25 results about "Program slicing" patented technology

The invention discloses a bug repairing method and system based on a grammar semantic tree, and the method comprises the steps: extracting a repairing mode fusing a grammar structure and a semantic intention from a historical bug repairing case, and guaranteeing the accuracy of the mode through program slicing guided by a key entity and iterative verification; on the basis of a double-constraint mechanism of grammar mergibility and semantic consistency, a specific mode is generalized to construct a hierarchical grammar semantic tree knowledge base; a knowledge base is used for generating a plurality of candidate patches for new vulnerabilities, single high-quality patches are generated through structure, grammar and semantic three-dimensional clustering and in-cluster integration and fusion, and the patches are output after heuristic sorting. According to the method, the defects that in the prior art, semantic understanding is shallow, and patch quality is unstable are overcome, and repairing accuracy, generalization ability and engineering practicability are remarkably improved.

The application discloses a vulnerability type adaptive program slicing method and device and electronic equipment, the method filters program statements in a patch through a large language model, identifies program statements related to a vulnerability, and accurately determines a vulnerability-related slicing standard; the large language model is used to analyze the patch description and the program statements, the type of the vulnerability repaired by the patch is judged, and a suitable slicing strategy is constructed according to different vulnerability characteristics; the large language model is used to judge the integrity of the sliced code segment, so that the sliced path explosion is prevented, and the high integrity of the sliced code segment is ensured. The application can achieve the purpose of adaptively realizing high-integrity program slicing according to the type of the vulnerability.

The application provides a computer network security software debugging method and system. A plurality of test nodes of a source program in target network security software are set, a key test node is located based on an execution state at each test node, and a target search domain when the source program is tested is constructed by program context information of the key test node. A plurality of program slices with abnormalities in the source program are determined based on a static control flow graph and a dynamic control flow graph of program code in the target search domain. Test overhead of each program slice is determined according to test coverage information of each program slice and path complexity when the source program is tested. Constraint levels of each program slice are determined according to all test overhead and membership between each program slice. Each program slice is debugged through the constraint level of each program slice. The above scheme can accurately point to a problem code line in a network security software debugging process based on constraint levels of each program slice.

The invention relates to the technical field of code similarity detection, and discloses a code similarity dynamic detection method based on ensemble learning and program slicing, which comprises the following steps: generating an abstract syntax tree and an original code attribute graph of each source code; obtaining grammar, data and control information of each node, and constructing an extended code attribute graph of each source code; performing program slicing on the extended code attribute graph to generate program slicing sub-graphs, and standardizing node names of the program slicing sub-graphs through an automatic mapping method based on semantics; inputting a standard program slice sub-graph into a graph neural network to capture graph features, and inputting the abstract syntax tree into a Transform encoder to extract sequence features; and finally, fusing and integrating the two generated features and carrying out a similarity detection task. According to the method, high-precision code classification and efficient semantic analysis in a cross-language scene are realized.

The invention provides a smart contract code auditing method fusing program slices and RAG, and relates to the technical field of smart contracts. According to the method, a verified vulnerability contract is obtained through a vulnerability knowledge base construction stage, and a labeling data set covering n types of OWASP standard vulnerabilities is constructed to form a vulnerability knowledge base as an RAG database; in the online detection stage, firstly, a to-be-detected smart contract is dynamically sliced, and a three-dimensional feature vector of a code snippet of the to-be-detected smart contract is obtained; then adopting a hierarchical matching strategy, and using all features of the to-be-detected smart contract code snippets to perform retrieval recall in the RAG database; and finally, designing a chained reasoning framework to deeply fuse the matching result of the large language model LLM and code analysis, and giving the result to the LLM for reasoning to realize vulnerability detection. According to the method, a three-stage processing flow of retrieval, filtering and refinement is constructed in a detection stage, so that the detection precision and the calculation overhead are effectively balanced.

