People

David R. Mortensen is a computational linguist specializing in phonology, morphology, and language change and is Associate Research Professor in the Language Technologies Institute at Carnegie Mellon University. He is the head of ChangeLing Lab.

Shikhar Bharadwaj is a third year PhD student in ChangeLing Lab. His interests are speech processing and learning world knowledge from speech. He is currently involved in building phonetic foundation models.

Chin-Jou Li is a first year LTI PhD student. She worked on and universal phone recognition and atypical speech assessment during her master’s here, and she’s now continuing to explore these areas.

Atharva Naik is a PhD student advised by Prof. Carolyn Rose at the Language Technologies Institute (LTI), Carnegie Mellon University (CMU). He also collaborates closely with Prof. Daniel Fried and Prof. David Mortensen. His research focuses on easy-to-hard generalization for LLM agents. He studies how to improve agent capabilities through controllable-difficulty synthetic data generation, post-training with verifiable or meta-level supervision, and inference-time optimization. He applies these methods to coding agents for code quality (code review, linting, refactoring), data science agents (e.g., Kaggle-style challenges), and computational linguistics (e.g., forward reconstruction).

John Ternullo is a student in the Fifth Year Master’s Program in Language Technologies at CMU and a member of the ChangeLing Lab. He is interested in multilingual ASR and technologies for low-resource languages. His past work has applied computational methods to phonological typology.

Thomas Pei is a first-year Data Science and Machine Learning student at Carnegie Mellon University. He is currently working with the ChangeLing Lab on rho baselines.

Ivan Zhang is an undergraduate student in SCS at CMU. He’s interested in exploring natural-language reasoning capabilities of LLMs and how they are influenced by linguistic principles.

Daniel (Chengmin) Xu is an undergraduate student studying Linguistics at CMU. He is interested in the intersection between multilingual NLP and speech processing, and AI/ML at large. His past work includes research on Sino-Japanese phonology, audio synthesis, and computational conlang generation. He is currently implementing solutions to automate backwards reconstruction.

Eugene Hwang is a sophomore studying Information Systems, with plans to double major in Computer Science. He is interested in studying lexical change with natural language processing and machine learning. He is currently working on a project investigating and improving large language models’ ability to date texts from linguistic cues rather than memorization.

Sadhana Panthi recently graduated in Computer Engineering from Pulchowk Campus, Tribhuvan University, Nepal. Her research interests include NLP and speech technologies for low-resource languages, with a broader interest in multilingual AI.

Collaborators

Alumni

PhD

•   •   Maria RyskinaMIT
•   •   Brendon BoldtEvenUp

MS & Undergrad

•   •   Katherine J. Zhang (MA GCAT)HoYoverse
•   •   Nathan Anderson (MLT)BYU
•   •   Xinyu Zhang (MLT)
•   •   Chenxuan Cui (MIIS)NVIDIA
•   •   Youngmin Kim (MIIS)Meta
•   •   Nate Robinson (MLT)JHU
•   •   Yanlin Feng (MLT)Megagon
•   •   Anna Cai (MSML)Meta
•   •   Kalvin Chang (BS CS, MLT)Berkeley
•   •   Kwanghee Choi (MLT)UT Austin
•   •   Keer Xu (MIIS)NetApp
•   •   Liang Lu (BS CS)
•   •   Ting Chen (BS StatML)IU Bloomington

Visitors & Collaborators

•   •   Leonie WeissweilerLeipzig University
•   •   Eunjung YeoUT Austin
•   •   Haeji JungUBC
•   •   Yoonjae Kim

Based out of Seattle, Rithvik is an applied ML scientist at Blue Origin, where he leads efforts to make rocket factories more efficient by building and scaling agentic systems. His interests range from social NLP all the way to reconstructing lost languages.