171 results about "Codebase" patented technology

Systems and processes are disclosed for enhancing cybersecurity and optimizing software repositories through integration of web crawling, web scraping, feature engineering, and advanced machine learning algorithms to detect phishing attempts, prevent account takeover fraud, and identify unused code in repositories. The system collects and refines data from various sources, including transaction logs, customer databases, device details, external data sources, and historical fraud data, to build comprehensive datasets. Feature engineering creates new, meaningful features from the refined data, which are used to train and evaluate machine learning models. The best-performing models are deployed in production to monitor incoming communications and transactions in real-time, flagging suspicious activities and optimizing codebases. This processing ensures timely detection and prevention of security threats while maintaining efficient software development processes. Robust protection is provided against evolving cyber threats and enhances software performance and security through continuous learning and adaptation.

The invention discloses a large language model reasoning acceleration method and device based on two-stage speculative decoding and a storage medium, and the method comprises the steps: constructing and initializing a Trie tree, and inserting a historical corpus, and phrase sequences in a document library or a code library into the Trie tree one by one; in the reasoning process, longest prefix matching is carried out based on a Trie tree, and a candidate draft sequence is generated by adopting branch backtracking and recursive search; performing confidence evaluation on the candidate draft sequence, calculating a joint confidence score of the sequence through probability multiplication and a Top-K screening mechanism, and judging whether the joint confidence score reaches a confidence threshold; if the accumulated confidence of the candidate sequence reaches a threshold value, skipping a small model generation stage, and directly entering large model verification; otherwise, entering a small model draft completion stage; and the final large model takes the replaced and updated draft sequence as final output. According to the method, adaptive acceleration of the decoding process can be realized, and the long text reasoning delay of the large language model is remarkably reduced while the generation quality is ensured.

The invention relates to the technical field of computers, in particular to a code sample optimization method and system based on a language large model, and the method comprises the steps: firstly, obtaining a source code sample, determining a code sample boundary, and generating a mark sequence; then, based on a compiling result, a static analysis result and a unit test result, a quality label is generated, and a training sample set is formed; then, taking a Transform neural network model of bidirectional context coding as an encoder, and carrying out training under the supervision of quality annotation to obtain a quality evaluation model; and finally, screening according to a preset quality score threshold to obtain a high-quality code sample set. According to the technical scheme, by combining the deep grammar analysis and context understanding of the code samples, the accuracy and efficiency of automatic screening are improved, manual intervention is reduced, and the method can adapt to automatic evaluation of a large-scale code library.

Techniques for implementing an AI threat modeling tool are disclosed. A static analysis tool is used to extract a candidate code snippet from a code repository. The candidate code snippet is identified as potentially being a security relevant code element. The static analysis tool generates additional context associated with the candidate code snippet. An LLM prompt is generated. This prompt is structured to include the candidate code snippet, the context, and a directive to assign a classification to the candidate code snippet. The classification includes a source classification, a sink classification, a sanitizer classification, or a flow step classification. The LLM operates on the prompt to generate output comprising a specific classification for the candidate code snippet. The output is formatted into a data extension file that is consumable by the static analysis tool.

The invention discloses a program defect repairing method and device, equipment and a storage medium. The method comprises the steps that a defect code segment is obtained from a first program code which is reported to be wrong, and a similar defect code segment and a similar repair patch corresponding to the defect code segment are inquired in a historical code library; constructing a first prompt word based on the defect code segment, the similar defect code segment and the similar repair patch, generating a repair patch based on the first prompt word, and repairing the first program code through the repair patch to obtain a second program code; and under the condition that the verification of the second program code is not passed, regenerating a repair patch based on a defect code segment appearing in the verification process of the second program code, and repairing the second program code through the newly generated repair patch. According to the method, the big language model is guided to generate the high-credibility repair patch through the historical similar defects and the repair patch thereof, and the problem that in the prior art, the repair patch is unreliable due to the fact that the big language model is separated from an actual repair case is solved.

The embodiment of the invention discloses an intelligent fuzzy testing method and device based on a large language model and a medium, and relates to the technical field of fuzzy testing, and the method comprises the steps: obtaining and analyzing a source code of a target code library to construct a structured knowledge graph, and evaluating a function vulnerability risk through the large language model and the structured knowledge graph, generating a test target sorting list; performing multi-dimensional code analysis on the selected test targets in the test target sorting list to generate seed input so as to construct a compilable test driver through the seed input; performing a test driver in an isolated environment, performing a fuzz test using a seed input, collecting a crash report to analyze the crash report, identify a crash cause, and generate a minimization replicator; and generating a candidate patch based on the crash reason, and verifying the validity of the candidate patch through a minimization reproducing device and a structured knowledge graph so as to realize the fuzzy test of the source code of the target code library.