The invention discloses a Rust program slicing method and system based on a high-order function, and relates to the technical field of functional programming and program analysis. The method comprises the steps of program preprocessing; analyzing in the process; inter-process analysis and program slice calculation. According to the method, slice information in the process is obtained by utilizing high-order function characteristics, and the function abstract expressed in the high-order function form is generated, so that function parameters in the function abstract can be conveniently substituted at a calling point as required during inter-process calling, and a corresponding relation between the function parameters and actual parameters can be quickly obtained; compared with a traditional abstract-based analysis method, more information can be carried in the parameter transmission process, so that the inter-process slicing efficiency can be effectively improved, various hypergraphs do not need to be constructed, the problem of low analysis efficiency caused by huge graphs and complex composition process is avoided, and the analysis efficiency is improved. Therefore, the space overhead in the analysis process can be effectively reduced, and the inter-process program slicing efficiency is improved on the premise that the detection precision is not lost.

This invention discloses a statement-level code vulnerability detection method based on fused feature prompt tuning. First, static analysis tools are used to parse source code, and program slicing technology is used to generate sliced ​​code segments. Statement-level hard and soft prompts are then combined to form fused feature prompts. Slice features and statement features are then fused and embedded into a model. Finally, statement-level vulnerability detection is achieved using fused feature prompt tuning and a pre-trained model. This method designs prompt templates based on the characteristics of vulnerable statements, fully leveraging pre-trained knowledge to learn the structure and attribute information of vulnerable code. This allows the model to more accurately distinguish between vulnerable and non-vulnerable statements, reducing the likelihood of false positives. Furthermore, this method can fully leverage the pre-trained model's ability to understand context, capturing subtle grammatical and semantic information in the code.

The invention discloses a smart contract vulnerability detection method based on a slice contract attribute graph, which comprises the following steps: firstly, obtaining a vulnerability detection sample set, respectively constructing a slice contract attribute graph for a smart contract in each vulnerability detection sample in the vulnerability detection sample set, and taking the slice contract attribute graph corresponding to each smart contract as input; the label corresponding to the smart contract serves as expected output, a training sample is formed, a vulnerability detection model is trained, when vulnerability detection needs to be conducted on a certain smart contract, a slice contract attribute graph of the smart contract is constructed, the slice contract attribute graph is input into the trained vulnerability detection model, and a detection result whether vulnerability exists or not is obtained. According to the method, a program slicing technology and a bidirectional sliding window are adopted to add dependent edges in a control flow graph and delete irrelevant nodes to generate a slice contract attribute graph, so that the accuracy and robustness of intelligent contract vulnerability detection are improved.

The application provides a vulnerability false alarm elimination method for a continuous integration platform, which captures vulnerability warning code features generated by program static analysis through a deep learning neural network, sorts and classifies the warnings, thereby eliminating false alarms in the warnings, and updates and optimizes the model using incremental warning data generated in the project continuous integration process. The steps include: obtaining historical warning data and corresponding warning label data of a Java project on a continuous integration platform; performing program slicing, tokenization and vectorization on the program corresponding to the warning; training a deep learning neural network to obtain a vulnerability false alarm elimination model; and adding the model to the continuous integration process of the project to enable the model to be continuously updated and optimized with project iterations. The application reduces the time cost of developers checking program static analysis vulnerability warnings in the project continuous integration process and improves the efficiency of processing warnings.

This invention belongs to the field of code security detection and automated repair technology. Specifically, it is a closed-loop system for automated repair and verification of code vulnerabilities based on unit tests. It completes vulnerability root cause localization and feature extraction through program slicing and static analysis. Then, based on the vulnerability root cause, it generates reproducible test cases with multi-dimensional assertions and regression test cases that meet coverage standards. Candidate patches are generated through retrieval enhancement and invalid patches are filtered out by static pre-verification. Patch verification is completed in an isolated sandbox using core and auxiliary dual-dimensional rules. Failure results are classified into scenarios and structured semantic feedback is generated to guide iterative repair. Finally, successful cases are structured and stored in a knowledge base in the form of triples to achieve self-evolution. This invention solves the core problems of existing technologies, such as unstable reproducible test cases, low iterative convergence efficiency, inability to identify overfitted patches, and lack of self-evolution capabilities. It can achieve fully automated repair and verification of code vulnerabilities, meeting the requirements for production environment application.

