Thesis supervisor: Gábor Iván Berend
Location of studies (in Hungarian): SZTE Abbreviation of location of studies: SZTE
Description of the research topic:
The field of computational semantics has witnessed tremendous progress recently, however, an array of important research questions still remains open. Although popular neural approaches can obtain high performance in a variety of natural language understanding (NLU) benchmark tasks, they also come with several shortcomings. The expected outcome of the research topic is to propose critical evaluations and deep analysis on the governing computational approaches for computational semantics, and to propose new algorithms for overcoming the shortcomings identified by the analyses. The proposed algorithms are primarily (but not exclusively) expected to mitigate the data hunger, high computational demand and the brittleness of the predictions of the most popular model architectures. Another important research goal for the given topic is to better understand the inner working mechanisms of the already existing algorithms, both from theoretical and practical aspects, so that they can be applied more efficiently.
Required language skills: English Number of students who can be accepted: 3
Deadline for application: 2022-03-15
2024. IV. 17. ODT ülés Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).
Login
