PeopleofSuperintelligence

Researcher

Tanya Goyal

Tanya Goyal is a computer scientist whose research develops ways to evaluate generated text and large language models. She studied mathematics and computing at IIT Guwahati, completed a computer science PhD at UT Austin in 2023, and received UTCS’s Bert Kay Dissertation Award for her work on text-generation evaluation. Her papers study human judgments of document edits, factuality in generated summaries, and FALTE, a tool for annotating long text. She worked as a postdoctoral scholar at Princeton’s Language and Intelligence Center from 2023 to 2024, then joined Cornell as an assistant professor in July 2024. At Cornell, she studies natural language processing, reliable LLM evaluation, factuality, and how training data and alignment shape model behavior.

Works
66
Citations
777
h-index
13

Publication numbers from OpenAlex, as of 27 October 2026.

Timeline

  1. 2011–2015

    Mathematics and computing

    She earns a B.Tech. in mathematics and computing at the Indian Institute of Technology, Guwahati, in 2015.

  2. 2017

    Document edit importance

    In 2017, she coauthors a study of how people judge the importance of document edits and models the weight of each change between versions.

  3. 2020

    Dependency arc entailment

    In 2020, she and Greg Durrett introduce dependency-arc entailment to test whether relations in generated text are supported by the input and to locate factual errors.

  4. 2021

    Fine-grained summary factuality

    In 2021, she and Durrett study factual errors in abstractive summaries at word, dependency, and sentence levels. Their paper finds that fine-grained human labels give models a stronger training signal for factuality.

  5. 2022

    FALTE annotation toolkit

    In 2022, she coauthors FALTE, a web-based toolkit that lets researchers collect fine-grained crowdworker judgments of long generated text.

  6. 2023

    A PhD and dissertation award

    She completes a computer science PhD at the University of Texas at Austin in 2023, advised by Greg Durrett. Her research on evaluating text generation models earns UTCS’s Bert Kay Dissertation Award.

  7. 2023–2024

    Princeton postdoctoral research

    From 2023 to 2024, she works as a postdoctoral scholar at Princeton Language and Intelligence Center.

  8. April 2024

    FABLES for book summaries

    In April 2024, she coauthors FABLES, a human evaluation of faithfulness and content selection in LLM-generated summaries of fictional books.

  9. Since July 2024

    Assistant professor at Cornell

    In July 2024, she joins Cornell’s Department of Computer Science as an assistant professor. Her current work covers reliable evaluation of large language models, factuality, and how training data and alignment shape model behavior.

Career

  1. Since 2024Cornell UniversityAssistant professor
  2. 2023–2024Princeton UniversityPostdoctoral scholar at Princeton Language and Intelligence Center

Education

  1. 2023University of Texas at AustinPhD in computer science
  2. 2015Indian Institute of Technology GuwahatiB.Tech. in mathematics and computing

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Tanya Goyal yet. Send us a free-licensed photo.

  1. Tanya Goyal tagoyal.github.io. Checked 27 October 2026.
  2. An Empirical Analysis of Edit Importance between Document Versions - ACL Anthology aclanthology.org. Checked 27 October 2026.
  3. Evaluating Factuality in Generation with Dependency-level Entailment - ACL Anthology aclanthology.org. Checked 27 October 2026.
  4. Annotating and Modeling Fine-grained Factuality in Summarization - ACL Anthology aclanthology.org. Checked 27 October 2026.
  5. FALTE: A Toolkit for Fine-grained Annotation for Long Text Evaluation - ACL Anthology aclanthology.org. Checked 27 October 2026.
  6. Tanya Goyal | Department of Computer Science | Cornell Bowers cs.cornell.edu. Checked 27 October 2026.
  7. FABLES: Evaluating faithfulness and content selection in book-length summarization arxiv.org. Checked 27 October 2026.
  8. Meet the new faculty: Tanya Goyal | Cornell Bowers bowers.cornell.edu. Checked 27 October 2026.