Embodiments of the present application provide a chip program debugging method and system without passive code and a medium. The method comprises: obtaining a chip program, slicing the chip program based on a program slicing technology to obtain a plurality of program slices; receiving a debugging instruction, analyzing function information of the debugging instruction based on the debugging instruction, and executing a corresponding debugging program according to the function information of the debugging instruction; linking the plurality of program slices to generate a plurality of debugging groups, performing program debugging on the plurality of debugging groups based on the debugging program, and generating a sub-debugging result corresponding to each debugging group; integrating the plurality of sub-debugging results to obtain a final debugging result, and saving the final debugging result to a register. The chip program is sliced by the program slicing technology, the plurality of program slices can be grouped and debugged or debugged in parallel, the debugging flexibility is improved, the number of program instructions analyzed by a developer is reduced, and the work efficiency is improved.

The embodiment of the invention provides an application program slicing processing method, terminal equipment and related products, and the application program slicing processing method comprises the following steps: in response to a slice establishment request initiated by an application program, determining whether the slice establishment request meets a preset slice establishment condition; when it is determined that the slice establishment request meets the preset slice establishment condition, the slice establishment request is sent to the network side device, and a response message returned by the network side device is obtained; judging whether the slice parameter information contained in the response message is matched with the slice requirement of the application program or not, and determining whether to trigger the network side equipment to execute slice establishment operation for the slice establishment request or not based on a judgment result. Through the embodiment of the invention, the risk of abnormal collapse of a chip and an operating system or network signaling storm caused by blindly initiating the session establishment request to the network by the terminal equipment can be avoided, the success rate of slice establishment is improved, and the application program obtains expected network performance and service quality.

The present application belongs to the technical field of software vulnerability mining, and specifically relates to a kind of Android application user privacy leakage vulnerability identification method.The present application includes: two-way program slicing analysis, privacy information identification based on correlation analysis and machine learning method assistance, and non-injective data transformation analysis.Two-way program slicing analysis is used to detect data received by Android application from the server end and judge whether they flow to user interface function;Privacy information identification based on correlation analysis and machine learning method assistance can comprehensively analyze which of the above server end returned information belongs to the privacy information of other users;Non-injective data transformation analysis is used to detect how user data is transformed on the client side before being presented to the user, and then judge whether the "data minimization" principle is followed and whether there is a privacy risk.

The invention provides a computer program vulnerability detection method and system, and the method comprises the following steps: obtaining a source code, converting the source code into a plurality of IR instructions, and selecting a plurality of to-be-recognized instructions from the IR instructions; obtaining a program slice based on the to-be-identified instruction, and obtaining a semantic vector set and a metric vector set of the program slice; constructing an instruction relation graph corresponding to the program slice, and obtaining a relation dependency vector of the instruction relation graph; and carrying out dimension unification on the semantic vector set, the metric vector set and the relationship dependence vector, further carrying out weighting fusion to obtain a final vector, and judging whether the program slices have vulnerabilities or not based on a neural network model and the final vector. By combining the semantic vector set, the metric vector set and the relation dependency vector into the final vector, point, line and plane three-dimensional detection of computer program vulnerabilities is realized, the comprehensiveness of computer program vulnerability detection is ensured, and the vulnerability detection accuracy is improved.

