Scholar
Shikhar Murty
Google Scholar ID: ubAcojQAAAAJ
Senior Research Scientist, Google DeepMind
Citations & Impact
Citations
1,078
H-index
14
i10-index
17
Publications
20
Co-authors
9
list available
Publications
Resume (English only)
Academic Achievements
- Co-organizer for REALM (Research on Agent Language Models) at ACL 2025; Gave talks about Building the learning-from-interaction pipeline for LLMs at institutions like Together AI, MIT, Harvard, and Brown; Organized a workshop at NeurIPS 2024 on 'System-2 Generalization at Scale'; Invited to lead a session on 'Intelligent Agents' at Foundation Capital AI Unconference, 2024 in San Francisco; Published several papers covering topics such as unsupervised learning, guided exploration, encoding recursive structure, etc.
Research Experience
- Part-time visitor at ServiceNow Research from Oct 2024 to Dec 2024, under the guidance of Alexandre Lacoste and Dzmitry Bahdanau, working on post-training for LLM browser agents; Research Intern at DeepMind from June 2023 to Feb 2024, mentored by Mandar Joshi, Kenton Lee, and Pete Shaw, focusing on unsupervised browser control with LLMs; Research Intern at Microsoft Research during Summer 2022, supervised by Marco Tulio Ribeiro and Scott Lundberg, working on fixing model bugs with language feedback.
Education
- Ph.D. in Computer Science at Stanford University from 2019 to present, advised by Prof. Christopher D. Manning; B.Tech in Electrical Engineering from Indian Institute of Technology, New Delhi from 2013 to 2017, with a thesis on Inference over Knowledge Bases with Deep Learning.
Background
- A final year CS PhD candidate working on Deep Learning and NLP, focusing on building LLMs that can generalize out-of-distribution through structured inductive biases or by interacting with their environments.
Miscellany
- Blog: https://skylerhallinan.com/; Google Scholar profile available; GitHub: MurtyShikhar; Twitter: @ShikharMurty
