Senior Research Scientist · DeepMind

Roma Patel

I'm a Senior Research Scientist at DeepMind working on understanding safety, behavior and alignment of LLMs , with a particular interest in how increasingly capable language-model systems can be trained, evaluated and stress-tested in a way that prevents their failures from being consequential. I finished my PhD at Brown University advised by (the incredible) Ellie Pavlick, and defended my thesis with the most excellent committee I could have asked for; with Stefanie Tellex, George Konidaris, Michael Littman and Felix Hill, focusing on language grounding, representation, and interpretability: foundations that continue to shape how I think about language understanding. As an undergrad, I was advised by Ani Nenkova and Byron Wallace in various areas of machine learning and language processing.

My research asks how we can understand failures in language models, improve alignment objectives and training methods, and build systems that proactively surface emerging risks.

Apart from work, I enjoy reading vast amounts of literature, going to new mountains, and to coffeeshops. Feel free to reach out with research related questions or otherwise!

Research Agenda

My current work is organized around three connected questions:

Understanding model failures

How do we build evaluations to uncover when and why models exhibit harmful, biased, sycophantic, or otherwise undesirable behavior in real human–AI interactions?

Improving alignment mechanisms

How can feedback, reward design, and learning objectives better represent diverse or conflicting human values and produce robust behavior?

Anticipating emerging risks

How can adversarial, multi-agent, and automated evaluation systems discover failure modes that static evaluations or human red teams may not anticipate?

Where I've Been

Google DeepMind, San Francisco
Safety, alignment, and reinforcement learning methods for LLMs.
2026-current.
DeepMind, London
Language grounding, reinforcement learning, multi-agent and game-theoretics methods for LLMs.
2022-2025.
Microsoft Research: Microsoft Turing Academic Program
Worked with Dean Carignan, Saurabh Tiwary, Pooya Moradi, Ali Alvi and others at MSR.
Summer 2021.
DeepMind, London: Research Intern (Multi-agent Reinforcement Learning)
Worked with Angeliki Lazaridou, Edward Hughes, Ian Gemp and Yoram Bachrach.
Summer 2020.
Google AI, Mountain View: Research Intern (Vision and Language Reinforcement Learning)
Worked with Alex Ku and Jason Baldridge.
Summer 2019.
Johns Hopkins University: Jelinek Summer Workshop on Speech and Language Technology (JSALT)
Worked with Ellie Pavlick, Sam Bowman and Tal Linzen.
Summer 2018.
Max Planck Institute: Cornell, Maryland, Max Planck Pre-doctoral Research School (CMMRS)
Summer 2018.
University of Pennsylvania: Undergraduate Researcher
Worked with Ani Nenkova and Byron Wallace.
Summer 2017-18.
Princeton University: Program in Algorithmic and Combinatorial Thinking (PACT)
Led by Rajiv Gandhi.
Summer 2016.
Rutgers University: Undergraduate Researcher
Worked with Rajiv Gandhi.
2016-2018.

What I'm Recently Upto

September 2024
I'm giving a talk on safety and ethical implications of LLMs at the M2L Summer School.
July 2024
I'm giving a talk at a CogSci workshop on understanding the conceptual structure of language models.
June 2024
I'm at the Mexican NLP Summer School co-located with NAACL and on a panel talking about PhDs in NLP.
November 2023
I'm giving a guest lecture at Brown CS 1460 on grounded language learning.
October 2023
I'm at the DIMACS workshop talking about language models and multi-agent reinforcement learning can benefit from each other.
In today's garden path sentences: The prime number few.