In various examples, a technique for resolving a standards violation includes receiving a violation notification of a standards violation detected in a software codebase. The technique also includes determining additional information relevant to the standards violation and included in one or more information sources. The technique further includes generating a prompt based at least on the violation notification and the additional information, generating, using a machine learning model and based at least on the prompt, one or more corrective suggestions associated with the standards violation, and modifying the software codebase based at least on the one or more software code changes.

Systems, computer program products, and methods are described herein for configuring data using advanced computational models for data analysis and automated processing. The present disclosure is configured to train a large language model (LLM), wherein training the LLM comprises using system-specific data comprising feed data, process run logs, historical events, code base, existing permissions, and data classification rules. The present disclosure is configured to determine prone data, wherein the prone data comprises a log file comprising sensitive information, and wherein the prone data is determined via a prone module. The present disclosure is configured to configure the prone data using a generative artificial intelligence (GenAI) module, wherein the GenAI module configures the prone data by masking the sensitive information using a masking procedure. The present disclosure is configured to determine the masking procedure via a decentralized autonomous organization (DAO).

Assessing computer system vulnerabilities and exposures by periodically querying data sources to gather information pertaining to computing system vulnerabilities and exposures (CVEs), such as, for each CVE, an identification of the CVE, a number of corresponding references to the CVE, and a number of code repositories that can be used to exploit the CVE. Compiling a datastore of the information. Periodically querying the datastore about the information and generating one or more views of a lifecycle of each CVE in response thereto.

The application provides an adaptation method and device and a computer readable storage medium, relates to the technical field of application software, and can provide functional support for automatic code replacement execution and scheme between source adaptation content and target adaptation content. The method comprises the following steps: when an adaptation application request is submitted, target parameters are configured, including an adaptation environment, basic data, a code library address and a function list; in the case of completed configuration, the adaptation application request is submitted to an automatic adaptation service through a form; the code library address information is transmitted to a machine learning service through the automatic adaptation service, and machine learning training is performed through the machine learning service; the function set in the code file information and adaptation requirements are used for rule matching on an adaptation rule library; in the case of successful matching, a code replacement operation is performed on user code; and based on the file name and file position information of the replaced user code, a code submission request is triggered. The application is applied to automatic code replacement execution.

The application discloses a code library UML diagram automatic generation method and system based on cooperation of abstract syntax tree and large language model, comprising an AST analysis and structure extraction layer, which is used for being responsible for abstract syntax tree analysis on source files in the code library and extracting skeleton structure information of the code library, and a large language model semantic analysis layer, which is used for receiving structured code skeleton data output by the AST analysis layer and submitting the structured data to the large language model for semantic analysis through a carefully designed prompt word (Prompt), wherein the application extracts structured skeleton information of the code library through AST analysis instead of directly processing original source code, compresses the information volume to 10% to 20% of the original code amount, greatly improves the effective information density of the input large language model, enables a larger scale code library to be covered in a limited context window, and solves the problem that the existing large language model scheme cannot process a large scale code library due to Token waste.

