Grace Luo
I'm a final-year PhD student at
BAIR
advised by
Professor Trevor Darrell
. I've also worked closely with
Anna Rohrbach
,
Aleksander Holynski
,
Jacob Steinhardt
, and
Alec Radford
. Before that, I received my B.S. and M.S. degrees in EECS at UC Berkeley.
My research is on interpreting and controlling generative models. For example, I've worked on
re-purposing image generators
for computer vision tasks, and
meta-modeling language activations
for better LLM probing and steering.
Research
Selected
All
* denotes equal first authorship,
†
denotes equal advising
Learning a Generative Meta-Model of LLM Activations
Grace Luo
, Jiahai Feng, Trevor Darrell
†
, Alec Radford
†
, Jacob Steinhardt
†
ICML 2026
paper
/
website
/
code
Constantly Improving Image Models Need Constantly Improving Benchmarks
Jiaxin Ge*,
Grace Luo*
, Heekyung Lee, Nishant Malpani, Long Lian, XuDong Wang, Aleksander Holynski, Trevor Darrell, Sewon Min, David M. Chan
ICLR 2026
paper
/
website
/
code
Dual-Process Image Generation
Grace Luo
, Jonathan Granskog, Aleksander Holynski
†
, Trevor Darrell
†
ICCV 2025
paper
/
website
/
code
Vision-Language Models Create Cross-Modal Task Representations
Grace Luo
, Trevor Darrell, Amir Bar
ICML 2025
paper
/
website
/
code
Readout Guidance: Learning Control from Diffusion Features
Grace Luo
, Trevor Darrell, Oliver Wang, Dan B Goldman, Aleksander Holynski
CVPR 2024
(Highlight)
paper
/
website
/
code
Diffusion Hyperfeatures: Searching Through Time and Space for Semantic Correspondence
Grace Luo
, Lisa Dunlap, Dong Huk Park
†
, Aleksander Holynski
†
, Trevor Darrell
†
NeurIPS 2023
paper
/
website
/
code
Shape-Guided Diffusion with Inside-Outside Attention
Dong Huk Park*,
Grace Luo*
, Clayton Toste, Samaneh Azadi, Xihui Liu, Maka Karalashvili, Anna Rohrbach, Trevor Darrell
WACV 2024
paper
/
website
/
code
G^3: Geolocation via Guidebook Grounding
Grace Luo*
, Giscard Biamby*, Trevor Darrell, Daniel Fried, Anna Rohrbach
Findings of EMNLP 2022
paper
/
slides
/
code
Twitter-COMMs: Detecting Climate, COVID, and Military Multimodal Misinformation
Giscard Biamby*,
Grace Luo*
, Trevor Darrell, Anna Rohrbach
NAACL 2022
paper
/
slides
/
code
NewsCLIPpings: Automatic Generation of Out-of-Context Multimodal Media
Grace Luo
, Trevor Darrell, Anna Rohrbach
EMNLP 2021
(Oral Presentation)
paper
/
slides
/
code
Industry Experience
Google Labs
: Student Researcher
(Summer '23 - Spring '24)
Descript - Lyrebird Team
: Research Intern
(Summer '22)
Google Cloud - Vertex AI Team
: Software Engineering Intern
(Summer '21)
Other Experience
ASUC Student Government
: Chief Technology Officer
(Fall '20 - Spring '21)
Berkeleytime
: Frontend Developer & Team Lead
(Fall '18 - Spring '21)
HKN
: Student Relations Committee
(Spring '19 - Fall '19)
Academic Service
Workshop on Knowledge in Generative Models
: Co-organizer
(ECCV '24)
Intro to Machine Learning (CS189)
: Graduate Student Instructor
(Fall '24, Spring '25)