6 results about "Self feedback" patented technology

The invention discloses a multi-level cache and self-feedback driven retrieval enhancement generation reasoning acceleration system. The system comprises an input processing module, a multi-level cache management module, a retrieval enhancement generation module and a self-feedback optimization module. Wherein the multi-level cache management module adopts a three-level cache structure and is used for pre-filling and caching completely matched historical interaction data, similar matched associated reasoning results and hotspot knowledge of a knowledge base, and redundant calculation is reduced by multiplexing cache data in the reasoning stage; the retrieval enhancement generation module has dual functions: on one hand, a cache or a knowledge base is retrieved based on a query vector generated based on a user request, a retrieval result and user input are fused to generate an enhancement cue word, and a large language model is driven to perform reasoning so as to generate targeted output; on the other hand, when there is no user request, the hot knowledge of the knowledge base is actively recognized, knowledge arrangement type cue words are generated through a preset template, and a large language model is driven to generate preprocessing results (such as a core abstract and a preset question and answer pair) and store the preprocessing results in a three-level cache. And the self-feedback optimization module dynamically adjusts retrieval parameters, generation parameters and a cache strategy by evaluating the accuracy, correlation and timeliness of a generated result to form closed-loop optimization, so that invalid calculation and redundant information generation are reduced, the reasoning speed and the output quality are improved, and the collaborative efficiency of a large language model is remarkably improved.

The invention relates to the technical field of natural language processing and information retrieval, and discloses a retrieval enhancement generation method and device based on self-feedback driving, and the method comprises the steps: determining a to-be-queried problem through a pre-obtained query request; retrieving in the knowledge base according to the to-be-queried question, and generating an answer of the round of retrieval; calculating an information quality index of the current round of retrieval according to the answer and the to-be-queried question, and if the current round of retrieval is not the first retrieval, calculating an information gain rate of the current round of retrieval according to an information quality index of the previous round of retrieval and the information quality index of the current round of retrieval; and if the information gain ratio does not meet the preset condition, returning to determine the to-be-queried question in combination with the pre-acquired query request to perform the next round of retrieval, and stopping iteration and outputting the final answer until the information gain ratio meets the preset condition. The problems that an existing retrieval enhancement generation method depends on a fixed retrieval threshold value and lacks dynamic judgment capacity are solved, and the information utilization efficiency and the answer accuracy are improved.

The invention relates to the field of water immersion tests of radioactive substance transportation containers, and discloses a water immersion test system and method for a radioactive substance transportation container, and the test system comprises a test pressure cabin which is used for accommodating a to-be-tested container and carrying out a water immersion test; the water inlet pipeline is used for injecting water into the test pressure bin; the drainage pipeline is used for draining water in the test pressure bin; the pressure sensor is arranged at the top of the test pressure bin and is used for detecting the pressure in the test pressure bin in real time; the gas-liquid booster pump is used for pressurizing the interior of the test pressure bin; and the control end is electrically connected with the pressure sensor and the gas-liquid booster pump, and is used for displaying the pressure data of the test pressure bin detected by the pressure sensor in real time and controlling the gas-liquid booster pump to start and stop. According to the invention, real-time detection, monitoring and dynamic adjustment of the water pressure in the test pressure bin are completed through the pressure sensor control end and the gas-liquid booster pump, so that self-feedback adjustment of the radioactive substance transportation container water immersion test system is realized.

Systems and methods are provided for generating, managing, adapting, and delivering speech therapy to a user via a mobile device in an at-home or out-of-clinic setting. Users may be persons experiencing aphasia or other speech conditions. For example, prompts may be communicated to a user, and their spoken responses recorded and analyzed. Based on analysis of the responses, these systems and methods may determine a mode of response including but not limited to playback of the user's spoken response to allow the user to recursively self-assess and / or self-correct. User performance and improvement trends may be assessed, and utilized to adapt future therapy sessions.

Systems and methods are provided for generating, managing, adapting, and delivering speech therapy to a user via a mobile device in an at-home or out-of-clinic setting. Users may be persons experiencing aphasia or other speech conditions. For example, prompts may be communicated to a user, and their spoken responses recorded and analyzed. Based on analysis of the responses, these systems and methods may determine a mode of response including but not limited to playback of the user's spoken response to allow the user to recursively self-assess and / or self-correct. User performance and improvement trends may be assessed, and utilized to adapt future therapy sessions.

The invention discloses a CTCS and CBTC overlap area temporary speed limit autonomous synchronization method and system. The method comprises the steps that a scheduling system judges main control equipment of a speed limit system according to a speed limit starting point; the speed limiting system comprises main control equipment and controlled equipment which are respectively used for different track systems. The scheduling system sends a control command to the main control equipment according to an operation principle of the main control equipment, so that the main control equipment generates feedback information of the party and sends the feedback information to the scheduling system; meanwhile, the main control equipment forwards the control command to the controlled equipment through a data format conversion module, so that parameters in the control command are converted into an expression mode suitable for the controlled equipment, and the controlled equipment generates and sends opposite side feedback information to the main control equipment; and the main control equipment generates comprehensive feedback information according to the feedback information of the local party and the feedback information of the opposite party and sends the comprehensive feedback information to the scheduling system. According to the method and the system, the operation procedure of speed limit control is simplified, the potential safety hazard of train running is eliminated, and the problem that electronic maps of different systems cannot be aligned is solved.