11 results about "Memory analysis" patented technology

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as financial science and technology and medical health, and discloses a memory analysis question and answer data construction method and device, equipment and a medium. The method comprises the steps of extracting a fact information set containing entities, relationships and values based on structured message representation, generating candidate question and answer pairs by utilizing a language model, executing multi-source verification to form a target question and answer set, performing semantic extension and question method transformation to generate a diversified question and answer set, and completing credibility screening based on a verification result to obtain a memory analysis question and answer set. And generating a memory analysis data set in combination with the user message sequence. According to the method, through multi-stage extraction, generation, verification and screening, reliability and diversity unification of question and answer content are achieved, the constructed memory analysis data set has the advantages of being consistent in context, accurate in fact, rich in expression and the like, and therefore the credibility of memory ability evaluation of the large language model is improved.

The application discloses a memory detection method and system of a virtual machine. The method is applied to a memory detection system, and the memory detection system is used for running a main process to execute the following method: in response to an inter-process communication request of at least one detection process, determining to-be-detected memory of a virtual machine; performing memory analysis on the to-be-detected memory according to a detection item, and obtaining an analysis result, wherein the analysis result comprises memory information of the to-be-detected memory under an operating system where the virtual machine is located; transmitting the analysis result to the detection process through inter-process communication between the main process and the detection process, wherein the analysis result is used for detection according to the detection item in the detection process, and a memory detection result is obtained; and obtaining the memory detection result corresponding to the analysis result through inter-process communication. The application solves the technical problem that the stability of a detection system for detecting memory of a virtual machine by relying on proxy software is poor.

Methods and apparatus consistent with the present disclosure may use instrumentation code that remains transparent to an application program that the instrumentation code has been injected into. In certain instances, data sets that include executable code may be received via packetized communications or be received via other means, such as, receiving a file from a data store. The present technique allows a processor executing instrumentation code to monitor actions performed by the program code included in a received data set. Malware may be detected by scanning suspect program code with a malware scanner, malware may be detected by identifying suspicious actions performed by a set of program code, or malware may be detected by a combination of such techniques.

The invention provides an AI platform data recommendation method and device and computer equipment, and relates to the technical field of data processing, and the method comprises the steps: obtaining a historical consultation dialogue record set of a target user, carrying out progressive multi-round dialogue iterative memory analysis, and determining the understanding category of the target user; obtaining a real-time active reasoning dialogue sequence; when a backtracking detection module is called to carry out recommended data completeness detection on the real-time active reasoning dialogue sequence, a recommended data completeness detection result is obtained; and triggering a recommendation output instruction according to the recommendation data completeness coefficient, and pushing recommendation data based on the recommendation output instruction. The data recommendation method and device have the advantages that the technical problems that in the data recommendation process in the prior art, in-depth interaction with the user cannot be dynamically conducted, initiative questioning in accordance with the actual situation of the user cannot be achieved, the recommendation quality is low, and the real-time requirement of the user cannot be met easily are solved, and the technical effect of improving the data recommendation quality is achieved.

The application relates to a heap memory analysis method and device, computer equipment, a chip and a chip module. The method comprises the following steps: in response to a heap memory processing instruction for a to-be-analyzed heap memory, determining the array identifier of a preset array corresponding to the to-be-analyzed heap memory according to the heap memory address corresponding to the heap memory processing instruction; updating the array information of the preset array according to the heap memory processing instruction and the array identifier, to obtain the updated array information of the preset array; and generating a memory analysis report corresponding to the to-be-analyzed heap memory based on the updated array information. The method can improve the analysis efficiency of the heap memory.

Methods and apparatus consistent with the present disclosure may use instrumentation code that remains transparent to an application program that the instrumentation code has been injected into. In certain instances, data sets that include executable code may be received via packetized communications or be received via other means, such as, receiving a file from a data store. The present technique allows a processor executing instrumentation code to monitor actions performed by the program code included in a received data set. Malware may be detected by scanning suspect program code with a malware scanner, malware may be detected by identifying suspicious actions performed by a set of program code, or malware may be detected by a combination of such techniques.

