Christian Wolf

Memory Caching: RNNs with growing memory Behrouz et al. arxiv.org/abs/2602.24281 Take different RNNs and, and at time t don't read out only M_t, but cached previous hidden memories. Neat and effective. Tested on different modern RNN variants.

Memory Caching: RNNs with Growing Memory arxiv.org

When projected onto a map, the attention maps correlate with scene structure and with affordances related to the scene structure. These maps have been computed on real nav data in a real environment (like ALL results in this paper). More results in the paper: arxiv.org/abs/260…

A scalar per patch from pre-trained ViTs enables fast moving navigation in the real world arxiv.org

This is work at Naver Labs Europe by - Philippe Weinzaepfel (@weinzaepfelp.bsky.social) - Yours truly, - Mert Bulent Sariyildiz (@mbsariyildiz.bsky.social) - Guillaume Bono, - Gianluca Monaci arxiv.org/abs/2606.21562 7/8

Compressing Observation History into Agent Memory: Distilling Transformers into Recurrent Transformers arxiv.org

Understanding the Impact of Geometric Foundation Models on Vision-Language-Action Models Yang,Lin,Martin-Martin,Labrie,Gayaka,Kuo,Carlone Put VGGT into VLA. Early fusion (before VLM) > late fusion (after VLM) Impact of 3D (VGGT) highest in single camera setup (!?!) arxiv.org/a…

Understanding the Impact of Geometric Foundation Models on Vision-Language-Action Models arxiv.org

ECCV 2026 Workshop on Embodied Multimodal Reasoning in Physical Environments emr-workshop.github.io Sept 8th, 2026 Full paper DL has passed, but extended abstracts can still be submitted: DL July 24, 2026 (23:59 AoE)

EMR Workshop @ ECCV 2026 emr-workshop.github.io

The person who pitched gen-AI to Hayao Miyazaki the way it was done here didn't do the world a favour. That's like pitching the interest of books by showing examples of random words put together. No wonder he was disgusted. www.youtube.com/watch?v=TGwC...

Hayao Miyazaki's thoughts on an artificial intelligence youtube.com

AI bubble vs. FIFA world cup vs. Summer Heat wave

For your Embodied AI task you want a recurrent model with constant complexity per step, but you don't want to lose the power of transformers (which store the full obs history and attend to it)? Do not despair, we have your back. We distill transformers into recurrent transformers 1/8

My Dad made fun of me for my use of "please" when talking to Claude. His point: the AI does not care. My point: this is not about the AI but about me.

Excellent Keynote by @akorba.bsky.social at Cap-Rfiap in Montpellier, France, on machine learning and distances between distributions.

Goodhart's law states that "When a measure becomes a target, it ceases to be a good measure". Conference peer-review is a measure, arguably now mostly implemented through LLMs, and will become a target. Ergo, future academic research will target what Claude finds interesting and acceptable.

It is 2030 and we ran out of English phrases which we can still use because of they are not frequently employed by LLMs.

ICRA panel on the impact of AI and the paper avalanche on robotics conferences and journals. The same problems arrise in ML and in CV of course.

In 2026 the accepted papers are the ones which were reviewed by the same LLM which the authors had used to check the paper before submission.

In the book "The Diamond Age" (1995) by Neal Stephenson, a working class child jumps social classes with the help of a unique interactive book, a personal trainer which answer all her questions and allows her to learn first. For centuries we waited for this. Anti-AI crowd: AI is anti teaching!! 😡

People: trees are connected to each other and they can react to insects by releasing chemical signals. They span a big consciousness. Also people: an artificial intelligence can never ever be conscious.