US 12,314,266 B2
Method, device, and product for searching data
Jiacheng Ni, Shanghai (CN); Bin He, Shanghai (CN); Tianxiang Chen, Shanghai (CN); Zhen Jia, Shanghai (CN); and Zijia Wang, Weifang (CN)
Assigned to Dell Products L.P., Round Rock, TX (US)
Filed by Dell Products L.P., Round Rock, TX (US)
Filed on Oct. 13, 2023, as Appl. No. 18/486,616.
Claims priority of application No. 202311238046.2 (CN), filed on Sep. 22, 2023.
Prior Publication US 2025/0103599 A1, Mar. 27, 2025
Int. Cl. G06F 16/00 (2019.01); G06F 16/2457 (2019.01); G06F 16/2458 (2019.01)
CPC G06F 16/2457 (2019.01) [G06F 16/2462 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
encoding a search input into a first dense vector based on a first multi-modal search model of a model store of a storage system;
determining, based on the first dense vector, a distilled data item corresponding to the search input from a distilled dataset corresponding to the first multi-modal search model, wherein the distilled dataset is generated from an original dataset utilizing a data distillation model of the model store of the storage system, the first multi-modal search model of the model store of the storage system being constructed based on the distilled dataset, and wherein the first multi-modal search model has an accuracy that is less than that of a second multi-modal search model constructed based on the original dataset;
encoding, based on the first multi-modal search model, an original data item in an original data subset corresponding to the distilled data item in the original dataset corresponding to the distilled dataset into a second dense vector; and
determining, based on the second dense vector, an original data item from the original data subset as a search result corresponding to the search input.