The invention provides a computer memory release system based on data analysis, and relates to the technical field of computer memory management. The system comprises a memory data acquisition module, a multi-dimensional memory analysis module, an intelligent release decision module, a hierarchical release execution module and a memory fragment optimization module. The memory data acquisition module acquires and preprocesses data by adopting kernel mode and user mode dual channels; the multi-dimensional memory analysis module realizes process memory consumption mode recognition and release potential calculation through feature matrix construction, improved LSTM network prediction and entropy weight method evaluation; the intelligent release decision-making module realizes dynamic decision-making based on an AHP model in combination with scene adaptation; the hierarchical release execution module adopts a three-stage mechanism of pre-cleaning, verification and release to guarantee safe release; and the memory fragment optimization module realizes free block integration and closed-loop feedback. The system solves the problems that a traditional memory is slow in release response, blind in release and poor in stability.

This invention discloses a vulnerability exploitation agent system, method, and apparatus based on the driving engineering framework, relating to the field of artificial intelligence, to address the vulnerability problem in the autonomous execution of long-chain tasks by vulnerability exploitation agents. It constructs a vulnerability exploitation agent system comprising an environment perception and object representation layer, an execution layer, an orchestration layer, a memory layer, a semantic management layer, and a security layer. Based on the logic of global constraint pre-positioning – progressive information disclosure – full-link behavior control – autonomous closed-loop execution – standardized output of results, it leverages the core mechanisms of the driving engineering framework, such as event-driven interception and verification bus, dynamic loading of prompts, memory compression, autonomous task execution loop, and tool permission gating, to form a fully traceable vulnerability exploitation scheme. This application can solve the problems of difficult binary semantic analysis and memory analysis in long-chain vulnerability exploitation tasks, improving task execution efficiency and accuracy.

The invention belongs to the technical field of data analysis and artificial intelligence, and particularly relates to a multi-scene switching audio loudspeaker tone quality adaptive adjustment system. Comprising a four-dimensional perception layer, a mixed scene-content-user-copyright fusion analysis module, an auditory memory analysis module, a multi-user preference arbitration module, an extreme environment compensation unit, a copyright-tone quality cooperative adjustment module, an equipment aging adaptation module, a dynamic decision-making layer, a feedback execution layer, a hearing quantitative evaluation engine and a cloud cooperative engine. The dynamic decision-making layer fuses multi-module output according to the constraint priority to generate a compliance adjustment instruction; the feedback execution layer executes adjustment and realizes partitioning; the hearing quantitative evaluation engine optimizes the adjustment effect through objective indexes and user feedback; the cloud collaboration engine synchronizes global data, provides support for decision making, and forms full-closed-loop self-adaptive adjustment of perception-processing-decision making-execution-evaluation-collaboration.

The application discloses an implementation method of a market electric refrigerator compatible with different power supply stability, comprising the following steps: formulating a refrigeration operation rule one of a power stability area and a refrigeration operation rule two of a frequent power failure area; the refrigerator records the total power-on time length and the power-off times in the total power-on time length through a memory chip; the memory chip analyzes and judges the total power-on time length and the power-off times in the total power-on time length; the refrigerator control center gives different judgment result instructions according to the comparison results of the total power-on time length and the preset judgment time length and the comparison results of the power-off times and the preset judgment times; and the refrigerator executes corresponding control operation rules based on the different judgment result instructions. The method is convenient to operate, does not need additional investment, meets the requirements of generalization, modularization and standardization, reduces the consumption of manpower and material resources, reduces the risk of wrong installation, and improves work efficiency.

This invention relates to the field of electronic digital data processing technology and discloses a software component verification method based on runtime memory analysis. The method includes: reading a virtual reference address and the instruction stream to be verified when the target memory page attribute switches from non-executable to executable; dividing the instruction stream to be verified into continuous memory blocks according to the hardware cache line alignment length; counting the number of instruction start boundaries within each memory block and serializing them in ascending address order to generate a feature vector representing load density; matching the feature vector with a reference library and sending an interception signal to the kernel scheduler when the distance metric exceeds a threshold. This invention utilizes the physical boundaries of the hardware cache lines to capture load structure distortions caused by code obfuscation, avoiding complex semantic parsing logic, effectively reducing system latency caused by verification, and synergistically improving the verification confidence for dynamically mutated components.