About
I am a fifth-year PhD student in computer science at the University of Chicago, advised by Dr. Allyson Ettinger. I also work closely with Dr. Claire Donnat and Dr. Kanishka Misra. I study how LLMs can learn over time and how LLM-based agents can work together.
My research spans data selection, continual learning, and multi-agent communication and evaluation. Two questions drive it: Which data should a model learn from, and how can it gain new capabilities without losing what it already knows? And when does communication help a group of agents solve a problem, and how should we evaluate that collaboration?
Before my PhD, I worked in industry on distributed file systems and data replication. I did my M.Tech. in Computer Science at IIT Kanpur.
I'm looking for industry research roles and am open to collaborations. Feel free to reach out at lcpandia@uchicago.edu.
News
- SAFE-DRIFT will appear at NeurIPS 2026.
- Presented “Rethinking Data Selection” at the Data Problems for Foundation Models workshop at ICLR 2026.
Publications
- SAFE-DRIFT: Data Selection for Supervised Fine-tuning with Controllable Off-Target Drifts
- Rethinking Data Selection: The Importance of Coverage over Difficulty in Generative Fine-Tuning
- Sorting Through the Noise: Testing Robustness of Information Processing in Pre-trained Language Models
- Pragmatic Competence of Pre-trained Language Models through the Lens of Discourse Connectives
Industry experience
-
System ArchitectDec 2018 – Jul 2022Huawei Technologies
- Designed the replication module for the Huawei file system.
- Designed a distributed data manager for cross-cluster storage replication.
-
Member of Technical Staff IIJul 2017 – Jul 2018NetApp
- Built data-backup functionality for moving data from storage servers to backup devices.
-
System EngineerDec 2012 – Jun 2015Tata Consultancy Services
- Built Java client-server applications for insurance policy management.