1. The name of the design product: input box search mode switching display module of search interaction graphical user interface for electronic device. 2. The use of the design product: an electronic device. 3. The design points of the design product: the part in the graphical user interface, the part outside the dashed line is the claimed part. 4. The picture or photo that best shows the design points: design 1 front view. 5. Design 1 is designated as the basic design. 6. The use of the graphical user interface: for search interaction, the claimed part is used to display input box search mode switching. 7. The human-computer interaction mode of the graphical user interface: in design 1, the user can click the "search" / "code base" option at the bottom left of the input box interface in the front view of design 1 to switch the search mode, or click the "full network search" control behind the mode option to select whether to open or close Internet search. The human-computer interaction description of design 2 is the same as that of design 1. In design 3, the user can click the "search" / "knowledge base" / "code base" option at the bottom left of the input box interface in the front view of design 3 to switch the search mode, or click the "full network search" control behind the mode option to select whether to open or close Internet search. The human-computer interaction description of design 4 is the same as that of design 1. In design 5, the user can click the "search" / "code base" option at the bottom left of the input box interface in the front view of design 5 to switch the search mode, or click the "select code base" control behind the mode option to retract the expanded code base list. The user can also click any option in the code base list module expanded below the input box to perform the check operation, or click the "add warehouse" control in the upper right corner of the module to perform the corresponding operation. In design 6, the user can click the "search" / "knowledge base" / "code base" option at the bottom left of the input box interface in the front view of design 6 to switch the search mode, or click the "select code base" control behind the mode option to retract the expanded code base list. The user can also click any option in the code base list module expanded below the input box to perform the check operation, or click the "add warehouse" control in the upper right corner of the module to perform the corresponding operation. In design 7, the user can click the "search" / "knowledge base" / "code base" option at the bottom left of the input box interface in the front view of design 7 to switch the search mode, or click the cross control behind the selected knowledge base name behind the mode option to delete the selected option. The user can also click any option in the knowledge base list module unfolded below the input box to perform a check operation, or can click the "add knowledge base" / "clear options" control in the upper right corner of the module to perform the corresponding operation. The human-computer interaction description of design 8 is the same as that of design 7. In design 9, when the user clicks the "code base" option at the bottom left of the input box of the main view interface of design 9, the code base list is unfolded below the input box, and the interface change from the main view of design 9 to the interface change state diagram of design 9 is presented. In design 10, when the user clicks the "knowledge base" option at the bottom left of the input box of the main view interface of design 10, the code base list is unfolded below the input box, and the interface change from the main view of design 10 to the interface change state diagram of design 10 is presented.

A system includes a memory configured to store a software codebase and an instance of vulnerabilities associated with the software codebase. The system further includes a processor coupled to the memory and configured to access the software codebase and execute a scan to identify sets of metadata associated with the software codebase. The processor is configured to execute a first machine-learning model trained to generate a prediction of feature clusters based on the sets of metadata. The plurality of feature clusters includes a ratio of a vulnerability to a time-intensiveness associated with remediating the vulnerability. The processor is configured to execute a second machine-learning model trained to generate a prediction of associations between the plurality of feature clusters and the instance of vulnerabilities. The prediction of associations includes an estimate of a time-intensiveness for remediating the instance of vulnerabilities.

The application discloses a business code processing method and device, a storage medium and an electronic equipment. It relates to the field of financial technology, and the method comprises the following steps: in the case that a business code repository is detected to have a business code modification event, a target code repository is matched according to a code file name corresponding to the business code modification event; in the case that a code class identical to the code file name is not matched, an attribute tree of a first code class and attribute trees of a plurality of second code classes are respectively constructed; similarity between the first code class and each second code class is calculated according to the attribute tree of the first code class and the attribute trees of the plurality of second code classes, and a target code class is determined from the plurality of second code classes according to the similarity. Through the application, the problem that the processing accuracy of business code is low when the code repository is adaptively modified in the related art is solved.

In some implementations, a system for automatically enforcing review rules within a code repository may receive, from a developer device, a pull request for a code branch. The system may perform a first check to determine whether the code branch has at least one review, in the code repository, associated with the code branch. The system may perform a second check to determine whether the at least one review is associated with a non-contributing user for the code branch. The system may perform a third check to determine whether the non-contributing user is associated with a same organization as the code branch. The system may selectively execute the pull request based on results from the first check, the second check, and the third check

The invention discloses an automobile software code change influence analysis method and system, and relates to the field of software development, and the method comprises the steps: S1, constructing a multi-source unified vector enhancement knowledge graph data index; comprising the following steps: respectively executing data extraction and analysis on a code library, an ARXML file and a demand and defect document, and establishing a corresponding index; wherein a unified atlas data model is established based on the extracted entities and relationships, a semantic vector is generated for each entity node, and fusion of the atlas and the vectors is realized; s2, multi-path influence propagation analysis is carried out; comprising the following steps: analyzing code difference of a version control system, and positioning an initial change code node set; influence propagation analysis is carried out through four influence paths including grammar / structure dependence, semantic similarity, interface contract and demand / defect tracing; the analysis process of the four influence paths is executed by calling an influence propagation engine by an analysis path planning engine; s3, carrying out backtracking and iterative analysis; and S4, generating a structured analysis context.

