iiinova — where your digital self keeps evolving

Igor Gilitschenski @igilitschenski.bsky.social · 16/09/2026

Huge congratulations to my phenomenal graduate student, Ziyi Wu, for completing his PhD at the University of Toronto. 🎉 Ziyi has done some of the pioneering work on endowing generative models for image and video generation with fine-grained control and better alignment. 1/n

120

Igor Gilitschenski @igilitschenski.bsky.social · 08/09/2026

If LLMs are great at generating code to control robots, their best use is probably helping generate controllers and, subsequently, data for pre-training robot models. Similar to how they currently help interface with game/physics engines used to help train video models.

000

Igor Gilitschenski @igilitschenski.bsky.social · 28/08/2026

Today was the last day of our brilliant research coordinator, Jazmin Diaz, at the University of Toronto. She kept @davelindell.bsky.social's lab and mine running smoothly. We can't imagine what things would be like without her. She is leaving giant shoes to fill and will be dearly missed. 1/2

130

Igor Gilitschenski @igilitschenski.bsky.social · 31/07/2026

📢 We are looking for Postdocs at @vectorinstitute.ai Join us to work with my group or any of our other brilliant researchers. Apply by Aug 31 to receive full consideration for the fall cohort: workforcenow.adp.com/mascsr/defau...

052

Reposted by Igor Gilitschenski

Claas Voelcker @cvoelcker.bsky.social · 30/07/2026

This is a robot failing to grasp a ball. Almost every robot lab produces clips like this daily… and almost all of them get thrown away. This is the most abundant but underused resource in robot learning. We’re collecting all of it now as “OopsieData”, please join us at oopsie-data.com! (1/13)

17413

Reposted by Igor Gilitschenski

U of T Department of Computer Science @uoftcompsci.bsky.social · 29/07/2026

A new method from @utoronto.ca computer scientists trains AI faster and more reliably using less data and computing power. This could help bring smarter robots and autonomous systems into everyday life. Learn more: uoft.me/cyu @cvoelcker.bsky.social @igilitschenski.bsky.social

Collage showing robotics simulations and a video game environment used to evaluate an AI training method developed by University of Toronto researchers.

273

Reposted by Igor Gilitschenski

University of Toronto @utoronto.ca · 23/07/2026

‘Nobel Prize of mathematics’: #UofT mathematician Jacob Tsimerman awarded prestigious Fields Medal 🏅 uoft.me/jacob

Jacob Tsimerman receives the Fields Medal (photo by Allie Ippolito)

2178

Igor Gilitschenski @igilitschenski.bsky.social · 04/06/2026

I am optimistic on 3d for robotics. It matters for policy learning, as some sensing modalities are inherently 3d; for HRI because words are not always enough; for scenario design in simulation & testing; and for bootstrapping training data.

000

Reposted by Igor Gilitschenski

Marcel Hussing @marcelhussing.bsky.social · 25/04/2026

At #ICLR2026 presenting our first poster in the morning on Relative Entropy Pathwise Policy Optimization. Stop by at #4613. 🧑‍🎓 @cvoelcker.bsky.social, @axelbrunnbauer.bsky.social, Michal Naumann, Pieter Abbeel, @ericeaton.bsky.social, Radu Grosu, @sologen.bsky.social @igilitschenski.bsky.social

0164

Igor Gilitschenski @igilitschenski.bsky.social · 20/04/2026

Excited to share that the code and weights for our #NeurIPS2025 relighting (UniRelight) and lighting estimation (LuxDiT) works are now public. 📢 Code 🧑‍💻 💡 github.com/nv-tlabs/Uni... 📷 github.com/nv-tlabs/Lux... Weights 🏋️‍♀️ 💡 huggingface.co/nvidia/UniRe... 📷 huggingface.co/nvidia/LuxDiT

020

Igor Gilitschenski @igilitschenski.bsky.social · 28/03/2026

Enjoyed my trip to Philadelphia to speak at @upenn.edu GRASP Lab Robotics Seminar this week. Thank you, Kostas Daniilidis, for hosting me. It was wonderful meeting with Antonio Loquercio, Rachel Holladay, Dinesh Jayaraman, Pratik Chaudhari, Lingjie Liu, and many of your brilliant mentees.

060

Igor Gilitschenski @igilitschenski.bsky.social · 05/03/2026

Every year around this time, I wish for introducing a matching procedure for CS grad school as in medical residency. This would be a mental health improvement for everyone involved. How are doctors better at this than computer scientists? Where is our occupational pride?

071

Igor Gilitschenski @igilitschenski.bsky.social · 19/02/2026

Had a great time speaking at Stanford's Center for Image System Engineering and Vision seminars over the last two weeks. 👨‍🏫 Thank you for hosting me, Gordon Wetzstein and Wenlong Huang.

