19 results about "Verification problem" patented technology

The invention relates to the technical field of block chain technology and artificial intelligence collaboration, and discloses a verifiable multi-agent collaboration method based on smart contract driving, comprising the following steps: task planning generates a directed acyclic graph and a zero-knowledge proof circuit template, submits the directed acyclic graph and the zero-knowledge proof circuit template to a planning contract, and deploys an exclusive zero-knowledge proof contract; the AI agent collaboratively executes the subtasks, and zero-knowledge proof is generated and submitted through a Veckrii auction tool; zero-knowledge proof contract chain verification is carried out, and after verification is passed, a signature is received and multi-layer gating verification is executed; and the governance contract carries out pledge management and cost distribution on the agents, the providers and the experts. According to the method, zero-knowledge proof and a trusted execution environment are combined, and an under-chain execution behavior is converted into an encryption certificate capable of being verified on a chain, so that the problem that the execution process cannot be verified due to the lack of distributed trusted evidence storage in a traditional centralized framework is effectively solved; and after each subtask is completed, double signature and multi-layer gating verification are forcibly executed.

This application discloses a puzzle generation method, apparatus, device, and readable storage medium. The method obtains a structured specification representation and seed parameters of a seed puzzle; generates a new problem using the seed parameters and the structured specification representation; verifies the accuracy of the structured specification representation using the new problem; if accurate, uses a puzzle generator to load the structured specification representation, reorganize the logical formula, and randomly sample specified data according to defined constraints to obtain a SMT problem; after verifying the SMT problem's logical reliability and solvability using a solver, obtains new configuration parameters based on the SMT problem; fills the new configuration parameters into the structured specification representation and renders it using a natural language template to obtain a target puzzle with a problem instance and a standard answer. This application, through pre-coding, reproducing, and verifying puzzles, can automatically generate accurate and reliable new puzzles in batches.

The invention relates to the technical field of blockchain technology application, and discloses a technology transfer information blockchain management system based on a smart contract. The system comprises a local verification node, a cloud policy engine, a partition permission controller, an encryption transmission channel, a policy divergence analysis module and a hierarchical execution module. The local verification node triggers a preset intelligent contract rule in the fixed verification partition to perform initial transaction compliance verification; if the initial verification fails, the cloud policy engine extracts verification problem features and stores the verification problem features in the temporary cache region, and activates the remote verification partition take-over permission; the partition permission controller locks and fixedly verifies the partition data write-in permission and reads related data; the encryption transmission channel uploads data to the cloud through a security protocol; the strategy divergence analysis module generates a multi-stage verification strategy; and the hierarchical execution module loads a multi-level strategy hierarchical verification risk, and issues a block chain binding permission according to a result. The system enhances the flexibility and security of technology transfer transaction verification, and improves the efficiency.

The application discloses a kind of mathematical application question solving method and system based on multi-path search, belong to problem solving technical field, the method utilizes mathematical solving model to solve mathematical application question, after determining initial answer;Using initial answer replaces key numerical condition in mathematical application question, obtains verification problem;Using mathematical solving model solves verification problem, determines verification problem answer;The consistency between key numerical condition and verification problem answer is calculated, to determine whether initial answer is accurate;When determining that initial answer is not accurate, according to initial answer to construct negative prompt information;Utilize negative prompt information to revise erroneous reasoning path, until the initial answer that mathematical solving model solves is determined to be accurate;The initial answer that is determined to be accurate is used as the final answer of mathematical application question.Accurate solution to mathematical application question is realized, and the problem that current method is not accurate to solve mathematical application question is solved.

The invention discloses a confidence-based verification method for the steering angle safety of an unmanned driving system, and belongs to the field of verification of the safety of the unmanned driving system. The method comprises the following steps: constructing a product network by using a self-combination technology, converting global robustness verification of an unmanned neural network into a standard verification problem, then linearizing an activation function of the neural network, and executing safety verification of an unmanned steering angle by using a neural network safety analysis tool Marbou. According to the technology, super attribute verification of the unmanned driving system can be realized, and the reliability is high; calculation is efficient, feedback can be quickly given out for the situation that the safety attribute is not met, and the method is easy to achieve in an existing unmanned driving system.

