Shivam Raval

Shivam Raval

AI Research Scientist

Generalist. Eternally curious.


Interests

  • AI Interpretability & Safety
  • High-dimensional Visualization
  • Interactive ML Systems

Get in touch: sraval@g.harvard.edu

AboutResearchResumeWritingVisuals

Research

Selected publications and workshop papers.

MoE Lens — An Expert Is All You Need

MoE Lens — An Expert Is All You Need

Practical downstream application of interpretability for compute optimization.

Marmik Chaudhari, Idhant Gulati, Nishkal Hundia, Pranav Karra,Shivam Raval

ICLR SLLM Workshop 2025★ Advisory capacity
Augmenting X-ray Astronomical Representations with Scientific Knowledge through Contrastive Learning

Augmenting X-ray Astronomical Representations with Scientific Knowledge through Contrastive Learning

Multimodal representation alignment for astronomical data.

Juan Rafael Martínez-Galarza, Nicolò Oreste Pinciroli Vago,Shivam Raval, Carolina Cuesta-Lazaro, Melanie Weber, David Alvarez-Melis, Alberto Accomazzi, Cecilia Garraffo, Joshua Knutson, Ryan Thill, Christopher B Green, Imantha Ahangama

ICLR Re-Align Workshop 2025
Hypertrix: An Indicatrix for High-Dimensional Visualizations

Hypertrix: An Indicatrix for High-Dimensional Visualizations

Visualizing and identifying anomalous distortion in visual projections.

Shivam Raval, Fernanda Viegas, Martin Wattenberg

IEEE VIS 2024★ Best Short Paper Award

Full publication list

Including citations, preprints, and workshop papers.

Google Scholar