Swarnashree Mysore Sathyendra
I am a machine learning and natural language processing researcher working on post-training,specifically alignment and evaluation for large language models.
I’m currently a Member of Technical Staff at Reflection AI, where I work on post-training and alignment research for frontier language models. I am especially interested in developing methods that make large language models more reliable when they generate answers, follow instructions, use provided context, and abstain when they do not have enough evidence.
Before Reflection AI, I was an ML Research Engineer at Scale AI, where I worked on evaluation systems for human-annotated data quality, reward modeling, RLHF, rubric-guided preference data selection, and alignment pipelines for frontier AI models. I have also interned at Amazon Alexa AI on fairness in entity resolution models, and worked at Goldman Sachs as a software development engineer.
I obtained my masters from the Language Technologies Institute, CMU.
Publications
Chasing the Tail: Effective Rubric-based Reward Modeling for Large Language Model Post-Training
Junkai Zhang, Zihao Wang, Lin Gui, Swarnashree Mysore Sathyendra, Jaehwan Jeong, Victor Veitch, Wei Wang, Yunzhong He, Bing Liu, Lifeng Jin
International Conference on Learning Representations (ICLR), 2026
Multi-Dimensional Evaluation of Text Summarization with In-Context Learning in LLMs
Sameer Jain, Vaishakh Keshava, Swarnashree Mysore Sathyendra, Patrick Fernandes, Pengfei Liu, Graham Neubig, Chunting Zhou
Findings of the Association for Computational Linguistics: ACL 2023
Swarnashree Mysore Sathyendra*, Rajdeep Pal*, Ranjana Seshadri*, S. Natarajan
24th International Conference on Advanced Computing and Communications (ADCOM), IIIT Bangalore, 2019
Real-time Text-Search on Encrypted Data
Presented in association with Goldman Sachs at Grace Hopper Conference India, 2019
| Scholar | Mail: ms dot swarnashree at gmail dot com |
last updated: May 31, 2026