US 12,314,318 B2
Enhanced searching using fine-tuned machine learning models
Rahil Bathwal, San Francisco, CA (US); Daniel Fernando Campos, Hudson, NY (US); Ashwin Devaraj, Menlo Park, CA (US); Seth Michael Li, Foster City, CA (US); Yash Pande, San Francisco, CA (US); Vivek Raghunathan, Palo Alto, CA (US); Rajhans Samdani, Belmont, CA (US); and Danmei Xu, Santa Clara, CA (US)
Assigned to Snowflake Inc., Bozeman, MT (US)
Filed by Snowflake Inc., Bozeman, MT (US)
Filed on Feb. 16, 2024, as Appl. No. 18/444,078.
Claims priority of provisional application 63/446,750, filed on Feb. 17, 2023.
Prior Publication US 2024/0281446 A1, Aug. 22, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/2457 (2019.01); G06F 16/248 (2019.01); G06F 16/9032 (2019.01); G06F 16/93 (2019.01); G06F 16/9538 (2019.01); G06F 16/955 (2019.01)
CPC G06F 16/90328 (2019.01) [G06F 16/24575 (2019.01); G06F 16/248 (2019.01); G06F 16/93 (2019.01); G06F 16/9538 (2019.01); G06F 16/9558 (2019.01)] 20 Claims
OG exemplary drawing
 
8. A method comprising:
receiving, by one or more hardware processors, a search query via an interface;
accessing a pre-trained large language model designed to respond to the search query; and
performing a plurality of iterations, using the pre-trained large language model, to generate a task-specific generative model, each iteration comprising:
performing domain-specific pre-training on an index to fine tune the pre-trained large language model;
employing the task-specific generative model to generate a search result to the search query;
analyzing the search result based on a performance metric associated with the task-specific generative model; and
refining the task-specific generative model based on the analyzing of the search result.