Skip to main content

Amil Dravid

Headshot of Amil Dravid
Program Year:
4
University:
University of California, Berkeley
Field of Study:
Computer Science
Advisor:
Alexei Efros
Degree(s):
B.S. Computer Science, Northwestern University, 2023
Personal URL:
avdravid.github.io

Practicum Experience(s)

Lawrence Livermore National Laboratory (2024)

Practicum Supervisor(s):
Shusen
Liu
Practicum Title:
Extracting and Aligning Knowledge from Foundation Models

Summary of Research

I am interested in enhancing our scientific understanding of deep learning. Lately, I have been thinking about linearity in weight space, learning dynamics, universality in representations, and the role of data. My goal is to leverage these insights to guide the design of better algorithms and models.

Annual Program Review Abstracts

Publications

1) Amil Dravid, Yasaman Bahri, Alexei A. Efros, Yossi Gandelsman. "Neuron Populations Exhibit Divergent Selectivity with Scale." NeurIPS 2026.
2) Nick Jiang*, Amil Dravid*, Alexei A. Efros, Yossi Gandelsman. "Vision Transformers Don't Need Trained Registers." NeurIPS 2025 (Spotlight).
3) Amil Dravid*, Yossi Gandelsman*, Kuan-Chieh Wang, Rameen Abdal, Gordon Wetzstein, Alexei A. Efros, Kfir Aberman. "Interpreting the Weight Space of Customized Diffusion Models." NeurIPS 2024.
4) Assaf Shocher, Amil Dravid, Yossi Gandelsman, Inbar Mosseri, Michael Rubinstein, Alexei A. Efros. "Idempotent Generative Network." ICLR 2024.
5) Amil Dravid*, Yossi Gandelsman*, Alexei A. Efros, Assaf Shocher. "Rosetta Neurons: Mining the Common Units in a Model Zoo." ICCV 2023.

Awards

1) NeurIPS 2025 Spotlight (Top 3% of Papers)
2) Goldwater Scholarship, 2021
3) CRA Outstanding Undergraduate Researcher Finalist, 2022
4) Caltech Summer Undergraduate Research Fellowship (SURF), 2022
5) Northwestern Summer Undergraduate Research Grant (SURG), 2020
6) Tau Beta Pi, 2022
7) Sigma Xi, 2021