Tanay Nagar - तनय नगर

Independent researcher | NLP · Human-Centered AI · Computational Social Science · AI Safety

Independent researcher NLP · Human-Centered AI · Computational Social Science · AI Safety

My work focuses on developing equitable, socially-aware language technologies that prioritize language equity for low-resource languages, uphold the socio-cultural nuances of diverse linguistic communities, and ensure robust human-AI dynamics.

I'm particularly interested in how we can:

better evaluate/understand social subtleties across languages and cultures, both through and within language technologies;
embed socio-ethical dynamics/intelligence within the creation of language technologies.

I earned a B.S. in Computer Science (Honors), a B.S. in Philosophy, and a certificate in leadership from the University of Wisconsin-Madison, and I am currently preparing for graduate study.

I'm grateful to have learned from and collaborated with Dr. Nitesh V. Chawla, Dr. Sarah Jung, Dr. Shamya Karumbaiah, Dr. Jacob Thebault-Spieker, and Dr. Margarita Ruiz Olázar across the University of Wisconsin-Madison, UW Health, University of Notre Dame, and Universidad Comunera.

I'm always happy to connect; feel free to reach out via email or the links at the top of the page.

NLP Human-Centered AI Computational Social Science AI Safety

Featured publications and presentation

Exploring Bias in Letters of Recommendation using NLP Techniques

Tanay Nagar, Sarah Jung, Peter Wirth, Alyssa Schappe, Amorn N. Salyapongse

Oral presentation: ACEPS 2025 · Extended abstract published in PRS Global Open 2025.

Hina and Faisal Mushtaq Scholarship

Talk Slides PRS Global Article Google Scholar DOI

We use natural language processing to analyze plastic surgery residency letters of recommendation, surfacing systematic patterns in wording that may reflect bias and influence selection decisions. The work highlights how computational text analysis can support fairer, more transparent evaluation of applicants.

Bridging Language Gaps: Code-Switched Fine-Tuning for Multilingual LLMs

Tanay Nagar, Grigorii Khvatskii, Anna Sokol, Nitesh V. Chawla

Poster: NAACL 2025 Student Research Workshop (SRW); ND Summer Research Symposium 2024 · Full paper on arXiv 2025.

Bromley Conference Travel Award CDIS Student Travel Award

NAACL SRW Poster Visualization arXiv ND Symposium Poster

We study code-switched fine-tuning strategies for multilingual language models, focusing on how mixing languages in realistic ways affects performance. The work explores how these techniques can narrow performance gaps for low-resource and multilingual communities.

Decoding Bias in Letters of Recommendation: A Word Embeddings Approach

Tanay Nagar

Senior Honors Thesis, University of Wisconsin-Madison, 2025 · Oral presentation: UW Honors Research Symposium 2025.

Advisors: Dr. Fred Sala, Dr. Sarah Jung, Dr. Shamya Karumbaiah

Hilldale Undergraduate/Faculty Research Fellowship Richard Ralph Phoenix Rising Humanitarian Scholarship

Honors Symposium Talk Fellowship Proposal

Using 1,585 de-identified letters of recommendation (~919k tokens), my thesis combines Word2Vec and fine-tuned Mistral-7B embeddings to measure gender-directional framing via bias axes, WEAT-style tests, and analogy probes. It turns long-standing concerns about biased LoR language into quantitative metrics that can be used for reviewer calibration, dashboards, and faculty workshops to support fairer admissions and evaluations.

/etc

Animals I’ve met along the way: Field Notes