TY - GEN
T1 - Vector-to-sequence models for sentence analogies
AU - Wang, Liyan
AU - Lepage, Yves
N1 - Funding Information:
This work was supported by grant number 18K11447 from the Japanese Society for the Promotion of Science (JSPS) entitled "Self-explainable and fast-to-train example-based machine translation using neural networks".
Publisher Copyright:
© 2020 IEEE.
PY - 2020/10/17
Y1 - 2020/10/17
N2 - We solve sentence analogies by generating the solution rather than identifying the best candidate from a given set of candidates, as usually done. We design a decoder to transform sentence embedding vectors back into sequences of words. To generate the vector representations of answer sentences, we build a linear regression network which learns the mapping between the distribution of known and expected vectors. We subsequently leverage this pre-trained decoder to decode sentences from regressed vectors. The results of experiments conducted on a set of semantico-formal sentence analogies show that our proposed solution performs better than a state-of-the-art baseline vector offset method which solves analogies using embeddings.
AB - We solve sentence analogies by generating the solution rather than identifying the best candidate from a given set of candidates, as usually done. We design a decoder to transform sentence embedding vectors back into sequences of words. To generate the vector representations of answer sentences, we build a linear regression network which learns the mapping between the distribution of known and expected vectors. We subsequently leverage this pre-trained decoder to decode sentences from regressed vectors. The results of experiments conducted on a set of semantico-formal sentence analogies show that our proposed solution performs better than a state-of-the-art baseline vector offset method which solves analogies using embeddings.
KW - Decoder
KW - Sentence analogies
KW - Sentence embeddings
UR - http://www.scopus.com/inward/record.url?scp=85099749697&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85099749697&partnerID=8YFLogxK
U2 - 10.1109/ICACSIS51025.2020.9263191
DO - 10.1109/ICACSIS51025.2020.9263191
M3 - Conference contribution
AN - SCOPUS:85099749697
T3 - 2020 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
SP - 441
EP - 446
BT - 2020 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 12th International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
Y2 - 17 October 2020 through 18 October 2020
ER -