Erik Wijmans

Thinking Machines

About Me

I am an AI research working at Thinking Machines. My research interests are focused on reinforcement learning, scaling systems, and their applications toward the emergence of intelligence. I currently work on scaling Tinker, our fine-tuning service that allows anyone to train at massive scale, and making fine-tuning models (including Inkling) easier and more effective.

Before Thinky, I was at Apple working with Vladlen Koltun. I received my PhD in computer science from the Georgia Institute of Technology in 2022. I was advised by Irfan Essa and Dhruv Batra and had the pleasure of working closely with Devi Parikh.

Interests

Embodied AI
Computer Vision
Reinforcement Learning

Education

PhD, Computer Science, 2022

Georgia Institute of Technology

Thesis: Emergence of Intelligent Navigation Behavior in Embodied Agents from Massive-Scale Simulation
Bachelor of Science, Computer Engineering, Summa Cum Laude, 2017

Washington University in St. Louis
Bachelor of Science, Engineering Physics, Magna Cum Laude, 2017

Juniata College

Select Publications

All Publications

1) Characterizing Emergent Intelligence

Robust Autonomy Emerges from Self-Play

Marco Cusumano-Towner*, David Hafner*, Alex Hertzberg*, Brody Huval*, Aleksei Petrenko*, Eugene Vinitsky*, Erik Wijmans*, Taylor Killian, Stuart Bowers, Ozan Sener, Philipp Krähenbühl, Vladlen Koltun

International Conference on Machine Learning (ICML), 2025.
Acceptance: 3,260 out of 12,107 submission = 26.9%

Paper

Does Spatial Cognition Emerge in Frontier Models?

Santhosh Kumar Ramakrishnan, Erik Wijmans, Philipp Krähenbühl, Vladlen Koltun

International Conference on Learning Representations (ICLR), 2025.
Acceptance: 3,704 out of 11,672 submissions = 31.7%

Paper

Emergence of Maps in the Memories of Blind Navigation Agents

Erik Wijmans, Manolis Savva, Irfan Essa, Stefan Lee, Ari Morcos, Dhruv Batra

International Conference on Learning Representations (ICLR), 2023.

Notable Top-5% (Oral), 91 out of 4900 submissions = top-1.8% of submitted papers

Outstanding Paper Award, 4 out of 4900 submissions = top-0.08%

Paper Code Project Press Coverage

VER: Scaling On-Policy RL Leads to Emergence of Navigation in Embodied Rearrangement

Erik Wijmans, Irfan Essa, Dhruv Batra

Neural Information Processing Systems (NeurIPS), 2022.
Acceptance: 2665 out of 10,411 submissions = 25.6%

Paper Code Project

2) Large-scale Machine Learning

Cut Your Losses in Large-Vocabulary Language Models

Erik Wijmans, Brody Huval, Alexander Hertzberg, Vladlen Koltun, Philipp Krähenbühl

International Conference on Learning Representations (ICLR), 2025.
Acceptance: 3,704 out of 11,672 submissions = 31.7%

Oral talk, 207 out of 11,672 submissions = top-1.8% of submitted papers

Paper Code

Large Batch Simulation for Deep Reinforcement Learning

Brennan Shacklett, Erik Wijmans, Aleksei Petrenko, Manolis Savva, Dhruv Batra, Vladlen Koltun, Kayvon Fatahalian

International Conference on Learning Representations (ICLR), 2021.
Acceptance: 860 out of 2997 submissions = 28%

Paper Code

DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Frames

Erik Wijmans, Abhishek Kadian, Ari Morcos, Stefan Lee, Irfan Essa, Devi Parikh, Manolis Savva, Dhruv Batra

ICLR, 2020.
687 out of 2594 submissions = 26.5%

Paper Code Video Press Coverage

3) 3D Datasets and Fast Simulation for Embodied AI and Robotics

Habitat 2.0: Training Home Assistants to Rearrange their Habitat

Andrew Szot, Alex Clegg, Eric Undersander, Erik Wijmans, Yili Zhao, John Turner, Noah Maestre, Mustafa Mukadam, Devendra Chaplot, Oleksandr Maksymets, Aaron Gokaslan, Vladimir Vondrus, Sameer Dharur, Franziska Meier, Wojciech Galuba, Angel Chang, Zsolt Kira, Vladlen Koltun, Jitendra Malik, Manolis Savva, Dhruv Batra

Neural Information Processing Systems (NeurIPS), 2021.

Spotlight talk, top 3% of 9122 submissions

Paper Code Dataset Project Press Coverage

Megaverse: Simulating Embodied Agents at One Million Experiences per Second

Aleksei Petrenko, Erik Wijmans, Brennan Shacklett, Vladlen Koltun

International Conference on Machine Learning (ICML), 2021.
1184 out of 5513 submissions = 21%

Paper Code Project Press Coverage

The Replica Dataset: A Digital Replica of Indoor Spaces

Julian Straub, Thomas Whelan, Lingni Ma, Yufan Chen, Erik Wijmans, Simon Green, Jakob J. Engel, Raul Mur-Artal, Carl Ren, Shobhit Verma, Anton Clarkson, Mingfei Yan, Brian Budge, Yajie Yan, Xiaqing Pan, June Yon, Yuyang Zou, Kimberly Leon, Nigel Carter, Jesus Briales, Tyler Gillingham, Elias Mueggler, Luis Pesqueira, Manolis Savva, Dhruv Batra, Hauke M. Strasdat, Renzo De Nardi, Michael Goesele, Steven Lovegrove, Richard Newcombe

arXiv, 2019.

Paper Code Dataset Press Coverage

Habitat: A Platform for Embodied AI Research

Manolis Savva*, Abhishek Kadian*, Oleksandr Maksymets*, Yili Zhao, Erik Wijmans, Bhavana Jain, Julian Straub, Jia Liu, Vladlen Koltun, Jitendra Malik, Devi Parikh, Dhruv Batra

International Conference on Computer Vision (ICCV), 2019.

Oral talk, top 187 of 4303 submissions = top-4.3%

Best Paper Award Nominee, top 11 of 4303 submissions = top-0.25%

Paper Code Dataset Project Press Coverage