11953 results about "Exploit" patented technology

A system operating with a cybersecurity system to enhance cyber threat detection is described. The system features a first and second orchestrator modules. The first orchestrator module includes at least a first large language model and is configured to perform artificial intelligence-based simulations of cyber-attacks to determine (i) how a simulated cyber-attack might occur in a selected computing device and (ii) how to use simulated cyber-attack information to preempt possible escalations of an ongoing actual cyber-attack. The second orchestrator module includes at least a second large language model and is configured to (i) perform a remediation task to correct one or more misconfigurations in one or more components associated with the cybersecurity system and (ii) return the one or more components back to a trusted operational state.

An artificial intelligence system can be used to respond to natural language inputs. The AI System may, for example, receive a first user input for a LLM, generate a first prompt based on the first user input, transmit the first prompt to an LLM, receive an output from the LLM, and evaluate the output from the LLM with reference to one or more validation tests. Responsive to determining that the output from the LLM is not validated, generate a second prompt for the LLM, where the second prompt indicates at least an aspect of the output that caused the output to not be evaluated (e.g., a portion of the output that may need to be updated or corrected), transmit the second prompt to the LLM, and receive an updated output from the LLM. The AI system can include an application for testing functions that utilize interactions with language models.

A vulnerability detection and repair system utilize a classifier model to detect a software vulnerability in a source code snippet and the tokens in the source code snippet attributable to the vulnerability. A large language model is then given the vulnerable source code snippet, its vulnerability type, the vulnerability tokens, and a few-shot examples to determine whether or not the source code snippet includes the identified vulnerability. The few-shot examples include positive and negative samples of the type of vulnerability to guide the large language model towards the correct output.

The invention provides a large language model-driven code review and optimization method, which belongs to the field of computer science, and comprises the following steps of: firstly, performing intensive training on a model through rich code samples and accurate manual annotation to ensure that the model has accurate code understanding and analysis capability; and then, performing deep analysis on an input source code by utilizing the trained large language model, and constructing an abstract syntax tree (AST) to capture deep semantic information of the code. And finally, the model automatically analyzes the codes, identifies potential defects, non-standard coding practices and optimization possibilities, and generates specific optimization suggestions, including code repair schemes and guidance following optimal practices, according to identification results, so that efficient support is provided for developers, and the quality of software products is ensured.

The invention relates to the technical field of data processing, in particular to an intelligent document updating processing method and system.The intelligent document updating processing method comprises the steps that part parameters and coding rules are extracted from engineering design file metadata, a part list and a PDM and PLM system, the semantic association relation between parts is analyzed through a knowledge graph, and the mapping relation between logic identifiers and physical files is constructed; generating an initial version file library; and monitoring file names and version changes in real time based on a micro-service architecture, calling a simulation service in combination with a knowledge graph to verify parameter compatibility, screening an optimal version, updating the optimal version to a main version library, optimizing historical conflict data in a block chain evidence storage index by using a genetic algorithm to generate a standardized code, and outputting a cross-platform parameter mapping table. According to the method, the problems of file name change, non-standard coding conflict, version control missing and cross-platform adaptation incoherence are solved, and the data tracing efficiency, the version management reliability and the system compatibility are improved.

The invention relates to the technical field of server scheduling, and discloses a large model reasoning scheduling method based on an off-network computing power server, which comprises the following steps of: collecting hardware information of the off-network computing power server, and constructing a server resource vector; decomposing a large model reasoning task into a plurality of sub-tasks, and constructing task modeling; according to the hardware information of the off-network server and the task dependency relationship, performing task division by using an optimization method; performing dynamic task scheduling on the sub-tasks based on the current load state of the server; according to the real-time calculation load of the server, dynamically migrating part of tasks to other servers to ensure the load balance among the servers; and executing the distributed task, monitoring the running state of the server and the task completion condition, and dynamically adjusting a task distribution strategy according to feedback data. Through modeling task requirements and server resources, in combination with real-time monitoring and optimal scheduling, efficient calculation and resource utilization maximization of large model reasoning in an off-network environment are realized.

