Who's Who of AI

David Picard

178 trust researcher @davidpicard.eurosky.social · 2,888 followers
AI research Vision & synthetic media Compute & infrastructure
Why they matter

Researcher with public evidence across AI research, Vision & synthetic media, Compute & infrastructure.

AI signals
8
past 30d
Sources
7
distinct domains
Discusiones
39
past 30d
Latest signal
10d ago
View every signal from David Picard →
Professor of Computer Vision/Machine Learning at Imagine/LIGM, École nationale des Ponts et Chaussées @ecoledesponts.bsky.social Music & overall happiness 🌳🪻 Born well below 350ppm 😬 mostly silly personal views 📍Paris 🔗 https://davidpicard.github.io/
What they're sharing

Articles & links

↻ David Picard reposted
Diane Larlus @dlarlus.bsky.social

We just released our #ECCV2026 paper on Model Merging for Computer Vision 🎓 arxiv.org/abs/2604.12935 Joint work w @pdejorge.bsky.social Cesar De Souza @bjoernmichele.bsky.social @mbsariyildiz.bsky.social @weinzaepfelp.bsky.social Florent Perronnin & @skamalas.bsky.social See P…

Task Alignment: A Simple Proxy for Practical Model Merging Across Diverse Vision Tasks arxiv.org
AI Weekly's analysis →
  • A NAVER Labs Europe paper accepted at ECCV 2026 introduces a 'task alignment proxy' for selecting model-merging hyperparameters without training the decoder for each candidate.
  • The method targets heterogeneous vision tasks with trainable decoders, moving beyond the CLIP image classification setup that dominated prior model merging work.
  • The authors claim the proxy speeds up hyperparameter selection by orders of magnitude while retaining downstream performance.
Read full analysis →
 View on Bluesky →

Babe, stop everything! New favorite paper of the year is out! kyutai.org/fid-lottery/ arxiv.org/abs/2606.20536

The FID Lottery: Quantifying Hidden Randomness in Generative-Model Evaluation arxiv.org
AI Weekly's analysis →
  • Retraining a model with a different seed moves its FID score 3.2x more than resampling from a fixed trained network.
  • FID coefficient of variation stays within a 1-2% band even as compute or model size increases.
  • The authors recommend treating any FID gap below roughly 1.3% CoV as inconclusive and requiring multi-seed error bars.
Read full analysis →
View on Bluesky · ♥ 26 ↻ 6 ↩ 2 · 2 from the directory shared this · 59d ago

The E.T.👽 workshop submission is finally open! We aim at promoting scientific theory building in deep representation learning 2 tracks: 1p storyline & regular 14p full paper Deadline: Aug 1st 🖥️ empiricaltheory.github.io 📜 openreview.net/group?id=the... Let's do science! @eccv…

ECCV 2026 Workshop ET | OpenReview openreview.net
View on Bluesky · ♥ 8 ↻ 4 ↩ 0 · 2 from the directory shared this · 42d ago
↻ David Picard reposted
Shubhendu Trivedi @shubhendu.bsky.social

Don't know the author, but have become quite a fan of her work. Is always quite cool and original (sometimes conceptually, sometimes in terms of theoretical machinery &c.) arxiv.org/abs/2407.02458

Statistical Advantages of Oblique Randomized Decision Trees and Forests arxiv.org
AI Weekly's analysis →
  • Eliza O'Reilly's paper proves oblique Mondrian forests achieve minimax optimal convergence rates on ridge-function data where axis-aligned trees cannot.
  • For general ridge functions, no weighting of axis-aligned splits can match the rate oblique splits obtain, regardless of covariate distribution.
  • The analysis uses random tessellation theory from stochastic geometry, tying convergence to the relevant feature subspace rather than ambient dimension.
Read full analysis →
 View on Bluesky →
↻ David Picard reposted
@si-cv-graphics.bsky.social

𝗦𝘂𝗿𝗳𝗹𝗼: 𝗖𝗼𝗻𝘀𝗶𝘀𝘁𝗲𝗻𝘁 𝟯𝗗 𝗦𝘂𝗿𝗳𝗮𝗰𝗲 𝗙𝗹𝗼𝘄 𝗠𝗼𝗱𝗲𝗹 𝘄𝗶𝘁𝗵 𝗚𝗹𝗼𝗯𝗮𝗹 𝗦𝘁𝗮𝘁𝗲 Antoine Guédon, Shu Nakamura, Nicolas Dufour ... Angjoo Kanazawa arxiv.org/abs/2606.13644 Trending on scholar-inbox.com

Surflo: Consistent 3D Surface Flow Model with Global State arxiv.org
AI Weekly's analysis →
  • Surflo encodes a variable number of unposed RGB views into a fixed global state of K=128 tokens, then decodes 3D surface points via flow-matching ODEs.
  • From a single encoder pass the model can sample any number of oriented surface points, up to roughly one million, without committing to a fixed grid.
  • The authors report state-of-the-art results across eight benchmarks, including a Tanks and Temples Chamfer Distance of 0.0056 and F1 of 86.40 on 8 views.
Read full analysis →
 View on Bluesky →
↻ David Picard reposted
@lebellig.bsky.social

The last project of my PhD is finally out! 🪴 It was a pleasure collaborating with Aimi on this work! We introduce A²BM: Alignment-Aware Bridge Matching, a new framework for image-to-image translation with weakly aligned image pairs. Paper 📄: arxiv.org/pdf/2607.16294

arxiv.org View on Bluesky →

1/ MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiency. Wed, Jul 8, 2026 • 10:30 AM – 12:15 PM KST 📜 arxiv.org/abs/2510.25897 🖥️ nicolas-dufour.github.io/miro/ 🧬 huggingface.co/nicolas-dufo... With @arrijitghosh.bsky.social and @lucasdegeorge.bsky.so…

nicolas-dufour/miro · Hugging Face huggingface.co
View on Bluesky · ♥ 8 ↻ 2 ↩ 1 · 43d ago
Their own posts

Recent commentary

Interestingly, I have no idea if the training of this generative model works or not 😅

View on Bluesky · ♥ 15 ↻ 0 ↩ 2 · 78d ago

During my vacation, I was staying at a (lovely old school) hotel that had this amazing AI slop deck of cards in one of the rooms. The more you look at it, the more it's uncanny. I should have asked to buy it because those will be the artefacts of our time. In 10y you won't be able to recognize them.

View on Bluesky · ♥ 11 ↻ 0 ↩ 2 · 27d ago

J'étais là, Gandalf, quand la station MIR devait s'écraser sur Paris. Bon, en vrai, je chargeais des camions dans un entrepôt pour gagner un peu de thunes. On est sorti regarder. J'avais que 2 paires de lunettes de soleil et j'y ai gagné un tout petit point gris au milieu de ma vision. Mais cool!

View on Bluesky · ♥ 2 ↻ 0 ↩ 1 · 5d ago
Their network

In David Picard's orbit

Center = David Picard. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.