1218 results about "Generative model" patented technology

The invention relates to the technical field of electrical digital data processing, and discloses a system conflict point intelligent review system based on a knowledge graph and a large language model, which comprises the following steps: constructing a dual-mode storage space containing an unstructured index and a structured logic graph, analyzing target text extraction features and triggering graph-based generation logic; converting the topological structure of the associated sub-atlas into a natural language instruction sequence to construct a forced logic constraint template, filling the template with a text, and inputting a pre-training language model to generate a verification result; according to the method, the discrete atlas topology is mapped into the linear logic constraint, random divergence of the generative model is restrained on the calculation principle, and precise decoupling and dynamic evolution of unstructured semantics and structured logic are achieved.

The invention discloses an unknown pollutant non-target identification method and system based on mass spectrum and a generative model, and is applied to the technical field of environmental monitoring and analytical chemistry. The method comprises the following steps: collecting mass spectrum data of a water body sample to be detected; preprocessing the original mass spectrum data, and outputting standardized features; inputting the standardized features into a pre-trained generative model to generate a plurality of candidate molecular formulas; using chemical and physical constraints to eliminate candidate molecular formulas which do not conform to rules; executing a rule-driven algorithm to obtain candidate pollutant molecular structures; carrying out comprehensive scoring and sorting on candidate pollutant molecular structures through chemical prior and environmental prior; and semi-quantitative or relatively quantitative concentration determination is carried out. According to the method, a rule-driven expert system and a data-driven generation model are combined, unknown pollutants which do not exist in a standard library can be effectively recognized and analyzed, and full-process automatic processing from original mass spectrum data to pollutant structures and concentrations is achieved.

The disclosed concepts relate to training a machine learning model to provide help sessions during a video game. For instance, prior video game data from help sessions provided by human users can be filtered to obtain training data. Then, a machine learning model can be trained using approaches such as imitation learning, reinforcement learning, and / or tuning of a generative model to perform help sessions. Then, the trained machine learning model can be employed at inference time to provide help sessions to video game players.

The invention provides a digital twin three-dimensional model construction method based on a generative large model, and the method comprises the steps: obtaining multi-source monitoring data of a power distribution network, and processing the multi-source monitoring data into a training data set; the method comprises the following steps: mapping multi-source monitoring data into a multi-scale tensor subspace through tensor wavelet structured transformation, adaptively extracting spatial features through a learnable wavelet kernel, and keeping the structural continuity of a physical field in combination with a geometric prior regular term; constructing and training a generative adversarial network through a training data set; inputting and analyzing the physical parameter vector of the target scene, and if the topological similarity score is lower than a preset threshold value, adjusting noise vector regeneration; and if yes, outputting a three-dimensional model tensor and importing the three-dimensional model tensor into a digital twin platform, and driving real-time physical field visualization. According to the method, characteristics of a multi-scale space structure and a nonlinear physical field can be reserved, the physical rationality and generalization ability of the generated model are remarkably improved, and depth identification of topological attributes (such as hole connectivity and surface defects) and local geometric defects of the three-dimensional model is realized.

Approaches presented herein provide for the training and use of generative models to generate high quality image data for novel reconstruction views. A generative model such as a neural radiance field (NeRF) can be trained to generate such content. In order to train the NeRF to represent a specific scene with high accuracy, the NeRF can be trained using a diffusion model for score distillation guidance. The diffusion model can be trained using a large set of environment data from a variety of different views, then fine-tuned for a specific domain and / or scene. A parameter-efficient training process can be used to avoid overfitting of the diffusion model to the domain- or scene-specific training data. Once fine-tuned, the “expert” diffusion model can be used with the NeRF during training to effectively transfer the expert knowledge to the NeRF, enabling the NeRF to generate high quality image data for the scene from viewpoints corresponding to extreme novel views.

The invention discloses a public opinion event multi-mode semantic fusion modeling and abstract generation method and system, and relates to the field of natural language processing and social network analysis. Through the multi-mode semantic fusion technology, the short text understanding ability is improved, and the problems of semantic fuzziness and network language diversification are solved. Meanwhile, through a cross-window event cluster matching technology, an event evolution path with time continuity is constructed, and comprehensive capture of event dynamic characteristics is realized. Besides, the structured event abstract is automatically generated by utilizing the generative model, so that the consistency and the information density of the abstract are improved, and the actual application requirements are met. Through the innovations, the defects in the aspects of semantic comprehension, dynamic modeling and abstract generation in the prior art can be effectively overcome, a more efficient and accurate solution is provided for monitoring and analysis of public opinion events, and the method has wide application prospects in the fields of public opinion monitoring, emergency early warning, social media data analysis and the like.