The invention provides an AI test case generation method, a server, a medium and a product. In the method, a server firstly collects code change data of a detected object, inputs the code change data into a correlation analysis model to obtain a current defect risk level, then establishes a mapping relation between the defect risk level and a generation control parameter, and determines a matched semantic similarity threshold value and a slice weight coefficient. And then, the server performs program slice analysis on the code change data to obtain candidate program slices, and calculates priorities by using slice weight coefficients and attribute characteristics of the candidate program slices to determine a target program slice. And finally, the server generates a test intention description based on the target program slice, maps the test intention description into a multi-dimensional intention semantic vector, and generates a test case code only when semantic repetition does not exist by calculating a size relationship between a vector similarity and a semantic similarity threshold. According to the method, the problem that the generated test suite contains a large number of low-efficiency cases is relieved, and the overall test efficiency is improved.

The application provides a computer program vulnerability detection method and system, the method comprises the following steps: obtaining source code and converting it into a plurality of IR instructions, selecting a plurality of to-be-identified instructions from the IR instructions; obtaining program slices based on the to-be-identified instructions, obtaining a semantic vector set and a metric vector set of the program slices; constructing an instruction relationship graph corresponding to the program slices and obtaining a relationship dependency vector of the instruction relationship graph; unifying the dimensions of the semantic vector set, the metric vector set and the relationship dependency vector, and then weighting and integrating them into a final vector; judging whether the program slices have vulnerabilities based on a neural network model and the final vector. By merging the semantic vector set, the metric vector set and the relationship dependency vector into the final vector, point, line and surface three-dimensional detection of computer program vulnerabilities is realized, the comprehensiveness of computer program vulnerability detection is ensured, and the accuracy of vulnerability detection is improved.

The application discloses a software test problem positioning and repairing method and device. The method comprises the following steps: collecting log data in a software test process, taking a log event as a node, taking a causal relationship, a time sequence relationship and a module correlation relationship between events as edges, and constructing a test log correlation graph; adopting a graph neural network to update and classify node features of the test log correlation graph, adding an abnormal propagation weight factor in a node updating formula, positioning an abnormal source node in the test log correlation graph, and taking the abnormal source node as a log event corresponding to a problem root cause; based on the abnormal source node, combining code execution paths and data dependency relationships to verify the root cause, extracting a code segment corresponding to the abnormal source node through a program slicing technology, and tracing input data and an intermediate calculation process corresponding to the abnormal source node through data dependency analysis, and confirming the accuracy of problem root cause positioning.

This invention relates to the field of computer technology, and more particularly to a method and computer device for performance fault diagnosis. The method includes: acquiring at least two program slices corresponding to a target program; acquiring first performance information corresponding to each program slice; classifying each program slice based on the first performance information; confirming the classification result; determining the target process in the target program where a fault has occurred based on the classification result and the first performance information; and determining the cause of the performance fault in the target program based on the faulty target process and a pre-constructed performance fault decomposition architecture. This method only requires capturing the function call process during program execution, and then analyzing and outputting results based on the captured results for the target program slices, greatly protecting the security and privacy of the tested program and significantly improving work efficiency.

The invention belongs to the technical field of computer network security, and particularly discloses a code detection method and device based on a multi-dimensional model, equipment and a storage medium. According to the method, the target program slice is generated according to the vulnerability concern; model input parameters are generated according to the target program slices, and new function codes are generated based on the multiple code large models; calculating text similarity according to a plurality of text similarity algorithms, and calculating structural similarity according to a plurality of structural similarity algorithms; and performing vulnerability detection on the to-be-detected code based on the target machine learning model. Through the above mode, the advantages of deep mining of inherent semantic and structural information of codes by the code large models and efficient identification and classification of the machine learning model are fully utilized, the new function codes are generated based on the multiple code large models, and then vulnerability detection is performed based on the target machine learning model. Therefore, the code detection efficiency and accuracy can be effectively improved, and the method is suitable for scenes of various programming languages and vulnerability types.

This disclosure provides a code verification method, apparatus, device, and system, relating to the field of computer technology. The method includes: running runtime data of a benchmark test program based on a reference model to determine a target program slice in the benchmark test program; wherein the reference model is a software model that simulates hardware description code, and the first runtime data corresponding to the target program slice satisfies preset runtime characteristics; running the target program slice through the hardware description code to obtain second runtime data; and determining the verification result of the hardware description code based on the first and second runtime data. This method can improve verification efficiency.