The present invention discloses a system and method for performing artificial intelligence (AI) safety red-teaming with integrated policy fuzzing and adversarial prompting to systematically identify, characterize, and mitigate unsafe or non-compliant behaviors in AI models. The disclosed invention automates the process of generating, executing, and analyzing adversarial test cases through coordinated functional units comprising a policy fuzzing unit, an adversarial prompting unit, an execution sandbox, a telemetry processing unit, a scoring and triage processor, and a cryptographic provenance processor. The system applies grammar-driven and reinforcement-based fuzzing techniques to vary policy descriptors, model configuration parameters, and instruction hierarchies, while a learned adversarial prompt generator synthesizes contextually coherent adversarial prompts optimized for maximum policy violation likelihood. The generated prompts and policy vectors are executed in an isolated, instrumented sandbox that records input-output interactions, timing characteristics, and intermediate representations.

The embodiment of the invention provides an information generation method and device, electronic equipment and a storage medium. The information generation method comprises the following steps: in response to an obtained question and user information, calling a first agent to determine a target intention based on the question and the user information; a second agent is called to determine a to-be-executed task corresponding to the target intention, the second agent is called to execute the to-be-executed task to generate target information, the target information serves as reply information of the target intention to be sent to the client side, and the first agent and the second agent are responsible for intention confirmation and task planning respectively. According to the technical scheme, the intelligent agent has the learning ability and can make corresponding actions based on the change of the environment, so that the target intention can be determined according to the problem of the user and the experience of similar problems appearing before, the to-be-executed task is further determined, the efficiency of processing the user problem is finally improved, and the user experience is improved.

The invention discloses a computing power scheduling method and system based on dynamic load prediction and resource priority ranking. The computing power scheduling method comprises the following steps: collecting historical load data, task submission data and resource state data of each node in a computing power cluster; on the basis of the preprocessed multi-dimensional load feature data set, constructing an improved hybrid prediction model, optimizing model parameters through training, and predicting the load change trend of each computing power node in a future preset time period by using the trained model to obtain a node load prediction result; extracting a service level protocol parameter, a resource demand type and historical execution efficiency data of a to-be-scheduled task, and establishing a multi-dimensional resource priority evaluation index system; according to the computing power scheduling method, the problems of low resource utilization rate and high task response delay caused by low load prediction precision and mismatching of resource allocation and task priority in a traditional computing power scheduling method are solved, and the overall operation efficiency and service quality of a computing power cluster are improved.

High-level synthesis (HLS) is a complex design flow that exploits the benefit of modern language features such as loops, inheritance, templates, etc. for rapid prototyping of hardware designs. Hardware engineers explore various design space parameters in a time-consuming trial-and-error approach to meet a target design specification. This invention provides a framework where machine learning methods are integrated with Bayesian optimization (BO) to accelerate hardware design in HLS. This tool is built around the context of HLS pragma exploration and kernel transformation. It can produce best set of pragma installments to determine parameters by optimizing on multi-objective functions, such as reduction of digital signal processor, flip-flop, look-up table, power consumption, and latency. The method uses classical deep neural network and quantum neural network for reinforcement meta learning to predict the synthesis characteristics and synthesizability under a given time budget. A large language model can also assist transpilation to modify the original codes so that the best tradeoff between hardware complexity, latency, and power consumption is realized.

The invention relates to the technical field of front-end resources, in particular to an intelligent caching method and device for loading front-end resources on demand, equipment and a medium, and the method comprises the following steps: obtaining user behavior data, predicting an access page, identifying required resources, and constructing a resource dependency graph; calculating resource use frequency and importance, and determining a target resource; selecting a resource format and a compression mode according to the user equipment type; according to the resource dependency relationship graph, loading and caching target resources required by a next page possibly accessed by the user in advance; meanwhile, the caching strategy is dynamically adjusted based on the resource use frequency and the importance index; the method comprises the following steps: monitoring a network condition, dynamically adjusting a resource loading strategy, pre-loading resources before a user requests by using a resource dependency graph and a server pushing technology, and determining a target loading mode when the user accesses. Therefore, the problems that in the prior art, resource loading is low in efficiency, a caching strategy is rigid, environment adaptability is poor, user behavior analysis is insufficient, and performance optimization automation is lack are solved.

A process scheduling method and apparatus, a device and a storage medium, relating to the technical field of production scheduling. The method comprises: receiving a process scheduling request (S201); in response to the process scheduling request, determining an instruction being executed, an instruction to be executed, and scheduling resource information (S202); establishing an objective function and a constraint condition on the basis of the instruction being executed, the instruction to be executed, the scheduling resource information and corresponding actual scheduling information (S203); optimizing the objective function under the constraint condition by means of a swarm intelligence optimization algorithm to obtain an optimal solution of the objective function (S204); and using the optimal solution to determine an instruction execution policy, so as to perform process scheduling (S205). According to the process scheduling method, scheduling resources such as instruments and devices can be fully utilized, and the continuously growing demand for production scheduling is met.

