Computer Vision & Machine Learning researcher @naverlabseurope she/her - I have permanently moved there: 🦋
Grenoble, France
Joined June 2009
- Tweets865
- Following719
- Followers2.8K
- Likes5.6K
Diane Larlus retweeted
⛷❄️ Study by day, ski by dusk— AI meets the slopes!
🌟 Join us at the ELIAS-ELLIS-VISMAC Winter School on Deep Learning and Computer Vision 🌟
📍 Brunico, Italy
🗓 January 27-31, 2025
🔗 Event Website: elias-ai.eu/event/ellis-vism…
👉 Secure your spot now!
Diane Larlus retweeted
🌟 Keynote Spotlight from CoLLAs 2024 🌟
In "Lifelong Visual Representation Learning," Diane Larlus (@dlarlus) dives into continual learning for computer vision 🖼️ and explores how large pre-trained models are reshaping the field. She addresses challenges like catastrophic forgetting 🧠 and discusses methods to enable more robust, incremental learning 🔄✨.
📽️ Catch the full talk here: tinyurl.com/j847ytm9
#ComputerVision #LifelongLearning #AI #KeynoteSpeakers #CoLLAs2024
📢 Job Alert 📢 @naverlabseurope
My team is looking for a Research Scientist in Visual Representation Learning - More info tinyurl.com/3jz935fk
Diane Larlus retweeted
We are looking for a post-doc in machine learning for astrophysics/astronomy ! pepr-origins.fr/2024/10/09/p…
Diane Larlus retweeted
Do not miss an application deadline for #ALPS2025 on October 15!
lig-alps.imag.fr/index.php/a…
ALPS is an Advanced Language Processing School, held in French Alps, with wonderful speakers, inspiring discussions around NLP, and outdoor activities such as skiing and hiking!
Deadline for applications is Oct 15th but that would be for the 2025 edition @mgalle 😉 #ALPS2025
🗓️Event: March 30th-April 4th 2025
⛳️ Where: ⛷️ Aussois (French Alps)
➡️ Applications & info: lig-alps.imag.fr/
Diane Larlus retweeted
We presented the poster at the eXCV workshop at #ECCV2024 . Thank you to people stopping by. If you missed it and you are interested in explainability in CV (#xAI) here is our poster.
New paper on arXiv today. We study the problem of explaining with simple sentences the errors made by computer vision models - to "debug" their performance into the wild. Work led by Gabriela Csurka, with @TylerLHayes and @dlarlus (amazing team 🚀)
Paper: arxiv.org/abs/2408.04471
Diane Larlus retweeted
Do we really need multiple models for weight averaging (e.g., model soups)? Our answer: just two! We introduce Model Stock and share interesting findings on fine-tuned weights. #ECCV2024. Don’t miss our poster at AM 10:30 - 12:30, #110.
#ECCV2024 is already over .. Kudos to everybody involved in the organization ✨ I had a great time at the conference and really enjoyed all the discussions with old friends and new ones 🥰
Publicity chair spotted, surrounded by the women of NLE - @CSProfKGD @kgcs96
Diane Larlus retweeted
Ginger is presenting this morning how to derive a rich and versatile human pose representation by combining several modalities #ECCV2024
Our #ECCV2024 poster freshly up, and already some visitors! @mbsariyildiz and @WeinzaepfelP in action explaining our work on multiteacher distillation - see you at poster #10 !
Our #ECCV2024 paper ✨UNIC✨ is out !
Universal Classification Models via Multi-teacher Distillation
📰 arxiv.org/abs/2408.05088
💫 github.com/naver/unic
w @mbsariyildiz @WeinzaepfelP @ThomasLUC4S @skamalas - @naverlabseurope
Full thread 🔽
Diane Larlus retweeted
It is hard to prove discrimination when the algorithms indirectly infer some of your protected attributes.
In addition to that “new groups” are created by algs, e.g. dropping yours credit score.
E.g. insurance costs differently, depending if you have a letter in your address
Work from @Vinc3nt_Leroy Yohann Cabon and @JeromeRevaud
Paper: arxiv.org/abs/2406.09756
Code: github.com/naver/mast3r
