Tessera
Folgen

Vicky Kalogeiton

@vickykalogeiton.bsky.social

Assistant Professor at Ecole Polytechnique, IP_Paris// Before: Oxford_VGG, Inria Grenoble // multimodality, genAI enthusiast // happy mum+dog_mum // opinions: mine

10 Posts764 Follower268 Folgt
Lucas Degeorge@lucasdegeorge.bsky.social · 3mo ago

🎉 Our work MIRO is accepted to #ICML2026 @icmlconf.bsky.social We integrate human preferences directly during pretraining with multi-reward conditioning. ⚡MIRO is 19x faster than baselines and 370x cheaper at inference! 🤗 Try out the models: huggingface.co/spaces/nicol... See you in Seoul 🇰🇷 !

MIRO - a Hugging Face Space by nicolas-dufour

Multi-reward conditioned text-to-image diffusion (ICML 2026)

huggingface.co

Nicolas Dufour@nicolasdufour.bsky.social · 3mo ago

Thrilled to share that MIRO is accepted to ICML 2026 @icmlconf.bsky.social ! 🎉 By training on the reward scores, we can simply condition the model on high rewards at inference time to guarantee top-tier, aligned outputs. We’ve updated our paper with some additional results!

David Picard@davidpicard.eurosky.social · 3mo ago

🥳Accepted to ICML, by the way! Old paper here: arxiv.org/abs/2510.25897 (new version up soon) Going to pressure @nicolasdufour.bsky.social to release the code and models! 😅

Nicolas Dufour@nicolasdufour.bsky.social · 9mo ago

We introduce MIRO: a new paradigm for T2I model alignment integrating reward conditioning into pretraining, eliminating the need for separate fine-tuning/RL stages. This single-stage approach offers unprecedented efficiency and control. - 19x faster convergence ⚡ - 370x less FLOPS than FLUX-dev 📉

Vicky Kalogeiton@vickykalogeiton.bsky.social · 10mo ago

Very proud of our recent work, kudos to the team! Read @davidpicard.bsky.social’s excellent post for more details or the paper arxiv.org/pdf/2502.21318

arxiv.org

David Picard@davidpicard.eurosky.social · 10mo ago

🚨Updated: "How far can we go with ImageNet for Text-to-Image generation?" TL;DR: train a text2image model from scratch on ImageNet only and beat SDXL. Paper, code, data available! Reproducible science FTW! 🧵👇 📜 arxiv.org/abs/2502.21318 💻 github.com/lucasdegeorg... 💽 huggingface.co/arijitghosh/...

Dima Damen@dimadamen.bsky.social · 11mo ago

As an SAC for @neuripsconf.bsky.social, I don't agree with PCs approach to reject papers based on ranking. I ranked my accepted papers as requested and explicitly stated that I support the acceptance of all papers. I was not given an explanation of why papers at the end of the rank were rejected.

#ICCV2025@iccv.bsky.social · last yr.

#ICCV2025 is deeply committed to promoting diversity, equity, and inclusion within our community. As part of this commitment, travel support is available to help broaden participation. Applications will be reviewed on a rolling basis until August 20, 2025 (anywhere on Earth).

Bild
Vicky Kalogeiton@vickykalogeiton.bsky.social · last yr.

Join us at CVPR@Paris on June 6th! It's a local alternative to CVPR, aiming at addressing some sustainability and inclusivity concerns. All students and researchers are welcome to attend! Registration free: cvprinparis.github.io/CVPR2025InPa...

CVPR@Paris 2025 June 6th

cvprinparis.github.io

David Picard@davidpicard.eurosky.social · last yr.

🔥🔥🔥 CV Folks, I have some news! We're organizing a 1-day meeting in center Paris on June 6th before CVPR called CVPR@Paris (similar as NeurIPS@Paris) 🥐🍾🥖🍷 Registration is open (it's free) with priority given to authors of accepted papers: cvprinparis.github.io/CVPR2025InPa... Big 🧵👇 with details!

Vicky Kalogeiton@vickykalogeiton.bsky.social · last yr.

Text-to-image models are trained on billions of data. But, is it necessary? Our "How far can we go with ImageNet for T2I generation?‬" @lucasdegeorge.bsky.social @arrijitghosh.bsky.social @nicolasdufour.bsky.social @davidpicard.bsky.social shows that no, if we are careful arxiv.org/abs/2502.21318

Bild
David Picard@davidpicard.eurosky.social · last yr.

🚨 New preprint! How far can we go with ImageNet for Text-to-Image generation? w. @arrijitghosh.bsky.social @lucasdegeorge.bsky.social @nicolasdufour.bsky.social @vickykalogeiton.bsky.social TL;DR: Train a text-to-image model using 1000 less data in 200 GPU hrs! 📜https://arxiv.org/abs/2502.21318 🧵👇

Vicky Kalogeiton@vickykalogeiton.bsky.social · last yr.

Similar for Polytechnique, part of Institut Polytechnique de Paris: future open positions in AI for junior researchers. They come with a good starting package and research independence. Contact me by email or DM even for informal discussions.

David Picard@davidpicard.eurosky.social · last yr.

🚨 Future open positions in AI at my home institution* targeted towards young researchers. These are tenure track positions with a good starting package, aimed at guiding you towards an ERC StG within 4 years. Feel free to contact me in DM. *ENPC, part of Institut Polytechnique de Paris