The invention discloses a software function deep analysis method based on large model driving, and relates to the technical field of software engineering.The software function deep analysis method comprises the steps that multi-modal data of target software is collected, and the multi-modal data at least comprises code library data, demand document data, user interaction logs and test case data; and performing cross-modal alignment processing on the acquired multi-modal data, mapping the data from different modals to a unified vector space, and constructing a multi-dimensional tensor sequence for representing a software state. According to the method, a layered virtualization verification environment comprising a system layer, a business logic layer and a data layer is adopted, a safe, isolated and approximate-to-reality execution platform is provided for the automatically generated test case, and real-time monitoring and verification of software function implementation conditions are supported, so that the test influence on a production environment is reduced, and the test efficiency is improved. And the test feedback cycle is accelerated, and the flexibility and the result reliability of the software quality verification process are comprehensively improved.

The invention relates to a code generation method and system based on a large language model, and belongs to the technical field of computers. According to the method, dialogue is carried out on the basis of the generated first version code and the large language model to provide more detailed and specific input, so that the large language model generates a corresponding subprogram, and complete code logic is constructed; calling an example code stored in an example code library to further optimize the initial code based on multiple rounds of conversations, and generating a second version of code; and performing comprehensive quality evaluation on the generated second version code based on a preset code quality evaluation standard, automatically generating a repair rule based on an evaluation result, and generating a target code by using the repair rule. According to the method and the device, the repair rule is obtained according to the evaluation result obtained by performing comprehensive quality evaluation on the second-version code, the second-version code is repaired based on the repair, so that the high-quality target code is obtained, and the whole generation process is more intelligent.

The invention provides an artificial intelligence-based industrial design method and system, and the method comprises the steps: obtaining an open-source artificial intelligence large model, building a dialogue function with the artificial intelligence large model, carrying out the recording of a 3D design code for each part of a 3D model of an existing product based on 3D design software, building a 3D design code library, and carrying out the verification of the 3D design code library. The 3D design code library serves as a third-party database to be added to an existing database of the artificial intelligence large model, the product design requirement of a user is obtained based on the dialogue function, the artificial intelligence large model automatically generates a 3D design code of the product according to the product design requirement, and the 3D design software automatically generates a 3D model of the product according to the 3D design code; according to the method, the design code of the required product is automatically generated by combining the 3D design code of the existing product 3D model to optimize the database of the artificial intelligence large model, so that the 3D model of the required product is automatically generated, intelligent aided design is realized, the design work is simplified, and the design result is optimized.

In some implementations, a device may obtain a codebase, associated with an application and one or more services of the application, that is deployable via both a first framework of the cloud computing environment and a second framework of the cloud computing environment. The device may receive the traffic associated with the application. The device may monitor one or more metrics associated with the traffic. The device may deploy, based on the one or more metrics, the codebase via the first framework or the second framework in the cloud computing environment. The device may route, based on the one or more metrics, the traffic to the first framework or to the second framework.

The application discloses a defect code generation method and related device, and the method comprises the steps of: obtaining at least one code segment pair from a code library; analyzing the modification reason between the first code segment before modification and the second code segment after modification in each code segment pair through a first language model, and identifying the defect type of the first code segment in each code segment pair indicating defect repair; obtaining at least one code segment pair corresponding to each defect type as a target code segment pair, and calling a second language model to learn each target code segment pair with the generation of defect code as a learning target to obtain a learned second language model; and generating a defect code segment corresponding to the target code segment through the learned second language model. The method can realize the automatic generation of defect code, improve the generation efficiency of defect code, improve the generation quality of defect code, and enhance the comprehensiveness and authenticity of the generated defect code.

The invention belongs to the technical field of natural language processing and information retrieval, and relates to an intelligent question answering method based on structured semantic indexing and double-layer memory enhancement. The method comprises six steps of query preprocessing, Agent-based multi-tool dynamic routing, structured and semantic enhanced index construction, self-adaptive context assembly, double-layer memory management, and data closed loop and self-evolution, a differential segmentation strategy is adopted for texts, codes and multi-modal data, query optimization, intelligent tool routing and a mixed retrieval algorithm are combined, and the multi-modal data are subjected to self-adaptive context assembly. The technical defects of semantic rupture, low retrieval accuracy, no long-term memory and lack of self-optimization of an existing RAG system are overcome. According to the method, the code retrieval accuracy and the context utilization rate are improved, personalized long-term service and automatic operation and maintenance are achieved, and the method is suitable for scenes such as technology research and development, code library maintenance and intelligent question and answer.

