Disentangle syntax and semantics in sentence representation with decomposable variational autoencoder
US12039270B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Aug 5, 2020 |
| Grant date | Jul 16, 2024 |
| Priority date | — |
| Expiry date | Feb 26, 2042 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F40/211
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Described herein are embodiments of a framework named decomposable variational autoencoder (DecVAE) to disentangle syntax and semantics by using total correlation penalties of Kullback-Leibler (KL) divergences. KL divergence term of the original VAE is decomposed such that the hidden variables generated may be separated in a clear-cut and interpretable way. Embodiments of DecVAE models are evaluated on various semantic similarity and syntactic similarity datasets. Experimental results show that embodiments of DecVAE models achieve state-of-the-art (SOTA) performance in disentanglement between syntactic and semantic representations.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.