9 results about "Metamorphic testing" patented technology

This invention discloses a testing and output repair method for an instruction-guided image editing model based on metamorphic testing technology, comprising: collecting the original input image and editing instructions as original test cases; collecting the output images of the original test cases; comparing the input image and the output image to obtain the editing region of the original test cases; transforming the input image according to the metamorphic relationship to generate a candidate image set; selecting equivalent derived test images and combining them with the original editing instructions to form new derived test cases; collecting the output images of all derived test cases; comparing the input image and the output image to obtain the editing region of the derived test cases; determining whether the editing regions of all derived test cases are consistent with those of the original test cases; cross-validating the outputs of all derived test cases with those of the original test cases; selecting the output of the best test case and mapping it onto the original test case as the original test output for repair; summarizing the test results and outputting the repair results.

The invention discloses a test method for software containing a deep neural network, which comprises the following steps of: firstly, analyzing software characteristics of the software containing the deep neural network, and designing a mutation operator according to the software characteristics; and then applying the mutation operator to software containing the deep neural network for mutation to generate a software variant. And then analyzing and judging whether the current test case set meets the test sufficiency requirement or not by using the software variant. If yes, the current test case set is directly used for conducting software testing on the software containing the deep neural network. And if not, generating a new test sample based on the metamorphic test technology and the generative adversarial network model to expand the test case set. And finally, performing software testing on the software containing the deep neural network by using the expanded test case set, and analyzing and judging whether the expanded test case set meets the test sufficiency requirement or not. If yes, proving that the software is fully tested, and ending the test. And if not, repeating the steps to continue to expand the test case set.

The invention discloses a dynamic simulation execution extension and verification method for an instruction. The method comprises the following steps: 1) acquiring instruction semantic data of a target processor and defining a group of metamorphic relationships; 2) generating an instruction semantic function based on the instruction semantic information as a reference model; 3) generating a random test case based on the reference model; 4) mining a metamorphic relation of the instruction by using a large model; 5) formalizing each metamorphic relation into a logic constraint, and verifying whether the reference model conforms to the metamorphic relation or not; if the verification is passed, confirming that the corresponding metamorphic relationship is valid; if a counter example occurs, rejecting an unestablished metamorphic relationship; the method comprises the steps of (1) verifying a random test relation and a metamorphic relation, (2) generating a metamorphic test case based on each verified metamorphic relation, (7) completing simulation implementation of an instruction function according to an instruction semantic assembly interface, (8) executing the generated random test and metamorphic test cases and comparing an execution result with output of a reference model, if the execution result is inconsistent with the output of the reference model, readjusting instruction implementation, and otherwise, completing verification.

The invention discloses a black box testing method and system for an infrared image target detection algorithm, and belongs to the technical field of computer vision and algorithm testing. According to the method, the comprehensiveness, robustness and generalization of the detection algorithm are tested on the premise that the internal structure and code logic of the algorithm are not obtained. The method comprises the following steps: firstly, on the basis of an original data set with normal gray level, generating aspect data required by a test by using a condition coverage and metamorphic test metamorphic relation method and a mode of collecting other data sets; secondly, inputting the generated data set into an infrared image target detection algorithm, and drawing an ROC curve, a PR curve and a confusion matrix through an IoU value between an algorithm prediction frame and a real frame; then, according to a performance test result, evaluation indexes such as Precise, Recall and F1 values are calculated, the test performance of the algorithm on a corresponding data set is obtained, and the comprehensiveness, reliability, generalization and efficiency of the algorithm are verified through a test result.

The invention discloses a DBMS logic defect detection-oriented value semantic perception metamorphic test method and system, and relates to the technical field of database management, and the method comprises the following steps: designing a basic grammar for a to-be-tested DBMS to construct an original query; locally rewriting the original query to generate a variation query, wherein the original query and the variation query meet an approximate relationship; analyzing the original query and the variation query into an abstract syntax tree AST, searching a propagation link locally rewritten on the AST, and generating a query-level approximation relationship between the original query and the variation query; and executing the original query and the variation query on the to-be-tested DBMS, collecting an execution result, comparing the execution result with the query-level approximation relationship, and judging whether a logic defect exists or not. The invention provides a unified SQL (Structured Query Language) query approximation model fusing ensemble semantics and value semantics reasoning, and logic errors in a database management system are detected by changing SQL queries and spreading semantic approximation.

