response features to form fault feature data; performing similarity comparison on the fault feature data and a standard fault feature library to obtain a fault type and a matching degree; if the matching degree is lower than a threshold value, encryption sampling is performed in the sensitive frequency band, and extraction and matching are repeated after energy is increased; and calculating the fault distance by using the phase deviation difference of the at least two characteristic frequency points. The self-adaptive determination of the injection parameter is realized, and the detection success rate and the positioning precision of the hidden fault are improved.

To provide simple and appropriate responses to the person being monitored, depending on the situation. [Solution] The monitored terminal (100) transmits the acquired voice data and location information to the server (200). The server (200) converts the voice data into text and analyzes the acoustic features to generate a summary and speech intent (including urgency level), and sends notification data to the guardian terminal (300) that includes the summary, location information, and multiple response candidate codes (including action candidate codes) corresponding to the speech intent. The guardian terminal (300) displays the notification data as a push notification (500A), and displays the summary (503), map thumbnail (504), and response buttons (505A, 505B), etc., along with a display according to the urgency level (e.g., icon 501, color change). When the guardian selects a response button, the corresponding code is sent to the monitored terminal (100) via the server (200), and the monitored terminal (100) executes a predetermined output.

A method of processing radar responses of a radar system characterized by a plurality of beam patterns, to determine if a given response corresponds to a reflective object. The method including, for a first beam pattern, identifying a response at an identified angle relative to a center of the first beam pattern. Then, for a second beam pattern overlapping the first beam pattern, determining if a measured response in the second beam pattern at an angle relative to the center of the second beam pattern that corresponds to the identified angle relative to the center of the first beam pattern is within a predetermined error threshold of an anticipated response calculated using the second beam pattern and the first beam pattern. If the measured response in the second beam pattern is not within the predetermined error threshold, the first response is eliminated from display and / or further tracking.

A system is provided.SOLUTION: A system comprising: means for converting a user's voice to text in real-time; means for projecting the converted text to a display; means for displaying a plurality of response options and allowing a user to select any response; and means for speech synthesis and playing the selected response.SELECTED DRAWING: Figure 1