[UG] Summer Research Opportunity: Enabling Automatic Evaluation of Text-Based Self-Reflection

The following announcement is from [PhD student Emily Weiss and Professor Maja Matarić]. Please contact them directly if you have any questions.

We have two undergraduate research positions available for projects at the intersection of natural language processing, machine learning, and socially assistive robotics. We are interested in understanding how text-based self-reflections and robot facilitated self-reflections can be automatically analyzed to determine reflection quality.

This full-time position is for Summer 2026, with a start date of May 25, 2026, with optional opportunity to continue at part-time in the 2026-2027 academic year. 

 

Summary of work: 

Students joining these projects will help annotate natural language reflection datasets collected by the Interaction Lab. Students will help construct and evaluate hypotheses about how reflection quality is correlated with textual measures, and evaluate whether these measures are appropriate features for a quality detection model. Students may assist in simulating reflections using LLMs to further test their hypotheses on synthetic data representing combinations of personalities and behavioral traits not present in the original datasets.

Depending on experience and availability, students may also begin to incorporate their findings into a robot-facilitated reflection system and/or a feedback model for future experimentation.

 Requirements:

• Experience programming in Python.
• A strong background in probability/statistics, NLP, and/or machine learning.
• Willing to work a minimum of 20 and up to 40 hours a week in-person during typical business hours.

A paid stipend is available for the summer. We are also accepting volunteer researchers for this project.

 To apply: please email Emily Weiss (emilymwe@usc.edu) with the subject “[Research Application] Reflection”. 

In your email, please answer the following prompt: “Provide a short statement (one paragraph) describing your skills, interest, and why you would be a good candidate.”

Please also provide the following information in your email: your degree level (e.g. undergraduate – junior), department, a pdf of your resume or CV, and a pdf of your unofficial transcript.

Feel free to provide any additional information that you think would be helpful for us to know while considering your application.

To learn more about the Interaction Lab’s work, visit: https://uscinteractionlab.web.app/

 

Published on April 14th, 2026Last updated on April 14th, 2026