Vijay Murari Tiyyala

Hi, I’m Vijay Murari Tiyyala, a researcher in NLP and Machine Learning with a Master’s degree in Computer Science from Johns Hopkins University. Interested in my work? Check out my resume.

In my full-time research role at the Center for Language and Speech Processing at Johns Hopkins University, I work under Prof. Mark Dredze to enhance chatbot interactions in healthcare. Before this, I developed a multilingual medical concept dictionary with Dr. David Yarowsky. I also have experience as a Data Analyst at Deloitte.

Projects

Empathy-Enhanced LLMs: Refining AI Responses with Fine-Grained Human Feedback

In this project, I led the development of models to improve empathy, quality, factuality, and comprehensiveness in text generation. This involved fine-tuning models, creating innovative training datasets, and developing new algorithms to ensure healthcare chatbots provide responses that are not only accurate and relevant but also empathetic and comprehensive.

SAMOYEDS: Simulating Agents for Modeling Outcomes and Estimations to Direct Social-policy

Developed a tool using LLMs to simulate public responses to health policies, aiding decision-making during the COVID-19 pandemic.

Medical Terminology Translation and Multilingual Matrix Construction

Created a massive multilingual matrix for medical terms, enhancing machine translation for low-resource languages.

Cannabis Use Detection in Clinical EMR

Trained NLP models to detect cannabis use in EHRs with 97% accuracy, improving data quality and compliance.

Adverse AI: Automated Discovery of Adverse Event Reports

Led development of a tool to identify adverse events from unstructured text, achieving 97.5% accuracy and reducing manual review time.

Improving Code Editing through Self-Instruct

Enhanced code editing by fine-tuning code LLMs, streamlining workflows with a 37% pass@1 accuracy.

Public Health Data Science

I collaborated with John W. Ayers on multiple projects to leverage big data for public health insights. This included developing models to monitor trends in health behaviors and improve disease forecasting, contributing to impactful public health research.

Experience

NLP Researcher - Johns Hopkins University (Full-Time)

At Johns Hopkins, I engineered an empathetic medical chatbot using LLaMA3, boosting response accuracy and enhancing patient interaction quality. I also reduced training time by 50% using PyTorch/SLURM in a multi-GPU environment.

NLP Research Intern - Center for Language and Speech Processing

I developed a RAG chatbot with Apache Solr Cloud, reducing search time by 70% and increasing web traffic by 40%. Additionally, I optimized document retrieval and integrated re-ranking and chunk summarization.

Graduate Research Assistant - Johns Hopkins University

I improved machine translation accuracy for medical terminologies in low-resource languages, enhancing accessibility and precision.

Business Technology Analyst - Deloitte

At Deloitte, I developed stored procedures and scripts for integrating clients’ tax data via APIs, reducing processing time and boosting client retention.

Publications

Published

  • Kreyòl-MT: Building MT for Latin American, Caribbean, and Colonial African Creole Languages, NAACL 2024.

Under Review

  • ANALOBENCH: Benchmarking the Identification of Abstract and Long-context Analogies, submitted to ACL 2024.

Feel free to reach out for potential collaborations. You can find more about my work and projects on LinkedIn and GitHub.