Profile avatar
chriswolfvision.bsky.social
Principal Scientist at Naver Labs Europe, Lead of Spatial AI team. AI for Robotics, Computer Vision, Machine Learning. Austrian in France. https://chriswolfvision.github.io/www/
829 posts 5,811 followers 696 following
Regular Contributor
Active Commenter
Posts 22 Comments 28

🍿

submitted 7 hours ago • 0 comments

We have a new blog post on how we optimized end-to-end training of navigation in simulation with physical models, allowing fast and precise motion. The post is simplified, animated, and should be very accessible. Great work by the Spatial AI team, writing by Steeven, myself and the NLE Coms team.

submitted 2 days ago • 0 comments

The good news is that my Parisian colleagues will not be in mental state able to submit any jobs to our GPU cluster in the next couple of days...

submitted 5 days ago • 1 comment

Leaner Transformers: More Heads, Less Depth H. Saratchandran, @damienteney.bsky.social , S. Lucey

submitted 5 days ago • 3 comments

We had a family dinner yesterday (fish and frog restaurant at at the Saone river bank), including my lovely nieces on visit from Austria ... and we found out that we all seem to have the same taste in shoes 😅

submitted 10 days ago • 2 comments

🔊 Only a few days left to apply to the #PAISS2025 summer school !! This is a fantastic opportunity to learn and to network, especially for students 🎓

submitted 11 days ago • 0 comments

I am very late to this, but I find the work of Chris Olah + team on LLM explainability absolutely breathtaking. The amount of work is absolutely phenomenal. The attribution graphs are interactive and you can see features and links to training instances. transformer-circuits.pub/2025/attribu...

submitted 14 days ago • 1 comment

We have an open internship position on socially aware navigation, human aware world models etc. At @naverlabseurope.bsky.social in Grenoble (Meylan), France. careers.werecruit.io/en/naver-lab...

submitted 15 days ago • 0 comments

Lots of recent movies have insane long continuous shots over extremely complex scenes and I love it. "Sinners" is the latest example, I have seen it 2 days ago. An excellent movie! Others: - Babylon (2022), the "party scene" - 1917 (2019), obviously - Adolescence (2025 TV series), obviously

submitted 15 days ago • 3 comments

Are you based in the U.S. 🇺🇸 or Canada 🇨🇦 and thinking of doing research in Europe? ERC grants offer up to €4.5 million to support groundbreaking, investigator-driven frontier research in all fields. Watch the dedicated info session for researchers in your region👉 youtu.be/92QYink1Lik #ChooseEurope

submitted 17 days ago • 1 comment

When you implement a physics simulator and you don't get causality totally right... (from www.reddit.com/r/shittyaske...)

submitted 18 days ago • 0 comments

A conference officially adds LLM generated reviews to "peer" (sigh)-review. What could possibly go wrong, apart from lazy or incompetent reviewers being convinced by LLM output, and authors spending their rebuttal time on answering hallucinations which are probably not read by these same reviewers?

submitted 19 days ago • 5 comments

Added 2 conference deadlines: 3DV 2026: 3dvconf.github.io/2026/ August 18, 2025 (PT) Vancouver, Canada AAAI 2026: aaai.org/conference/a... Abs: July 25, 2025 (AoE) Paper: August 1, 2025 (AoE) Singapore

submitted 19 days ago • 0 comments

I don't know what's the best part of a post-submission deadline phase: the joy about the finished paper and work, the weekend off work, or the anticipation of finally being able to start the new stuff you have dreamed about doing.

submitted 20 days ago • 2 comments

submitted 20 days ago • 0 comments

I wonder how many of these NeurIPS registrations finally got submitted ... In any case, it will be rough for reviewers, ACs, PCs etc.

submitted 20 days ago • 0 comments

Anywhere on Earth (AoE) deadlines are actually quite nice for Europeans. The NeurIPS deadline is Friday noon. Pre-finished the paper yesterday evening, slept well, spending the morning polishing the paper. Easy-peasy, or as we say in French, "Finger in the nose!".

submitted 20 days ago • 0 comments

\vspace*{-3mm}

submitted 20 days ago • 0 comments

Ok, I am NOT a fan of the NeurIPS paper checklist. We are bit treated as toddlers here. 🙄

submitted 21 days ago • 2 comments

I am considering writing a dark rock ballade about cheating rogue scientists who overfit their ML results by removing training runs with non-optimal hyper parameters, and as performers I thought of Nick Cave and the bad seeds.

submitted 22 days ago • 1 comment

My wife tells me her students hide LLM usage in their reports (University software checks it automatically) by not directly pasting ChatGPT answers into them. They first paste into Word, then create a screenshot, paste the image into the report.

submitted 22 days ago • 2 comments

Overleaf is down 😬 Any correlation with a surge due to an upcoming deadline is probably unfounded.

submitted 22 days ago • 2 comments