Elena Koulidobrova
Central Connecticut State University, CT·Social Science
How do languages show where a speaker's information came from? Many languages have evidential markers, grammatical pieces that signal whether a statement rests on direct observation, inference, hearsay or some other kind of evidence.
- Fields
Social Science
Computer Science
- Topics
evidentiality
evidential markers
focus
- Grants
- 1 known grant, through Jul 2028
What they work on
- Build treebanks, collections of sentences annotated for grammatical structure in the Universal Dependencies format, that capture how evidential markers (grammatical signals of information source) and focus features appear in sentences.
- Train a deep-learning natural language processing model on the patterns around evidential markers so it can extract information from texts semi-automatically.
- Explain why evidential markers in some languages take on more than one function and appear in more sentence positions than those in other languages.
- Examine how flexible evidential and focus marking shapes the way speakers show evidence or emphasis across longer conversations.
More about the research
Studying how different languages encode the source of information can show how speakers judge information and can point to the thinking processes behind it.
Why it matters
Open-access annotated corpora and a trained language model could support language technology for languages that use evidential markers. The project names machine translation, text summarization, chatbots and other AI systems as possible uses. Better tools for detecting how a speaker signals the source of information could help these systems handle evidence and emphasis more accurately.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Frederik Hartmann
University of North Texas, TX·Social Science
Build artificial neural networks that learn to reconstruct proto-forms, the ancestor words of a language family, from their descendant words, using large datasets spanning many language families.
Emily Morgan
University of California-Davis, CA·Psychology
Build computational models of how speakers combine general constraints, such as the length of the receiver, with verb-specific preferences when choosing between the two English dative structures for verbs like give and send.
Eszter Ronai
Northwestern University, IL·Psychology
Identify the real-time processing signals of focus, the marking of new or emphasized information, in languages with fixed versus flexible word order, using eye-gaze data and statistical modeling.
The school
Central Connecticut State University: 1 labs on PI Pages.
Reach out
A start for your email. Fill in the brackets.