Methods and systems for assessing a software bill of materials (SBOM) include building a knowledge graph from repositories, using function fingerprints of software packages in the repositories. Dependencies of an application are identified using function fingerprints of the application and comparing to function fingerprints of the software packages. A quality score for an SBOM of the application is generated based on a comparison of the identified dependencies to claimed dependencies of the SBOM. An action is performed responsive to the quality score.

The invention discloses a memory fault repairing method and device, equipment, a medium and a computer program product, and relates to the technical field of computers.The memory fault repairing method includes the steps that memory error information sent by a memory controller is obtained, a fault target memory page can be positioned, and a standby memory page is obtained from a standby memory pool; the memory address mapping table is updated, the physical address of the target memory page with the fault is mapped to the physical address of the standby memory page, the repairing process does not depend on triggering of a system management interrupt mechanism, memory repairing in the system running stage is achieved, the system downtime caused by memory fault repairing is shortened, and therefore the memory repairing efficiency is improved. The problem of high delay in memory fault processing can be solved, and the technical effect of improving the memory fault repairing efficiency is achieved.

The invention provides a security data penetration test method. The method comprises the steps of pre-establishing a multi-thread application program race condition vulnerability feature library, performing static analysis on source codes of a target application program, identifying potential race condition vulnerability points, and obtaining a preliminary vulnerability candidate set; for monitored suspected race condition events, abstracting event context information into constraint conditions, exploring an input combination of suspected static conditions by using a constraint solver, and determining a minimum input set of stable triggering race conditions; aiming at memory models and synchronous primitives of different programming languages, obtaining a race condition inverse mode, and extracting universal utilization primitives covering various race condition vulnerabilities; in the penetration test process, static analysis, dynamic monitoring and symbolic execution are applied to mine race condition vulnerabilities of a target application system, the availability of the vulnerabilities is verified, and evaluation of system security is assisted to be completed.

The invention relates to the technical field of hard disk testing and verification, in particular to a multi-thread-based solid state disk reliability automatic testing method and system.The method comprises the steps that firstly, SMART information is deeply analyzed through microsecond-level high-granularity continuous performance monitoring, and multi-thread parallel processing is assisted; according to the method, fine performance fluctuation of the solid state disk under the concurrent load can be quickly captured, a fault mode can be identified, then early warning is realized by utilizing the extracted multi-dimensional features and a machine learning model, and a detailed fault diagnosis report is generated; and through dynamic error correction code strength verification and data integrity verification under pressure, an internal error correction mechanism of the solid state disk is actively detected and optimized. And finally, in combination with prediction reliability modeling, the system can estimate the remaining service life and predict faults, and provides product optimization suggestions for design, manufacturing and firmware optimization of the solid state disk, so that automation, intelligence and full life cycle management of the fault detection reliability of the solid state disk are realized.

The embodiments of the present application disclose a speculative decoding optimization method, an electronic device and a storage medium, wherein a speculative decoding optimization method is used for a draft model and a target verification model, the method comprising: inputting the draft content generated by the draft model in the previous round into the target verification model; performing parallel verification on the tokens in the draft content in the target verification model; using a lightweight draft length predictor to analyze the latent vector of the last received token position and the last token generated, and predicting the optimal draft length for the next round; and autoregressively generating a corresponding number of drafts based on the optimal draft length predicted by the lightweight draft length predictor through the draft model.

Systems, methods, and computer-readable media for workflow management operations are disclosed. A template selection is received for a predefined workflow to be implemented in an existing user's account having at least one internal resource, and a selection of a set of resources is received from a plurality of resources to implement the functionality of the predefined workflow. The selected set includes at least one internal resource of the existing user's account and at least one external resource accessible via an API. A secure storage area is queried to determine that credentials are non-existent for the at least one external resource, an authentication flow is initiated with the at least one external resource to obtain a token from an external resource, and the token is received. The token is stored in the secure storage area, and a configuration of the predefined workflow initiates functionality of the predefined workflow.