A data processing system implements receiving, via a user interface, report context and a request to generate insights of a data report; constructing a first prompt by appending the request a default system prompt, the report context, and the data report as a first instruction string; validating the first prompt using a second generative model by checking whether the first prompt is structured according to sections that contain one or more predetermined purposes and whether the default system prompt is responsive to the report context; when the first prompt is validated by the second generative model, providing the first prompt to the first generative model; generating, by the first generative model and according to the first prompt, an insight output; receiving the insight output from the first generative model; and providing the insight output to display on the user interface.

The invention discloses a document review method based on multi-agent collaboration and retrieval enhancement generation. The method comprises the following steps: firstly, analyzing a document review rule input by a user into a rule semantic intermediate representation through a natural language processing technology, and calling a retrieval enhancement generation (RAG) module to expand related knowledge to form a structured rule library; secondly, multi-modal analysis and chapter segmentation are carried out on a to-be-examined document, and elements such as texts, tables, pictures and formulas are expressed in a unified mode; tasks such as rule analysis, document segmentation, knowledge retrieval, matching comparison and report generation are completed through a multi-agent cooperation mechanism; finally, when ambiguity exists in rule and document matching, an RAG module is introduced to retrieve supplementary evidences from an external knowledge base, explanatory comparison is conducted in combination with the generative model, and the accuracy and authority of judgment are improved.

The invention relates to the technical field of unmanned aerial vehicle monitoring and trajectory processing, and discloses a low-altitude unmanned aerial vehicle dynamic trajectory tracking and predicting method based on multi-base-station cooperation. The method comprises the steps that signal parameters are obtained through multi-base-station collaborative observation, and a multi-modal position prediction set is generated; and performing classification and scoring according to the spatial distribution characteristics of the candidate points and the historical track points, and screening out an optimal prediction position point. And inputting the optimal prediction point and the historical trajectory into a generative model, and dynamically adjusting the number of trajectory points by analyzing the point distribution probability to form a preliminary smooth trajectory. And performing physical feasibility verification and fine adjustment on the trajectory according to kinematics constraints to obtain a final smooth continuous trajectory. And matching degree calculation is carried out by fusing alternative trajectories generated by a multi-kinematic model, and the most probable target trajectory is identified. According to the method, the robustness of trajectory prediction and the structural rationality of the generated trajectory in a complex observation environment are improved.

Systems and methods for image search result filtering can include obtaining a search query, determining a plurality of candidate image search results, processing the search query with a generative model to determine a plurality of search result criteria, and refining the plurality of candidate image search results based on determining whether the candidate results satisfy the plurality of search results criteria. The systems and methods can perform a plurality of determinations based on the output of the generative model.

The invention discloses a data deep learning and intelligent analysis method based on an AI artificial intelligence technology, and relates to the technical field of basic AI models, and the method comprises the steps: employing a multi-modal data preprocessing module to carry out the expansion of small sample data through a generative model, and combining with meta-learning to extract prototype features, meanwhile, an epsilon-differential privacy budget is dynamically allocated based on the data sensitivity level so as to inject dynamic noise; establishing a layered federated learning architecture, training a model by local training nodes through a loss function containing a self-adaptive regularization item, and performing sparse processing and gradient disturbance before uploading parameters; the global aggregation node adopts a weighted federated average algorithm to aggregate parameters, and dynamically adjusts the communication frequency according to the loss convergence speed; and a target model is obtained through iterative training, and a decision interpretation report containing the attention thermodynamic diagram and the desensitization identifier is generated when a result is output. According to the method, the problems of small sample overfitting, data islands and privacy disclosure are effectively solved, and the accuracy and practicability of the model are improved.