This application provides a fixed solution verification method, system, electronic device, and storage medium for PPK (Programmable Point Kinematics) analysis. The method includes: acquiring GNSS observation data from multiple epochs at a user station; performing ambiguity fixing calculations based on the GNSS observation data for each epoch to obtain candidate fixed solutions, where each epoch corresponds one-to-one with a candidate fixed solution; calculating the carrier phase double-difference integer characteristic corresponding to each candidate fixed solution; verifying the correctness of each candidate fixed solution based on the carrier phase double-difference integer characteristic across multiple epochs; and determining the positioning result based on the candidate fixed solution if it is correct. This method fully utilizes the correlation of multi-epoch observation information in PPK post-processing mode, transforming the verification problem of a single-epoch fixed solution into a check of the consistency of correlated information across multiple epochs, significantly improving the accuracy and reliability of fixed solution verification.

This invention discloses an automated testing method, system, equipment, storage medium, and program product for microscopic imaging measurements. It addresses the problem of objective verification of measurement results through a log-driven automatic extraction and judgment mechanism; establishes a geometrically dependent linkage verification link to solve the correctness verification problem in dynamic linkage scenarios; uses the start timestamp of the test step as a filtering anchor point for pattern grouping and matching to solve the problem of accurate location of effective logs; and constructs a unified "log parsing—geometric calculation—threshold judgment" process to solve the integrated verification problem of multiple types of measurement scenarios, forming a traceable link of "operation action—log fragment—geometric reference value—judgment result," thus solving the problem of difficulty in tracing back the cause of failure.

The invention discloses an oil field GIS modularization application program interface verification method and system, and the method comprises the steps: firstly, setting a uniform interface specification, guaranteeing that all modules follow the same standard during the interface development, and reducing the verification problem caused by the interface difference; the access point of each interface is provided with interface metadata used for describing the basic information, the parameter type, the value range and the like of the interface. The interface metadata is the basis of interface verification and provides all information required during interface calling. When the interface is called, the transmitted parameters are dynamically verified according to the interface metadata to ensure that the parameters conform to the definition in the interface metadata. And after the parameter verification succeeds, obtaining a corresponding business logic processing object and a data format according to the interface metadata. This ensures that the system can correctly process the interface request, avoiding problems caused by mismatching of data formats or processing logic errors. Therefore, the method provided by the invention can solve the problems in the prior art.

The present application relates to a kind of neural network verification methods based on error divide and conquer, the strict approximation function of network node in given input range is given using improved symbolic linear relaxation calculation, and using the error constraint segmentation strategy of over-approximation node divides the original verification problem into equivalent series of sub-problems, and the safety property of entire neural network is verified by carrying out complete verification to series of sub-problems.Through the method of the present application, the state space in neural network verification process is effectively reduced, the safety property verification and evaluation of neural network can be efficiently carried out, and the scalability of neural network trusted verification method can be improved.

The application provides a state machine-based multi-node interoperation process test system and a state machine-based multi-node interoperation process test method. The application aims at the interoperation process verification requirement among multiple nodes and more interactive information. First, the application solves the interoperation verification problem among multiple nodes through scene presetting, simulation, multi-role support and other methods. Second, the application overcomes the limitations of architecture and development language and expands the application scene through the basic model of the finite state machine model. Third, the application solves the syntax layer test and syntax layer error coverage problem in the verification process through the extraction of a general protocol analysis layer. Fourth, the application realizes the test support and error coverage of the pragmatic layer by constructing the preset scene containing the error process and the state machine conversion design, adopting the mode of mapping the actual application process with the preset process model, and displaying the difference between the actual process and the expected result.

The invention discloses a mathematical application question solving method and system based on multi-path search, and belongs to the technical field of question solving. The method comprises the steps that a mathematical solving model is used for solving a mathematical application question, and after an initial answer is determined; replacing a key numerical condition in the mathematical application question with the initial answer to obtain a verification question; solving the verification question by using a mathematical solving model, and determining an answer to the verification question; calculating the consistency between the key value conditions and the answers of the verification questions to judge whether the initial answers are accurate or not; when it is judged that the initial answer is inaccurate, negative prompt information is constructed according to the initial answer; correcting the incorrect reasoning path by using negative prompt information until the initial answer solved by the mathematical solution model is judged to be accurate; and taking the accurately judged initial answer as a final answer of the mathematical application question. The mathematical application problem is accurately solved, and the problem that the mathematical application problem is not accurately solved through a current method is solved.

Methods and systems are described for verifying an identity of a user through contextual knowledge-based authentication. The system described uses contextual knowledge-based authentication. By verifying an identity of a user through contextual knowledge-based authentication, the verification is both more secure and more intuitive to the user. For example, by relying on confidential and / or proprietary information, the system may generate verification questions, the answers to which are known only by the user.