The invention relates to a fuzzy testing method, device and equipment based on a large language model, and the method comprises the steps: analyzing and analyzing RFC standard documents of various network protocols through the large language model, and extracting protocol rules and constraint conditions; according to the protocol rule and the constraint condition, a large language model driving method is adopted to generate a test case conforming to protocol characteristics; and performing a differential test through the test case, and verifying the consistency between the actual software of the network protocol and the RFC standard document. According to the invention, the accuracy and security of implementation of the complex network protocol can be improved, and the test efficiency and coverage rate are optimized.

Accurate prompting improves the operation, efficiency, and output of computer models, such as language models. The disclosed systems and methods improve interaction with computer models by facilitating the generation of accurate prompts for targeted interactions with computer models. For example, disclosed systems can be configured to store prompts and responses generated with computer models in session memories. The system can use information in session memories to generate combined prompts when receiving prompts that refer to previous prompts or answers. The system improves prompt accuracy by generating combined prompts—formed by combining the context from the stored information (e.g., in context sub-prompts) and instructions in the new prompt (e.g., in instructions sub-prompts). The system can generate responses based on the combined prompts allowing the computer models to leverage the context in previous interactions, without burdensome or complicated prompts, for more meaningful or accurate responses.

The invention provides a supercomputing resource intelligent allocation method and system based on AI load monitoring, and the method comprises the steps: firstly obtaining a real-time load data set of a plurality of computing nodes in a supercomputing system, including a resource occupancy rate index and a task queue state parameter, then carrying out the load feature extraction processing of the real-time load data set, generating resource occupancy fluctuation characteristics and task queue evolution characteristics, performing resource demand analysis on the load characteristic set by using a pre-constructed resource demand prediction model, predicting resource demands of each computing node in a subsequent time window, determining resource allocation priorities and migration strategy parameters based on prediction results, and performing resource allocation according to the resource allocation priorities and the migration strategy parameters. The resource dynamic allocation instruction is generated and sent to the target computing node, the resource reallocation operation is triggered, intelligent and efficient allocation of super computing system resources is achieved, the resource utilization rate and the task execution efficiency are improved, and the overall energy efficiency ratio and stability of the system are optimized.

The embodiment of the invention discloses a causal chain-based front-end state anomaly detection method and system. The method comprises the following steps: S1, identifying a calling relation among an event monitor, an asynchronous task and a state change logic by analyzing a source code, and constructing an expected causal chain model; s2, dynamically verifying whether the actually occurring causal chain conforms to an expected model or not during the application running period; wherein in the step S1, the analysis of the source code comprises but is not limited to an AST technology; in the step S2, the hijacking mode of the key function comprises but is not limited to a Proxy mechanism, function packaging and a Hook technology. According to the method, the expected causal chain model is constructed, the actual path is dynamically verified during operation, system inconsistency caused by causal chain abnormity is avoided, the AI model is utilized to generate the repair strategy suggestions, a traditional low-efficiency troubleshooting process depending on logs and breakpoint debugging is upgraded to intelligent decision, and the repair efficiency is remarkably improved.

Systems and methods for detecting and mitigating prompt injection attacks on a generative LLM are disclosed. A deployment scenario is considered, in which the generative LLM supports a task automation function. Prompts are received and interpreted by the generative LLM, and outputs from the generative LLM are used to trigger automation actions. The prompts are constructed based on a combination of user input and external data and are, therefore, vulnerable to prompt injection attacks though manipulation of the external data. To mitigate this risk, a separate discriminative classification, decoupled from the generative LLM, engine is configured to identify malicious prompts, and filter out any malicious prompts before they reach the generative LLM.

A method, computer program product, and computing system for processing workload data associated with processing a plurality of requests for an artificial intelligence (AI) model on a processing unit. A maximum number of key-value (KV) cache blocks available for the workload data is determined by simulating the workload data using a simulation engine. A token utilization for the workload data is determined based upon, at least in part, the maximum number of KV cache blocks available for the workload data. Processing unit resources are allocated for the processing unit based upon, at least in part, the token utilization.

The invention discloses a software test case generation method and system based on a large language model, and relates to the technical field of software testing, and the system comprises a to-be-tested module detection subsystem which is used for constructing a to-be-tested module set; the test case generation subsystem is used for constructing a prompt template according to the to-be-tested module and generating a test case by utilizing a program generation model based on an adaptive focus context generation mechanism and the prompt template; the correction verification subsystem is used for performing multi-stage correction processing on the test case, eliminating errors and generating an initial test suite; and the test case supplement subsystem is used for carrying out coverage rate analysis and variation test on the initial test suite, constructing a model input text according to a result, and generating a supplement test case by utilizing a program generation model so as to update the initial test suite and obtain a final test suite. By combining the functional annotation of the to-be-tested module with the logical reasoning ability of the LLM, misjudgment caused by assertion errors can be reduced, and the accuracy of test feedback is ensured.