The invention discloses an optical diffusion plate microstructure defect detection method, electronic equipment and a storage medium. The method comprises the following steps: acquiring images of the same diffusion plate area under at least two different illumination angles and performing elastic registration so as to eliminate image dislocation caused by plate movement; decoupling and separating physical defect features stably existing under all illumination and shadow interference features changing along with illumination from the registered image; carrying out enhancement and dynamic up-sampling processing on the physical defect features, and generating a suspected defect candidate region set with a low confidence threshold; for each candidate region, reconstructing a virtual reference image when the region has no defect through a generative model by utilizing the peripheral texture of the candidate region, and judging authenticity and outputting a confidence coefficient by comparing residual errors; and determining a final alarm result according to whether the confidence exceeds an alarm threshold. According to the invention, artifacts and real physical defects caused by mechanical micro-vibration and micro-structure reflection can be effectively distinguished, and the false alarm rate is obviously reduced.

The invention relates to a pathological section intelligent auxiliary differential diagnosis system based on multi-modal fusion, in particular to the field of clinical pathology, semantic unification of multi-modal data is achieved through meta-task construction and a cross-modal alignment technology, and transferable diagnostic knowledge is extracted by utilizing a meta-learning framework; the method combines a generative model and knowledge constraints to generate high-quality synthetic data, and finally fuses real and synthetic samples through a self-adaptive diagnosis mechanism, thereby remarkably improving the differential diagnosis capability of rare lesions, effectively solving the problem of model generalization in a training data scarcity scene, and improving the accuracy of model identification. And efficient and reliable intelligent auxiliary decision support is provided for clinical pathological diagnosis.

Techniques for creating and updating natural language summaries representing personalized user knowledge (e.g., user interests, user affinities, user preferences, family structure, routines, and other insights) based on conversational interactions with and other natural language content available to an AI system are described. In some embodiments, to provide a more personalized service, a system can use a generative model to summarize learnings about a user and determine helpful nuanced insights about the user such as “the user is learning how to play guitar.” This “user knowledge” can be updated based on further (later) conversations with the user, where updating can involve negating or deleting stored information, adding to or modifying stored information, etc.

A processing system including at least one processor may obtain a natural language request for a network topology visualization associated with a communication network. The processing system may next generate a prompt based upon the natural language request in accordance with a prompt mapping function, apply the prompt as an input to a generative model to generate a query, and apply the query to a communication network database system to obtain a query result. The processing system may generate the network topology visualization from the query result. The processing system may then present the network topology visualization via at least one display device.

The invention discloses a business process multi-dimensional logic verification method and system based on a dynamic knowledge graph, and relates to the technical field of artificial intelligence and financial risk control. The method aims at solving the technical problems that a traditional rule engine cannot process unstructured data, and a large generative model has logic illusion and privacy risks. The method comprises the following steps: constructing a rule knowledge graph containing parameterized logic constraints (such as time sequence, mutual exclusion and numerical accumulation); converting the unstructured business operation record into a feature vector by using a locally deployed semantic model; performing global sequence alignment optimization by constructing a global semantic consistency objective function, mapping operation records to a theoretical decision path derived from a graph, and eliminating local matching ambiguity; and finally, on the basis of a mapping result, calling deterministic parameters in map edge attributes to execute strict mathematical logic verification (including complex workday calculation, mutual exclusiveness check and the like). According to the method, a'nerve-symbol 'dual architecture is adopted, accurate understanding of fuzzy semantics is achieved, logic verification certainty and data privacy security are guaranteed, and the method has a hidden process discovery capability.

An online system applies a visual language model and an optical character recognition model to a received image to generate descriptive information about unknown items in the image. The online system prompts a generative model with the descriptive information about unknown items in the image to separate the descriptive information into different bins each corresponding to a different unknown item in the image. For each unknown item detected in the image, the online system generates a target embedding from its descriptive information and performs a nearest neighbor search on an item catalog including embeddings for various items to find a set of candidate embeddings matching the target embedding. The online system retrieves item attributes of candidate items each corresponding to a candidate embedding of the set and prompts the generative model with this information to rank candidate items for the unknown item in the image.

