Hi, I'm Gabe Sarch

I'm a fourth-year Ph.D. student at Carnegie Mellon University in the Neural Computation and Machine Learning joint PhD program under the supervision of Dr. Katerina Fragkiadaki and Dr. Mike Tarr. My work is supported by the National Science Foundation Graduate Research Fellowship. Previously, I received a B.S. in Biomedical Engineering from the University of Rochester, where I studied the marmoset visual system under Dr. Jude Mitchell. I interned at Microsoft Research during Summer 2024, working with Andy Wilson, Bala Kumaravel, and Vibhav Vineet.

 profile photo

I am on the job market for postdoctoral and industry research opportunities

       




Research

VLM Agents Generate Their Own Memories: Distilling Experience into Embodied Programs of Thought

GH Sarch   L Jang   MJ Tarr   K Marino   W Cohen   K Fragkiadaki  
NeurIPS 2024 Spotlight (Top 2%)
Paper Project Page Code

Open-Ended Instructable Embodied Agents with Memory-Augmented Large Language Models

GH Sarch   Y Wu   MJ Tarr   K Fragkiadaki  
EMNLP Findings 2023
🔥[NEW!] in ICLR 2024 Workshop on LLM Agents: HELPER-X achieves Few-Shot SoTA on 4 embodied AI benchmarks (ALFRED, TEACh, DialFRED, and the Tidy Task) using a single agent, with just simple modifications to the original HELPER.
Paper HELPER-X Project Page Code

Brain Dissection: fMRI-trained Networks Reveal Spatial Selectivity in the Processing of Natural Images

GH Sarch   MJ Tarr   K Fragkiadaki*   L Wehbe*  
*equal advising
NeurIPS 2023
Paper Project Page Code

3D View Prediction Models of the Dorsal Visual Stream

GH Sarch   HF Tung   A Wang   JS Prince   MJ Tarr
CCN 2023
Paper

TIDEE: Tidying Up Novel Rooms using Visuo-Semantic Commonsense Priors

2023 CVPR Embodied AI Rearrangement Challenge winner
GH Sarch   Z Fang   AW Harley   P Schydlo   MJ Tarr   S Gupta   K Fragkiadaki
ECCV 2022
Paper Project Page Code Blog

Beyond Fixation: detailed characterization of neural selectivity in free-viewing primates

JL Yates   SH Coop   GH Sarch   R Wu   D Butts   M Rucci  Jude Mitchell  
Nature Communications
Paper

Move to See Better: Towards Self-Improving Embodied Object Detection

GH Sarch*   Z Fang*   A Jain*   AW Harley   K Fragkiadaki
*equal contribution
BMVC 2021
Paper Project Page Code

See all my publications