US 12,443,859 B2
Dialogue model training method and device therefor
Seok Jun Seo, Seoul (KR); Seung Ju Han, Seoul (KR); Beom Su Kim, Seoul (KR); Bu Ru Chang, Seoul (KR); and Enkhbayar Erdenee, Seoul (KR)
Assigned to Hyperconnect LLC, Dallas, TX (US)
Filed by Hyperconnect LLC, Seoul (KR)
Filed on Jun. 17, 2022, as Appl. No. 17/807,653.
Claims priority of application No. 10-2021-0112541 (KR), filed on Aug. 25, 2021; and application No. 10-2021-0161615 (KR), filed on Nov. 22, 2021.
Prior Publication US 2023/0080930 A1, Mar. 16, 2023
Int. Cl. G06N 5/022 (2023.01); G06F 16/3329 (2025.01); G06F 40/35 (2020.01); G06N 3/0455 (2023.01); G06N 3/096 (2023.01)
CPC G06N 5/022 (2013.01) [G06F 16/3329 (2019.01); G06F 40/35 (2020.01); G06N 3/096 (2023.01); G06N 3/0455 (2023.01)] 15 Claims
OG exemplary drawing
 
1. A method of training a dialogue model in an electronic device, the method comprising:
selecting a first context from a first dialogue data set including at least one pair of a context and a response corresponding to the context;
generating a first response corresponding to the first context through a first dialogue model, wherein the first dialogue model is a generative-based dialogue model that generates a response to a given context;
generating an augmented dialogue dataset by incorporating a pair of the first context and the first response corresponding to the first context into the first dialogue data set;
generating an augmented response set including a response of the first dialogue data set and the first response; and
training a second dialogue model based on the augmented dialogue dataset, wherein the second dialogue model is a retrieval-based dialogue model that searches for a response to the given context, wherein the training comprises:
acquiring a response set including a first response subset corresponding to a second context included in the augmented dialogue dataset and a second response subset selected arbitrarily,
calculating a first score for a response included in the response set with respect to the second context based on the first dialogue model,
calculating a second score for a response included in the response set with respect to the second context based on the second dialogue model, and
training the second dialogue model based on the first score and the second score.