A source code vulnerability detection method based on a multi-relation graph, which first cuts the C language source code data set from the standard vulnerability database SARD and the NVD data set, generates an AST and obtains its program slice by means of Joern; secondly, in order to capture the additional syntax information of the AST, the AST is expanded by using a compiler to add edges carrying additional information to enhance its structure and connectivity, and then the enhanced AST is converted into a multi-relation graph; finally, a multi-relation graph attention network is constructed to learn an embedding vector. A plurality of code relations are extracted from the source code by using the graph attention network, and a global embedding vector is learned by using the adjacency matrix and the initial node representation of the multi-relation graph, and the global embedding vector is normalized and fed into a softmax layer for processing. By means of the newly designed representation mode and the multi-relation graph aggregation learning, the application can provide a better code representation for the downstream software vulnerability detection task, so as to realize an accurate, efficient and more applicable vulnerability detection method.

The invention relates to a malicious program detection method based on symbolized slices and a deep neural network, and belongs to the field of computer science and technology. According to the method, various key code segments in a malicious program are taken as research objects, and feature indexes of malicious codes are determined based on a weighted comprehensive multi-segment texture fingerprint similarity matching method and in combination with a gray-scale graph. The method comprises the following steps: analyzing a to-be-analyzed program by applying a symbolized slicing technology, determining an operation structure of the to-be-analyzed program, extracting a key code segment, extracting characteristics of the to-be-analyzed program according to a symbolized slicing result, and realizing visualization of a working principle of a malicious program. Through the method of combining the deep neural network and the program slicing, the detection precision and speed can be improved. Therefore, visualization of the working principle of malicious software is achieved, and certain convenience is brought to follow-up targeted work such as malicious program prevention and control and monitoring.

This invention discloses a method for detecting multi-attribute incremental models in synchronous language programs, comprising the following steps: S1, performing lexical and syntactic analysis on the synchronous language program to obtain an abstract syntax tree, and then transforming the complex program into a simple program containing only a unique node with a memory depth of 1; S2, using a program slicing algorithm to delete data streams and equations unrelated to the attributes to be verified, resulting in a simplified program; S3, constructing basic formulas for the bounded model detection engine thread and inductive formulas for the inductive engine thread; S4, executing the bounded model detection engine thread, the inductive engine thread, and the program abstraction engine thread in parallel, and controlling the synchronization between these threads through a message passing mechanism. This invention can accelerate the verification speed, thereby alleviating the problem of slow verification speed in existing model detection technologies.

This invention discloses a defect localization method for program slices in deep learning frameworks. The method includes: inputting the original failed test cases that caused the defect; extracting the target mutation point; designing mutation rules for deep learning frameworks; generating mutated test cases with execution paths highly similar to the original failed test cases but with erroneous statements removed; recursively selecting the most suitable mutation operator by comprehensively considering historical iteration data and the current test case state, gradually eliminating code elements containing errors; executing the original failed test cases and mutated test cases, collecting coverage information during execution; and using a spectrum-based defect localization algorithm to locate the defect based on the collected coverage information and the pass / fail status of the mutated test cases, outputting accurate defect localization results. This invention can quickly locate and repair defects in the code of program slices, effectively improving the overall reliability of deep learning frameworks.

The application provides a runtime exception detection method fusing slice testing and formal verification, comprising the following steps: adopting program slicing to extract a minimized code segment related to a potential abnormal program point from a program to be detected; generating a test case based on the minimized code segment related to the potential abnormal program point; and executing the minimized code segment containing the test case to obtain a detection result; the method can not only detect known exceptions, but also cover more potential exceptions through automatic path analysis, especially complex input scenarios and dynamic behavior triggered exceptions, and the scheme can identify abnormal paths in advance and reduce post-analysis after actual exception occurrence.