This invention proposes a quantitative evaluation method for port access control policies based on distributed dynamic verification, used for network security baseline verification, belonging to the field of network information security technology. Addressing the port access control policy verification problem in network security baseline verification, this invention designs a quantitative evaluation method for port access control policies based on distributed dynamic verification, including: a distributed dynamic verification subsystem, a software distribution subsystem, a dynamic verification task scheduling subsystem, and a quantitative evaluation subsystem. This method achieves fine-grained measurement and evaluation of port access control policies in the network through distributed port open information collection and distributed dynamic verification design, thereby improving the coverage and accuracy of network security baseline verification.

The invention provides a verification code generation method, a verification code verification method and a computer program product. In response to a verification request of a user, a large language model is called to generate a semantic understanding type verification question and a plurality of candidate answers corresponding to the verification question; according to the method, a semantic understanding type verification problem is generated, and a common attacker cannot pass verification through technologies such as simple image recognition and must accurately understand the semantics of the verification problem, so that compared with a traditional verification code, the verification safety is greatly improved.

The application discloses a model optimization method and device and electronic equipment. The method comprises the following steps: a strategy model to be optimized is used to analyze a problem text set, a plurality of candidate answers corresponding to each problem text are obtained, and the strategy model is subjected to first-stage iterative training according to the plurality of candidate answers and a first loss function, wherein the plurality of candidate answers are different sampling results generated by the strategy model based on the same prompt, the problem text set is a set composed of verifiable problems, and the plurality of candidate answers corresponding to each problem text are a group of candidate answers; after the first-stage iterative training of the strategy model, error samples are determined from the plurality of candidate answers; and the strategy model is subjected to second-stage optimization training according to the error samples and a second loss function. The application solves the technical problems of insufficient learning signal utilization and low data utilization rate of a traditional reinforcement learning scheme when a large language model is trained to reason verifiable problems.

The embodiment of the invention provides a machine tool emergency rollback system, a control method thereof, electronic equipment and a medium. The method comprises the following steps: determining a configuration scheme of the machine tool emergency rollback system; executing an emergency rollback function test on a rollback shaft and a delay stop shaft of the machine tool based on the configuration scheme, and recording test data; analyzing the test data to obtain actual key function parameters of the machine tool emergency rollback system, and comparing the actual key function parameters with specified function parameters set in the configuration scheme to judge whether the configuration scheme meets the function requirements of the emergency rollback system or not. According to the embodiment of the invention, the function verification problem of the machine tool emergency rollback system configuration is solved, so that the efficient, accurate and reliable operation of the system is ensured.

This application discloses a quantum circuit verification method, device, storage medium, and electronic device. The quantum circuit verification method transforms the quantum circuit verification problem into an optimal control problem on a unitary manifold, calculates the optimal control trajectory and minimum action value using the minimum action principle, and determines the circuit equivalence by comparing these quantities. This method overcomes the bottleneck of exponential complexity in traditional full-state simulation methods, making the computational complexity of the verification process increase polynomially with the circuit size, thereby enabling efficient verification of large-scale quantum circuits.

The invention provides a correct rate verification method and device of an RAG knowledge base and a storage medium, and solves the problems that a traditional correct rate verification mode of the RAG knowledge base is low in manual sampling inspection efficiency, and an auditing result is separated from a statistical caliber, the method comprises the following steps: importing a plurality of verification questions and a standard answer of each verification question into a unified file template, the field integrity is ensured; the multiple verification questions are input into a large language model of the RAG knowledge base in batches and concurrently for retrieval, the similarity between the structured JSON format answer of each batch of verification questions and the standard answer of each batch of verification questions is obtained and calculated, and whether the structured JSON format answer is correct or not is marked based on the similarity; displaying the data on a visual auditing interface, and configuring a result judgment button for each verification problem in each batch of verification problems; and in response to the verification operation of the trigger result judgment button, updating the global correct rate of the RAG knowledge base.

The present application relates to the technical field of question and answer, and particularly relates to an answer relevance evaluation method, device, equipment and medium. The method comprises the following steps: obtaining an evaluation task carrying an original question and a to-be-evaluated answer corresponding to the original question; matching the to-be-evaluated answer of the evaluation task with historical answers in a cache; if the matching is successful, determining a first verification question list corresponding to the historical answer matched successfully; if the matching is not successful, calling a large language model to infer an answer of the to-be-evaluated answer, and obtaining a second verification question list; and evaluating the original question according to a target verification question list to obtain an evaluation result. The to-be-evaluated answer is matched with the historical answers in the cache in terms of semantics; after the matching is successful, the verification question list corresponding to the historical answer is directly reused, so that the process of generating the verification question list according to the answer by the large language model is skipped, and the step of similarity calculation and scoring is directly entered, thereby greatly reducing the cost and delay.