Disclosed herein are system, apparatus, article of manufacture, method, and / or computer program product embodiments, and / or combinations and sub-combinations thereof, for code migration and deployment in an Internet-of-Things (IoT) environment. A system server (e.g., a server of a cloud-based platform, etc.) may receive a codebase and operational information for a target device (e.g., an IoT device, a mobile device, a smart device, etc.). The codebase may be modified for compatibility with the target device based on functionality for libraries of the codebase mapped to functional elements that define the functional capabilities of the target device. A codebase migration window may be identified based on an indication that an operation of the target device satisfies an operational threshold and the operational information received from the target device. The modified codebase may be transferred to the target device during the codebase migration window.

A method and system of testing a fine-tuned LLM for domain specific code generation is disclosed. Further, a processor receives a test dataset corresponding to a domain from a code repository. Further, the processor determines an LLM generated problem statement corresponding to the test code using the fine-tuned LLM. The fine-tuned LLM is fine-tuned based on a training dataset. Further, the fine-tuned LLM is prompted based on the LLM generated problem statement to determine an LLM generated code for a corresponding test function. The accuracy level of the fine-tuned LLM is determined based on a percentage match between the LLM generated code with the test code for each of the set of test functions.

Presented herein are systems and methods of automatically generating files for defining computing clusters. A server may maintain a code base including a plurality of files to define an infrastructure of a computing cluster. The server may identify a first file including a first plurality of parameters for configuring the computing cluster. The server may generate, using the first file and the code base, a second plurality of parameters to define a corresponding plurality of resources for creation of the computing cluster. The server may create a second file to define the infrastructure of the computing cluster, using the first plurality of parameters and the second plurality of parameters. The server may establish the infrastructure of the computing cluster in accordance with the second file.

Techniques are described herein that are capable of responding to a query in a developer tool using semantically related keywords in relevant code chunks. A user-generated query regarding a location of an element in a codebase of a software development project is received. The codebase is parsed into code chunks. Semantically related keywords, including keywords from the user-generated query and other keywords that are semantically related to the keywords, are identified. Relevant code chunks are selected from the code chunks based on satisfaction of a relevancy criterion regarding the user-generated query. Execution of an instruction is triggered, which causes a visual representation of a response to the user-generated query to be generated. The execution of the instruction causes the visual representation to include at least portions of the relevant code chunks and further causes at least a subset of the semantically related keywords to be highlighted in the portions.

Embodiments of this disclosure provide, but are not limited to, solutions for protecting, sharing, verifying, and / or generating data resources such as documents. In exemplary embodiments, a data resource is represented, expressed, or defined as a plurality of hashed constituent segments. Using the segment hashes as leaves, a Merkle tree is generated representing the complete original version of the data resource. An edited version of the data resource can be shared, omitting one or more segments that the data administrator wishes to keep secret or does not wish to share in its original form. To verify the authenticity and / or completeness of the edited data resource, a validator is provided with sufficient data (e.g., the hash of each of the non-shared segments) to verify the presence of non-shared segments in the tree representing the complete original version, and an authorized signature that signed the root. In some embodiments, the data resource is a code library, and the segments are parts of computer code that perform specific subtasks. Shared code segments, when executed on a processor, may be selected to combine to perform larger tasks.

This application discloses a method, medium, and program product for automatically modifying test scripts, relating to the field of automated testing technology. By automating log analysis, script modification, and script verification for automated testing, it saves the human resource costs associated with these tasks. The method includes: identifying the executed test script and generating an error log; retrieving the test script from the code repository and downloading it to a verification directory; polling historical log information in the log database based on the script name corresponding to the error log to obtain the execution success log corresponding to the script name; retrieving the test script corresponding to the execution success log from the code repository; modifying the test script in the verification directory based on the test script corresponding to the execution success log; and verifying the modified test script in the verification directory; after successful verification, submitting the modified test script in the verification directory to the code repository.

Provided in the embodiments of the present specification are a code completion method and apparatus. The code completion method comprises: acquiring a current code snippet to be completed from a current code file; querying a code relationship graph for a target node matching the current code snippet, wherein the code relationship graph is constructed on the basis of a plurality of code files in a code library, the plurality of code files comprise code files other than the current code file, individual nodes in the code relationship graph represent code entities corresponding to code snippets, and connecting edges between nodes represent semantic associations between the code entities; on the basis of first code snippets corresponding to neighbor nodes of the target node, determining the code context of the current code snippet; and inputting the current code snippet and the code context into a code large language model, so that the code large language model generates a completion result for the current code snippet on the basis of the code context.