Andrei Bursuc

All correct Master David! 🏹

This is where dad got way more excited than kids at the figurine exhibition today 😅

CLIP's visual embedding projector is a few-shot cornucopia

💡Few-shot adaptation of VLMs by fine-tuning the last matrix of the vision encoder

📜https://arxiv.org/abs/2410.05270
💻code available ✅

by M. Fahes, @tuanhungvu.bsky.social, @abursuc.bsky.social, @ptrkprz.bsky.social, R. de Charette

a cat is playing in a hamster wheel in front of a television ALT: a cat is playing in a hamster wheel in front of a television

Scrolling back and forth through submissions in that infamous single-column 14-page template this week.

Deadline day today! Let’s go #eccv2026

Tune in to the live PhD defense of Loïck Chambon on "Efficient Representations for Autonomous Driving", a PhD co-advised by @mlia-isir.bsky.social & @valeoai.bsky.social
www.youtube.com/live/uFMAHn0...

At #CVPR2026, 16,092 submissions underwent the review process (this number excludes papers withdrawn or desk-rejected during the review process). The program committee recommended 4,090 papers for acceptance, resulting in an acceptance rate of 25.42%.

a volleyball player wearing a number 10 jersey ALT: a volleyball player wearing a number 10 jersey

Christian, every time I mention android vs ios recently 🙃

Researcher morale after three rejections in a row

Same here. My previous Android phone was tagging such numbers as potential spams and it was easy to just reject. No such service on iOS (at least not for free).
I don’t answer calls from unknown numbers anymore: if they really need me they call again or leave a voicemail.

Energized and inspired after our annual meetup to brainstorm on new exciting ideas and plan the projects ahead of us this year!
As always this is an excellent occasion to fit (almost) the entire team in a single photo

This year I'm teaching a new course on generative models for visual content (images, video, 3D, etc). It's mostly me rambling about recent papers, design choices I like/hate. The slides of the first lectures are here: davidpicard.github.io/teaching/
Use right arrow to navigate past the blank page.

Yeah! I discovered lichess just recently as I was looking for some solutions for learning and practice with my 11yo.

All the "you need to learn AI skills or you'll get left behind" things are patently nonsense. It's easy to use and only becomes easier to use over time. If there's skill it's in knowing what it does well and what is does poorly

a man is laying on a couch and crying . ALT: a man is laying on a couch and crying .

You caught me! Actually most of the times as AC I'm rather like that 😅

Are any of the ACs getting the first impulse of ending their messages to late reviewers with "Thank you for your attention to this matter"? 🙃

Bluesky Map Interactive map of 3.4 million Bluesky users, visualised by their follower pattern.

I made a map of 3.4 million Bluesky users - see if you can find yourself!

bluesky-map.theo.io

I've seen some similar projects, but IMO this seems to better capture some of the fine-grained detail

Hello nearest-neighbor! 👋

An email found me, and I was well at the time.

The ever increasing surge of verbosity.
Even authors are now more verbose than usual in message to ACs.

And yet again for #ausopen

Haha, I guess they use those non-formatted and messy references all the time on purpose to make sure I will never get to ask them to do that 😆

Google Scholar Button – Get this Extension for 🦊 Firefox (en-US) Download Google Scholar Button for Firefox. Lookup scholarly articles as you browse the web.

This one: addons.mozilla.org/en-US/firefo...

During deadline rushes when I'm cleaning up and looking up references just before submission I got blocked out of bot suspicion 😅

yeah, but so convenient to just click on your google scholar plugin :)

Yes, most often in fact. Now, other sources are not more accurate either, so you still have to check.
Changing `article={}` & journal={}` to `inproceedings={}` `booktitle={}` remains easier than parsing and deleting from other kB-sized bib entries

still, google scholar does the shortest exports out of all (fairly easy to check & minimize further for the paper).
compare that with bib entries that my students find: some go to give you the street name of the conference venue and the names of the parents of the program chairs 🙃

Maybe the most important thing I’ve learned over the past few years is that the solution to what looks like a collective action problem is to just start solving it. People will show up.

Seems like a bunch of people are joining BlueSky recently after the whole X situation, so maybe it's worth highlighting this again. Also, if you feel like you should be on this list, send me a message!

The bitter lesson of peer-reviewing?