10 results about "Rewriting" patented technology

Methods and systems for generating a frequently asked questions are provided, which include defining, by a computer program executed by a computer, a first large language model (LLM) with a user query and a feedback of the user query; refining, by the computer program, the user query to a question set based on the feedback, the question set comprising one or more sentences; defining, by the computer program, a second LLM to generate a first set of question and answer pairs from a source document; defining, by the computer program, a third LLM to generate a content set from the source document based on a rewriting of the source document; selecting, by the computer program, top questions from the content set to be provided to the second LLM; and generating, by the second LLM, a second set of question and answer pairs based on the top questions.

The application discloses a syntax robustness reinforcement training method for large language model sequence knowledge reasoning, comprising the following steps: firstly, sequence knowledge question and answer samples are acquired, and question samples with different syntax structures but semantic equivalence are constructed for the same sequence knowledge unit to form training data containing original questions, synonymous rewritten questions and standard answers; secondly, the large language model is trained in the first stage based on the training data, the direct answering ability of the original question to the standard answer is learned through multi-task supervision fine tuning, and a small amount of synonymous rewritten auxiliary samples are used to make the model preliminarily adapt to the compound output structure of "synonymous rewriting-answer"; thirdly, a group relative strategy optimization algorithm is used in the second stage, and a composite reward function composed of answer correctness reward, semantic maintenance reward and format reward is combined to optimize the "synonymous rewriting-answer" track generated by the model through reinforcement learning; finally, a lightweight direct answer interface alignment is performed to make the model converge to the final answering interface of "input original question-direct output answer". The application is suitable for sequence knowledge question and answer, time sequence question and answer, process reasoning, hierarchical relationship judgment, education question and answer and enterprise knowledge service scenes, and can improve the answering accuracy and consistency of the large language model under the condition of multiple syntax variants.

The application discloses an open source dialogue model-oriented automatic jailbreaking prompt word generation and attack method and system. The application first screens a prompt word set from an original prompt word set; secondly, based on an attack question, the prompt word with the optimal attack efficiency is screened out from the original open source prompt word set, multi-path parallel testing is carried out by using a proxy model, and the attack success rate and the prompt word length of different prompt words in the prompt word set are evaluated in real time through a dynamic evaluation mechanism based on a greedy selection strategy, the final attack prompt word is output to a target model, and after being spliced with the attack question, the attack prompt word is returned and an analysis report is output; finally, the prompt word in the analysis report is adjusted by using an automatic mutation and expert modification strategy, and the prompt word is reconstructed and iterated according to the feedback result of each round of attack, so that a new prompt word is generated. The application combines offline evolution and online decision-making to automatically generate a high-success-rate jailbreaking prompt word, controls the length and overhead, and optimizes mutation by using semantic rewriting and logic skeleton extraction.

The embodiment of the application provides a data operation bit width expansion method and device, relates to the computer technical field, and comprises the following steps: reading variable vector information in a vector width control register, determining a maximum data operation bit width after expansion based on the variable vector information.If the width of to-be-processed data is greater than an original data operation bit width and less than or equal to the maximum data operation bit width, then based on the width of the to-be-processed data and the bit width of a single operation unit, the target number of operation units used for processing the to-be-processed data is determined.The target number of operation units is written into an operation unit control register after being reduced by one, so as to control the start of the target number of operation units, and the to-be-processed data is processed.Two control registers are newly defined to realize variable vector expansion without changing the original mode, and the data operation bit width is expanded from the original data operation bit width to a larger data operation bit width, so that the parallelization capability of data processing is improved without the need of instruction set expansion and code rewriting.

The present disclosure provides a method and device for training a rewriting model, a storage medium and a computer program product, relating to the technical field of computers. In the present disclosure, by using sample conversation data with judgment information, the model can not only learn which rewriting is correct, but also learn which rewriting is incorrect. The introduction of positive and negative samples helps the model learn from both correct and incorrect rewriting during the training process. Compared with the related art which only learns from the perspective of correct rewriting, the training effect of the rewriting model can be improved, thereby the accuracy of the rewriting model in generating rewritten sentences can be improved.

The present application relates to the technical field of resource scheduling, in particular to a software project resource optimization allocation system and method, the system comprises, in the present application, the task execution priority sequence is compared with personnel pool language item, framework item, module history, code branch occupation, work time slot, concurrent upper limit is executed item by item, and the task number and personnel number are matched with the help of hungarian matching algorithm to form a row-column matching structure, the matrix position is locked under the joint participation of the completely adaptive bit, the substitute bit, the frozen bit and the limit bit, the batch task resource allocation table is counted, the continuous occupation segment, the module switching segment and the defect single insertion segment, the high load member and the low load member are circled, and the execution person number is replaced, the task sequence is rearranged, the continuous occupation time period is corrected, the module switching order is adjusted, and the iteration screening is executed with the help of simulated annealing algorithm, so that the resource reallocation process can continuously compare and retain the better arrangement state between local rewriting and overall arrangement.

The application provides a menu permission configuration method, device and equipment and a computer readable storage medium. The method comprises the following steps: obtaining menu data and each first permission rule corresponding to the menu data; determining the operation mode of each first permission rule according to the menu data; performing connection operation on each first permission rule according to the operation mode to obtain a target permission rule; and configuring the target permission rule to a menu item corresponding to the menu data. In the application, the extension of rules can be realized through connection operation on multiple rules, code rewriting is not required, the extension process of menu rule permission is simplified, and the extension difficulty of menu permission rule is reduced.

This invention relates to the field of natural language processing technology, providing an end-to-end multi-turn dialogue rewriting method, system, and storage medium that integrates dialogue detection. In the proposed end-to-end multi-turn dialogue rewriting method, a joint learning approach is used to train the detector and generator, solving the problems of cascading errors and incoherent sentences after matching and filling in the Pipeline method, thus improving the detector's detection performance. By having the detector ignore sentences that do not need rewriting and integrating dialogue detection information into the generator, the slow speed and repetitive encoding problems of the Seq2Seq method are solved. The dialogue rewriting process of this invention does not require the generator to repeatedly encode the text, improving the model's prediction speed. Simultaneously, integrating dialogue detection information into the generator improves the generation quality of the rewritten sentences.