2results about How to "Achieve privacy" patented technology

The application relates to an efficient and privacy-considered large model end-cloud collaborative inference method and system, which comprises the following steps: obtaining input and configuring KV cache control parameters through an end side; performing embedding calculation to generate a feature tensor and uploading the feature tensor to a cloud side; performing decoding layer inference on the cloud side according to authorization to call the KV cache and returning a hidden state; in a current generation step, performing speculative decoding prediction on subsequent tokens based on the hidden state on the end side, and calculating an end-side local inference result corresponding to a low-dimensional feature according to a dimension splitting rule; calculating a cloud-side local inference result corresponding to a high-dimensional feature on the cloud side, and returning the cloud-side local inference result after language self-adaptive cutting and encryption; fusing the two-end local inference results to obtain a complete distribution, sampling to generate an incremental token, uploading the incremental token to the cloud side for verification, and updating the KV cache.

The invention belongs to the technical field of higher education management, and discloses a semi-supervised learning-based college student consumption data mining and distinguishing method, which comprises the steps of obtaining college multi-source original consumption data, converting the data into a desensitization behavior unit according to a preset mapping interval, counting a single score of each dimension, and generating a monthly consumption toughness index; creating a consumption toughness time sequence atlas, mounting external event tags, and integrating to form a whole school consumption toughness time sequence atlas database; optimizing a toughness rule weight through semi-supervised learning and mining a recessive consumption mode combination to obtain an optimized five-dimensional consumption toughness weight matrix and an enhanced feature set, and further generating a pseudo tag set; performing economic difficulty state judgment on the to-be-judged students through a three-step fusion judgment rule chain, generating a judgment analysis report, and performing division to obtain an economic difficulty student candidate list, a non-economic difficulty list and an observation list; and performing three-dimensional synchronous feedback optimization in combination with an artificial rechecking result to form a discriminant management link of closed-loop evolution.