The invention relates to an industrial control system security policy adaptive configuration method and device and a readable storage medium. According to the method, a control instruction stream, a process variable sampling value and link layer message metadata are synchronously collected, an association hypergraph fusing network semantics, control semantics and physical semantics is constructed to represent a system state, and a sampling period is divided into a plurality of time slots with fixed lengths. In each time slot, self-supervised learning is utilized to extract a topology embedded vector group, and a system operation rule is represented. A robustness envelope interval is generated by calculating a statistical distance between a current time slot and a historical baseline, and is used for driving a reinforcement learning model to generate a security policy parameter set. And executing corresponding access control, traffic shaping and integrity verification operations in the next time slot, generating receipt vector feedback model update according to an execution result, outputting a micro-policy patch with a hash signature and a timestamp, and sending the micro-policy patch to the edge security gateway to realize deployment, thereby realizing closed-loop adaptive optimization of the security policy in the dynamic environment.

The invention discloses a test case determination method and device, a computer readable storage medium and a processor. The method comprises the following steps: identifying a source code of software by utilizing a pre-trained large model to obtain a plurality of functional modules in the software and a dependency relationship among the plurality of functional modules; in response to updating of the source code of the software, determining a code updating part in the source code; determining a test case matched with the code updating part from a test case library of the software based on the code updating part and a dependency relationship among the plurality of functional modules, the test case being used for testing the code updating part; based on a test result of the test case, adjusting a selection strategy of the test case; and re-determining the test case for testing the code updating part based on the adjusted selection strategy of the test case. The technical problem that a traditional regression test strategy is low in test efficiency is solved.

The invention relates to a multi-type industrial equipment PLC control method and device based on a hardware abstraction layer, and the method comprises the steps: constructing the hardware abstraction layer comprising a core layer and an adaptation layer; the core layer is provided with a standardized interface set, the adaptation layer is provided with a special adaptation module, and a control protocol is converted through a protocol analysis engine. And generating a unified instruction set based on the standardized interface set, and establishing a data interaction channel supporting priority scheduling. And identifying the newly-added equipment, matching the equipment configuration library, and generating a newly-added adapter drive if the equipment configuration library fails. And acquiring a device state by using a data interaction channel, and constructing a cooperative control logic to realize linkage. Equipment load and task queues are monitored in real time, and dynamic task scheduling and grading fault processing are carried out. And when an adapter configuration updating instruction is detected, the adaptive layer thermally loads the updated drive, and unified, intelligent, efficient and stable PLC control of multiple types of industrial equipment is realized through cooperation of the core layer and the adaptive layer.

The invention discloses a warehouse level code problem solving system based on a large language model, and relates to the technical field of software engineering, a source code file is extracted from a code warehouse, a static analysis tool is utilized to analyze codes, a warehouse level code graph is constructed, and then code entities and dependency relationships thereof are analyzed; based on the warehouse-level code graph, combining historical problem solving data, and constructing a dynamic evolution knowledge graph through continuous learning; through a context retrieval module, according to the context information of the current problem, dynamically retrieving related solution strategies and generating corresponding code patches, complementing or reviewing suggestions; the validity of the generated patch is verified through static analysis, unit testing and integration testing, and a problem solving strategy is optimized according to feedback of a developer. The problem solving efficiency and accuracy can be improved, the development cost is reduced, the solving process is gradually optimized through an intelligent feedback mechanism, a closed-loop feedback mechanism is formed, and therefore the increasingly complex software development requirements are met.

A system and method for deep learning using a large codeword model with homomorphically compressed and dyadically encrypted data is disclosed. The system preprocesses input data, applies homomorphic-dyadic compression and encryption, tokenizes the compressed data into sourceblocks, and assigns codewords using a codebook. These codewords are processed through a machine learning core, which can be either a conventional transformer-based architecture or a latent transformer core utilizing a variational autoencoder. The system enables secure operations on encrypted data, preserving privacy while allowing complex computations. The processed output is decrypted, decompressed, and translated to match the input modality. A neural upsampler may further enhance the output. The machine learning core is continuously trained using the processed data and additional training data, improving performance over time.