About
I am a second-year Sc.M. student in Computer Science at
Brown University, advised by
George Konidaris,
Daniel Ritchie, and
Thomas Serre. I’m
spending my final year visiting the
Kempner Institute
at Harvard University, advised by
Qianqian Wang. I also
work closely with
Victor Boutin.
Previously, I earned my B.S. in Computer Science from
FHSU, conducted research at
NUS, and worked as an AI Engineer at
MPTC.
My research spans
visual representation learning, generative modeling, and world
models for robotics. Broadly, I aim to understand how visual and generative models
represent the structure and dynamics of the physical world, and how
these representations can support prediction and planning in embodied
agents.
Applying to PhD programs for Fall 2027 & seeking
research internships from Spring 2027.
Selected Research
Sergio Orozco, Chealyfey Vutha, Nandika Auluck,
Anthony Ramirez-Roca, and George Konidaris
CoRL 2026 Workshop · Do Robots Need World Models?
Planning with world models can be computationally expensive when
searching over large action spaces. We learn action priors that
propose promising trajectories and refine them with model
predictive control for robotic manipulation.
Paper
Poster
Chealyfey Vutha, Jorge Chang, Bastien Le Lan,
Victor Boutin, and Thomas Serre
NECV 2026 · New England Computer Vision Workshop
How does the balance between discriminative and generative
learning affect human perceptual alignment? We study this
continuum within a fixed energy-based model architecture and find
that alignment is strongest at intermediate settings.
Paper
Chealyfey Vutha, Zheyu Shi, and Daniel Ritchie
NECV 2026 · New England Computer Vision Workshop
Linear interpolation in generative models can produce perceptually
uneven transitions. We optimize interpolation paths using learned
visual representations to encourage smoother, more semantically
consistent transitions between generated images.
Slides
Talk Recording
(Brown login required)
Recent News
For undergraduates and first-year master’s students:
I'm happy to chat about life, careers, or AI/ML research. I set aside 30 minutes
weekly, prioritizing underrepresented students.
Kindly email me!
Show older news
Hide older news