Shreya Rajpal

Shreya Rajpal

she/her

Ph.D. researcher in trustworthy neuro-symbolic reasoning

Shreya Rajpal

I am a Ph.D. student in Computer Science and Engineering at Michigan State University. I am advised by Prof. Parisa Kordjamshidi in the Heterogeneous Learning & Reasoning Lab, where I develop trustworthy neuro-symbolic methods for complex spatial reasoning. I am a University Distinguished Fellow and Dr. Frank B. Martin Cum Rec Endowed Fellow. Previously, I worked on efficient video understanding at the University of Tübingen, trustworthy medical-language-model explanations at IIT Madras, financial LLM applications with Georgia Tech's Financial Services Innovation Lab, and knowledge-graph question answering at the University of Hamburg. In summer 2026, I was an AI Science Intern at Intuit, where I worked on production recommendation systems and recommender migration research. I returned to Michigan State University in August 2026 to continue my Ph.D. My work has appeared at EMNLP, ACM FAccT, ACL workshops, and ISWC workshops.
I am open to collaborations. Let's connect on LinkedIn or email me at rajpalsh@msu.edu.

News

  • Oct 6, 2026

    Our Intuit work, Evidence Before Sampling, is now available. We propose an interpretable way to discover implicit negative candidates for recommendation, backed by evidence from observed customer behavior.

  • Sep 1, 2026

    Our Intuit work, From Gradient-Boosted Trees to Deep Recommenders, is now available. We share practical lessons from migrating a production customer-support recommender from CatBoost to deep learning.

  • Aug 2026

    Our works SATURN: Symbolic Spatial Reasoning for Multi-Perspective Grounding and Spatial Reasoning via Modality Switching Between Language and Symbolic Representations were accepted at the EMNLP 2026 main conference and EMNLP 2026 Findings, respectively.

  • Aug 2026

    I completed my AI Science internship at Intuit and am back at MSU to continue my Ph.D.

  • Jun 2026

    🎉 I joined Intuit as an AI Science Intern, working on recommendation systems.

  • Jun 2026

    🗣️ My paper on spatial reasoning via modality switching was presented at the SurgeLLM Workshop at ACL 2026.

  • Jun 2026

    📄 Our work on label-guided efficient procedural-video summarization was accepted to the SELVA Workshop at ACL 2026.

  • May 2025

    📄 LExT, our framework for evaluating trustworthy natural-language explanations, appeared at ACM FAccT 2025.

  • Aug 2025

    🎓 I started my Ph.D. at Michigan State University as a University Distinguished Fellow.

Selected Publications

Spatial Reasoning via Modality Switching Between Language and Symbolic Representations

Findings of the Association for Computational Linguistics: EMNLP 2026

PDF
Code

LExT: Towards Evaluating Trustworthiness of Natural Language Explanations

ACM Conference on Fairness, Accountability, and Transparency (FAccT 2025)

PDF