030

Igor Gilitschenski @igilitschenski.bsky.social · 13/02/2026

🚀 Excited to share REPPO, a new on-policy RL agent! TL;DR: Replace PPO with REPPO for fewer hyperparameter headaches and more robust training. REPPO, led by @cvoelcker.bsky.social, will be presented at ICLR 2026. How does it work? 🧵👇

12510

Igor Gilitschenski @igilitschenski.bsky.social · 09/02/2026

Given the diverse work on world models, there is a debate on where to draw the line between world models and other dynamic systems models. I'd argue that, like LLMs for general-purpose language modelling, world models are not designed for a single specific task or phenomenon.

020

Igor Gilitschenski @igilitschenski.bsky.social · 08/01/2026

2025/2026 may be remembered as the years when many grad students "stopped" and many professors "resumed" programming.

010

Igor Gilitschenski @igilitschenski.bsky.social · 01/12/2025

I'll be at #NeurIPS2025 in San Diego from Thu to Sat, and I am looking for PostDocs in Embodied AI, particularly in world modeling and simulator learning. Please reach out if you are interested.

051

Igor Gilitschenski @igilitschenski.bsky.social · 22/11/2025

I'm looking for graduate students to join my group in fall 2026. We work at the intersection of Computer Vision, Deep Learning, and Robotics. The goal of our work is to create and understand organic 🍀 simulation systems, i.e., controllable data-curation engines built from real-world data. 1/n

1104

Igor Gilitschenski @igilitschenski.bsky.social · 25/10/2025

Together with our brilliant graduate researcher Ziyi Wu and our fantastic undergrad, Brayden Zhang, we had a wonderful time at #ICCV2025!

050

Igor Gilitschenski @igilitschenski.bsky.social · 10/10/2025

Interesting @anthropic.com work on LLM poisoning. I'd argue that the glass is half full, and this can be interpreted as a positive result: Regardless of model size, the same number of samples seems to be sufficient to learn a new "skill". www.anthropic.com/research/sma...

anthropic.com

A small number of samples can poison LLMs of any size

Anthropic research on data-poisoning attacks in large language models

010

Igor Gilitschenski @igilitschenski.bsky.social · 01/10/2025

Excited to have been appointed as a faculty member at @vectorinstitute.ai. 🎉 This would not have been possible without the amazing work of my collaborators and the brilliant students in my group within @uoftcompsci.bsky.social. We all look forward to deepening ties with the Vector community. ❤️

2110

Igor Gilitschenski @igilitschenski.bsky.social · 01/10/2025

Want to help train the next generation of leaders, researchers, and innovators in CS? We are looking for teaching stream faculty to join us at @uoftcompsci.bsky.social. Check out our job posting and join a team of brilliant colleagues and amazing students. 🇨🇦❤️💻 academicjobsonline.org/ajo/jobs/30410

academicjobsonline.org

University of Toronto, Department of Computer Science

Job #AJO30410, Assistant Professor, Teaching Stream - Computer Science, Department of Computer Science, University of Toronto, Toronto, Ontario, CA

020

Igor Gilitschenski @igilitschenski.bsky.social · 04/09/2025

Proud German advisor moment: Discovered a pair of Birkenstocks under a grad student's desk.

020

Reposted by Igor Gilitschenski

Anastasiia Pedan @pedanana.bsky.social · 19/06/2025

Would you be surprised to learn that many empirical implementations of value-aware model learning (VAML) algos, including MuZero, lead to incorrect model & value functions when training stochastic models 🤕? In our new @icmlconf.bsky.social 2025 paper, we show why this happens and how to fix it 🦾!

183

Igor Gilitschenski @igilitschenski.bsky.social · 10/06/2025

Thrilled to share the work that our lab will present at @cvprconference.bsky.social. Check out the papers and meet @kai-he.bsky.social, @yashkant.bsky.social, @dazitu616.bsky.social, and Toshiya Yura in Nashville at their poster sessions!

040

Igor Gilitschenski @igilitschenski.bsky.social · 02/06/2025

Ai4Science is doing a great job at applying AI to accurately model physics for one (or a few) phenomena at a time. Video models sacrifice such accuracy in favour of huge diversity. The big challenge for robot learning researchers is to strike a balance between these two.

000

Igor Gilitschenski @igilitschenski.bsky.social · 28/05/2025

In Germany, there is a tradition of creating funny hats for doctoral graduates. 🎓 @cvoelcker.bsky.social brought this tradition to my group and, together with Umangi Jain, spearheaded the construction of a masterpiece for our first PhD graduate, @ashmrz.bsky.social. 1/2

1122

Igor Gilitschenski @igilitschenski.bsky.social · 22/05/2025

Compared to huge budget cuts, barring Harvard from enrolling any international students may seem unimportant. The opposite is true. Graduate students are the primary generators of research. Staying world-class requires the world's brightest minds. www.cnn.com/2025/05/22/u...

cnn.com

020

Igor Gilitschenski @igilitschenski.bsky.social · 12/05/2025

Today is one of the most exciting days of my academic career: my first PhD student graduated. 🎉 Huge congrats, @ashmrz.bsky.social! 🥂 I am grateful for the opportunity to have worked with you and could not have wished for a better start. 🚀 1/4

180

Igor Gilitschenski @igilitschenski.bsky.social · 30/04/2025

Congrats on your election victory, @liberalca.bsky.social and @mark-carney.bsky.social! If I had one wish for Canada's new government, it would be this: stop delaying visas and make it easier for the world's top scientists, including research-focused graduate students, to come to Canada. 🍁

022

Igor Gilitschenski @igilitschenski.bsky.social · 30/04/2025

Perfect time to catch up on some missed-out reading.

