Kevin Ros

Computer Science PhD Candidate

I am a Ph.D Candidate at The University of Illinois Urbana-Champaign

I study information retrieval under ChengXiang Zhai as part of the TIMAN Group. I am expecting to graduate in the summer of 2025.

My main research goal is to predict when people need information, and then to proactively resolve their needs with minimal user effort.

To contact me, please reach out at

About

I am a fifth-year computer science Ph.D. candidate studying information retrieval under ChengXiang Zhai as part of the TIMAN Group at The University of Illinois at Urbana-Champaign.

Prior to this, I graduated from Vassar College in 2020 with a B.A. in Computer Science and a B.A. in Mathematics.

Outside of academia, I enjoy playing volleyball.

Research

If you're like me, you're tired of constantly having to manually search for things. This is especially frustrating in settings where the search is redundant with respect to the context. For example,

  1. During a meeting, someone references a document. Then someone in the meeting needs to manually search to pull up this document and share it with everyone
  2. While reading a book, you come across a word you don't know. So you need to open your browser and search to define the word.
  3. During a class, the professor mentions a concept that's a bit confusing. So you try to find some online diagram that explains the concept
  4. In a conference presentation, the presenter mentions a related work that in unfamiliar to you. So you pull up the paper being presented, find the title, search for it online, and then skim the discovered paper.

I'd like to create a world where this information is provided automatically, so that you don't need to spend the time looking for it yourself. Towards this goal, we've built instinfo, as the first step towards practically addressing these challenges.

Concretely, this research leverages state-of-the-art language models, machine learning, data collection, benchmark and dataset creation, and user studies.