The invention provides a quantum compiler efficiency evaluation method based on a metamorphic test, and the method comprises the steps: obtaining a coupling graph Q and an allowable quantum gate set G of a target quantum chip, and setting the layer depth of a circuit; the evaluation stage comprises at least one of the following two tests: 1, generating an executable random circuit C meeting physical constraints based on a coupling graph Q, compiling the C to obtain a compiled circuit C ', rejecting a non-operational instruction from the C and the C', constructing a gate-level dependency graph, and performing isomorphic judgment on a graph with a label; 2, generating an executable front-section circuit according to the coupling graph Q, taking a physical bit corresponding relation at the end of the executable front-section circuit as a fragment entry layout, performing optimal compiling scheduling on randomly generated small fragments on the Q to obtain an optimal result C *, accessing the fragments according to the mapping of the C * to form a middle-section window, generating an executable rear-segment circuit on the Q by taking the physical bit corresponding relation at the end of the segment as a starting point, and combining the front end, the middle segment and the rear segment to obtain a test circuit C; c ''is obtained by compiling the Cs, the number of swap gates is used as a main index in a fragment window, the total number of two-bit gates or weighted depth is used as an auxiliary index, and the C' 'and the C * are compared to obtain a swap gate difference value; in the output stage, an evaluation report is generated based on an isomorphic conclusion in the function preserving test and a swap gate difference value in the optimal fragment benchmarking test; according to the method, the efficiency can be improved through two metamorphic relations of isomorphic maintenance and optimal comparison.

This invention provides a method and apparatus for risk testing of deep learning models based on metamorphic testing, belonging to the field of artificial intelligence technology. The method includes: acquiring source test cases; selecting a target-specific metamorphic relationship pattern from a dedicated metamorphic relationship pattern library, which includes various dedicated metamorphic relationship patterns for multi-dimensional risk testing; transforming the source test cases based on the target-specific metamorphic relationship pattern to generate derived test cases; inputting the source test cases into a deep learning model to obtain source output, and inputting the derived test cases into the deep learning model to obtain derived output; verifying whether the source output and derived output satisfy the expected relationship to determine the risk status of the deep learning model; updating the risk feature library and optimizing the dedicated metamorphic relationship pattern library based on the risk status, and dynamically adjusting the testing strategy of the deep learning model. This invention employs multiple testing modes, improving the accuracy and comprehensiveness of the testing.

This invention discloses a value semantic-aware metamorphic testing method and system for detecting logical defects in DBMS, belonging to the field of database management technology. The method includes the following steps: designing a basic syntax for the DBMS under test to construct an original query; partially rewriting the original query to generate a mutated query, wherein the original query and the mutated query satisfy an approximation relationship; parsing the original query and the mutated query into an abstract syntax tree (AST) and finding the propagation link of the partial rewrite on the AST to generate a query-level approximation relationship between the original query and the mutated query; executing the original query and the mutated query on the DBMS under test, collecting the execution results and comparing them with the query-level approximation relationship to determine whether there is a logical defect. This invention proposes a unified SQL query approximation model that integrates set semantics and value semantic reasoning, detecting logical errors in database management systems by mutating SQL queries and propagating semantic approximations.

The invention discloses a metamorphic test-based large language model iterative optimization training method and system, and relates to the technical field of artificial intelligence, and the method comprises the steps: collecting query-output pair data of a large language model on a plurality of original queries; performing transformation processing on each original query to generate variant queries, and obtaining corresponding variant output; identifying a plurality of causal metamorphic relationships based on the type of the transformation processing, the difference between the original output and the variant output and the semantic dependency relationship before and after the transformation; constructing an interpretable neural network verifier; constructing a causal constraint loss function; carrying out joint training by adopting a multi-layer optimization framework, and carrying out incremental learning on the large language model through a mechanism of combining meta-learning and continuous learning; and outputting the trained large language model. According to the method, through causal metamorphic relation constraint and a multilayer optimization training mechanism, the output consistency, robustness and interpretability of the large language model can be improved, and efficient iterative optimization and rapid task adaptation are realized.