The invention discloses a user track privacy protection method and system fusing a generative model and three-stage personalized differential privacy, and relates to the technical field of data security, and the method comprises the steps: obtaining real user track data, and carrying out the preprocessing of the real user track data to form standardized track representation; training a depth generation model to generate synthetic trajectory data; performing density clustering on the synthetic trajectory to obtain a feature region and calculating a region centroid; point-level privacy budgets are distributed according to three stages of the area proportion, the centroid distance and the global density; constructing an adaptive elliptical disturbance domain based on the regional covariance, sampling candidate confusion points in the disturbance domain and applying road feasibility screening; and calculating a score based on the privacy budget and the utility function, and sampling replacement points by adopting an index mechanism to output a privacy protection trajectory. According to the technical scheme, the track structure continuity and the road semantic consistency are remarkably improved under the condition that the differential privacy constraint is met, and the privacy intensity and the data availability are both considered.

Certain aspects of the present disclosure provide techniques and apparatus for generating a response to a query input into a generative artificial intelligence model. The method generally includes generating, based on an input query and a first generative model, a plurality of sets of tokens, each set of tokens in the plurality of sets of tokens corresponding to a candidate response to the input query; outputting, to a second generative model, the plurality of sets of tokens for verification; receiving, from the second generative model, an indication of a selected set of tokens from the plurality of sets of tokens based on the input query and the plurality of sets of tokens; and outputting the selected set of tokens as a response to the input query.

Provided are systems and methods for performing panoptic segmentation of images and videos using a denoising diffusion model. The panoptic segmentation task is formulated as a conditional discrete data generation problem. This is achieved by learning a generative model for panoptic masks, for example treated as an array of discrete tokens, conditioned on an input image. The generative model can also be applied to video data by including predictions from past frames as an additional conditioning signal. This enables the model to learn to track and segment objects automatically across video frames.

The invention provides a space intelligent non-planar reflection generation method based on surface normal guidance. Extracting multi-layer visual representation from an original RGB image containing a non-planar reflection medium scene, and identifying a non-planar reflection area by combining brightness distribution of the original RGB image and a spatial symmetry relation; analyzing non-linear deformation features of textures in the non-planar reflection area, and predicting pixel-level three-dimensional surface normal vectors and confidence distribution in combination with multi-layer visual representation to construct a surface normal graph; determining a line-of-sight vector based on a camera imaging model, and executing geometrical optical mapping operation on the line-of-sight vector, the surface normal graph and the confidence distribution to construct a reflection offset field; extracting an intermediate semantic feature representation corresponding to the non-planar reflection region, and generating a pre-distortion reflection feature representation according to the reflection offset field and the intermediate semantic feature representation; and injecting the pre-distortion reflection feature representation, the multi-layer visual representation and the initial reflection area mask into a generative model for image reconstruction, and generating a non-planar reflection image.

A training method and apparatus for a generative model. The training method comprises: acquiring a first training sample which comprises a first initial image and first information; inputting the first training sample into a generative model for model processing to obtain a first target image carrying steganographic information; inputting the first target image into a pre-trained discrimination model to obtain a first visibility score, wherein the first visibility score indicates the visibility of the steganographic information; and with the objective of reducing a first loss, adjusting parameters of the generative model, wherein the first loss is positively correlated to the first visibility score.

The invention relates to the technical field of acoustic performance testing of electronic products, in particular to an AI model-based automatic detection method and system for noise defects of electroacoustic devices such as a loudspeaker and a receiver. The system comprises a hardware acquisition module, a data processing module and a core AI analysis engine, and adopts a two-stage architecture of unsupervised anomaly detection and supervised defect classification: in the first stage, only a non-defective sample is needed to train a generative model, and high-sensitivity anomaly detection is realized by reconstructing a difference spectrogram; in the second stage, accumulated defective product data is utilized to train a classification model, and automatic identification of defect types such as air leakage and particle loosening is completed. According to the invention, cold start deployment, low omission ratio, interpretable diagnosis and model sustainable evolution are realized, and the method is suitable for automatic quality inspection of large-scale electroacoustic devices.

Systems and methods are disclosed comprising techniques for preliminary signal evaluation, such as generating a first session record for a first environment state for the user runtime session, retrieving a second session record for a second environment state for prior user runtime sessions, inputting the first and the second session records into a first generative model to generate a predicted session event set for the user runtime session, selectively determining a prioritized session event from the predicted session event set based on comparing realization parameters of the predicted session event set, inputting the prioritized session event and the first environment state into a second generative model to generate a preliminary session event for execution during the user runtime session, and when the environment state of user runtime session satisfies activation criterions of the preliminary session event, executing the preliminary session event prior to execution of the prioritized session event.

