Soroush Omidvartehrani

prof_pic.jpg
Ph.D. candidate in Computing Science at the University of Alberta, working on example-driven data wrangling. Specifically, I study how much a system can learn from a handful of well-chosen examples and how to carry what it learns across the rest of the data. I also work remotely as a Data Scientist at Dodona Data in Toronto.


Recent Publications

EdgeLM: Edge Demonstrations for Language Models' Table Understanding. Soroush Omidvartehrani, Mohammadamin Habibollah, Mohammadreza Daviran, Davood Rafiei. Conference of the European Chapter of the Association for Computational Linguistics (EACL) 2027. (submitted)

LDI: Localized Data Imputation for Text-Rich Tables. Soroush Omidvartehrani, Davood Rafiei. Tabular Data Analysis (TaDA) Workshop @ VLDB 2026, Boston, USA.

WebTableX: Efficiently Discovering Web Table Transformations Through Sampling. Soroush Omidvartehrani, Arash Dargahi Nobari, Davood Rafiei. Web Conference (WWW) 2025, Sydney, Australia.

GXJoin: Generalized Cell Transformations for Explainable Joinability. Soroush Omidvartehrani, Arash Dargahi Nobari, Davood Rafiei. European Conference on Advances in Databases and Information Systems (ADBIS) 2024, Bayonne, France.


Ask

Suggested questions

Every answer is written from the notes on this site and lists the passages it drew on. Answers are generated by a language model reading a small set of notes about my work, so treat them as a starting point rather than a citation. Follow the linked sources for anything that matters. Questions are not stored, and the assistant knows nothing beyond what is on this site.

Last update: Aug 2026