030

Igor Gilitschenski @igilitschenski.bsky.social · 11/04/2025

A rare sighting of @csprofkgd.bsky.social, showing his students some cool @vscode.dev productivity hacks.

1131

Igor Gilitschenski @igilitschenski.bsky.social · 06/04/2025

Ending the #RSS AC meeting on a high note with @mgb45.bsky.social, Rudolf Lioutikov, Jana Tumova, and Abhinav Valada. Can’t wait to see all of you again soon!

091

Igor Gilitschenski @igilitschenski.bsky.social · 04/04/2025

Excited to attend the RSS AC meeting for the first time. It’s held at #MIT this year. Great organization by @lucacarlone.bsky.social!

120

Igor Gilitschenski @igilitschenski.bsky.social · 01/04/2025

Spotted @csprofkgd.bsky.social trying out his (new?) "AR" glasses at @vectorinstitute.ai. Being a prof in Computer Vision is hard work.

1101

Igor Gilitschenski @igilitschenski.bsky.social · 24/03/2025

When you board the plane, and find your @uoft.bsky.social CS colleague sitting next to you. Great connecting with you for the first time in person, Bo Wang!

050

Reposted by Igor Gilitschenski

Claas Voelcker @cvoelcker.bsky.social · 18/03/2025

In case somebody needs older algorithms in a new dress, I open sourced my jax implementation of MBPO. It also contains my own vagram algorithm github.com/cvoelcker/ja...

github.com

GitHub - cvoelcker/jax_mbpo: A jax-based reimplementation of MBPO

A jax-based reimplementation of MBPO. Contribute to cvoelcker/jax_mbpo development by creating an account on GitHub.

2163

Igor Gilitschenski @igilitschenski.bsky.social · 17/03/2025

Learning the hard way why people who switched to fish complain that incompatibility to bash can be a real issue. I'd still argue that the benefits outweigh the costs, but maybe this is just my stubbornness.

010

Igor Gilitschenski @igilitschenski.bsky.social · 03/03/2025

I am excited to share that my students @kai-he.bsky.social, @yashkant.bsky.social, Ziyi Wu, and Toshiya Yura, our previous research visitor from Sony, will present papers at #CVPR2025. 🎉 Check out their amazing work! 1/🧵

182

Reposted by Igor Gilitschenski

Andreas Geiger @andreasgeiger.bsky.social · 28/02/2025

In NAVSIM #NeurIPS24, we use simulation metrics for end-to-end driving by unrolling BEV abstractions of scenes for a few seconds. The result? Open-loop evaluation that is better aligned with closed-loop evaluation compared to traditional open loop metrics. github.com/autonomousvi...

184

Igor Gilitschenski @igilitschenski.bsky.social · 13/02/2025

Young Assistant Prof at his first faculty meeting before he learned that he is not at a hire-to-tenure institution.

010

Igor Gilitschenski @igilitschenski.bsky.social · 13/02/2025

The global transformer shortage is a very serious issue. Please don't be wasteful! To do our part, the new @vectorinstitute.ai policy is to use at most 3 heads per multi-head attention layer.

030

Reposted by Igor Gilitschenski

Claas Voelcker @cvoelcker.bsky.social · 11/02/2025

Do you want to get the most out of your samples, but increasing the update steps just destabilizes RL training? Our #ICLR2025 spotlight 🎉 paper shows that using the values of unseen actions causes instability in continuous state-action domains and how to combat this problem with learned models!

3207

Igor Gilitschenski @igilitschenski.bsky.social · 23/01/2025

In robot learning research with foundation models, we adapt models that have initially been trained with language modeling and content creation in mind. I wonder what similarly expensive models would look like if the target use case was robotics, to begin with?

030

Igor Gilitschenski @igilitschenski.bsky.social · 02/01/2025

Excited to be starting the new year with a big change in life. After 25 years of loyalty, I'll be switching away from bash. Still debating how big the switch will be and deciding between zsh and fish.

140

Reposted by Igor Gilitschenski

Cathy Wu @cathywu.bsky.social · 22/12/2024

Yes, ready for some thinking and reading time during the holidays!! Got a list of research questions and books critiquing my field loaded up and ready to go.

4276

Igor Gilitschenski @igilitschenski.bsky.social · 21/12/2024

While OpenAI folks say that they are running out of sufficiently difficult benchmarks, my robots still can't clean up my kitchen.

0140

Igor Gilitschenski @igilitschenski.bsky.social · 17/12/2024

For simulation to become the future of data, data has to become the future of simulation.

030