Systems and methods for audience-based content modification can include obtaining a content item from a link notes interface, obtaining a user embedding associated with a particular user, determining to augment the content item based on the user embedding, processing the content item and the user embedding with a generative language model to generate an alternative content item, and rendering the alternative content item in place of the content item within the link notes interface. The systems and methods can leverage linguistic characteristic determinations and generative model predictions to generate model-generated content items that vary based on the viewing user.

A computer-implemented process for creating 3D objects from 2D images includes receiving an input image, conditioning a generative model using image-derived, voxelized three-dimensional features, generating, by a transformer-based rectified-flow generative model parameterized as a base network optionally coupled to one or more low-rank adapter modules activatable at inference, a volumetric latent of the target object, the volumetric latent including a sparse, feature-augmented volumetric lattice obtained by transporting an initial random sample toward a learned manifold via a rectified-flow sampling process, decoding the volumetric latent by mapping the volumetric latent to a feature-bearing sparse volumetric field consistent with the volumetric lattice, decoding the field to a continuous implicit surface function, and extracting a watertight mesh by isosurface extraction, estimating camera-pose parameters by render-and-compare alignment between silhouettes rendered from the mesh and silhouettes of the input image, and, performing style-preserving inverse rendering on the mesh that updates UV-space albedo and material maps.

The invention discloses a generative model redrawing attack-oriented image robust watermark traceability method and system and a computer readable storage medium, and belongs to the field of digital information security and artificial intelligence content governance. The method comprises the following steps: in a watermark embedding stage, performing error correction coding and digital signature processing on traceability information containing identity information and a timestamp to generate a load to be embedded, and embedding the load to be embedded and a synchronization template for geometric synchronization into a host image in a function separation manner by using a deep neural network; in the training stage, combined optimization is carried out on the watermark embedding and extracting process by introducing generative model redrawing and microsimulation of image distortion attack, so that the robustness of the watermark under a complex attack condition is improved; in the extraction and verification stage, under the condition that an original image is not needed, geometric synchronous correction and blind extraction of a watermark load are carried out on an image to be analyzed, and authenticity confirmation of traceability information is completed through error correction decoding and digital signature verification. By adopting the technical scheme of the invention, the problems of insufficient traceability information robustness, difficult source confirmation and incomplete evidence chain in a generative model redrawing scene in the prior art are solved, and verifiable traceability and credible evidence generation of the image source information are realized.

Approaches presented herein provide for the use of reinforcement learning to fine-tune a generative model, such as a motion diffusion model, for a specific objective, such as to generate representations of human motion corresponding to provided text input. A discriminator can be used to guide the training of the generative model. In at least one embodiment, the discriminator can compare the input text and generated motion representation (or embeddings of each) to determine an alignment value or match score, for example, which can then be used to adjust the network parameters or weights of the generative model to improve the alignment between input text and generated motion.

Some implementations relate to receiving input data; generating, using a low-rank representation of a machine-learned generative model, a generative output from the input data; determining, based on a machine-learned reward model, a corresponding reward from the generative output, and updating, based on the corresponding reward, one or more parameters of the low-rank representation of the machine-learned model. Further, some additional or alternative implementations relate to receiving input data associated with a client device; generating, using a general purpose agent, responsive content to the input data, wherein the general purpose agent is configured based on a machine-learned generative model and a low-rank representation of the machine-learned generative model; and causing the client device to render the responsive content.

The invention relates to the technical field of textbook construction and optimization, in particular to an automatic textbook construction and optimization method and device.The method comprises the following steps that original resources of multiple types of textbooks are collected, and core knowledge elements in the original resources are recognized; constructing a structured mapping relationship between the knowledge elements and the teaching resources based on the core knowledge elements; on the basis of the structured mapping relationship, constructing a knowledge graph covering multidisciplinary knowledge points and a preposition dependency relationship, and automatically identifying knowledge faults and complementing missing association so as to form a knowledge network with self-evolution ability; based on the knowledge network, utilizing a hybrid generation model to cooperatively work, and quickly generating a structured textbook framework containing knowledge veins, teaching cases and exercise resources; learning condition data including classroom behaviors, learning feedback and knowledge mastering conditions are obtained, and a quantity and chemical condition report is obtained after data processing; and updating data based on the quantitative chemical situation report and the knowledge graph.