computer vision | @UniofOxford, @ELLISforEurope member | phd @TU_Muenchen also on LinkedIn: https://nitter.gq/t.co/0xKyKz9F1s
Oxford, United Kingdom
Joined April 2023
- Tweets39
- Following157
- Followers422
- Likes289
Pinned Tweet
What if attention wasn't about matching tokens, but operating in function space?
Glad to share our #ICML2026 paper:
📄 Functional Attention: From Pairwise Affinities to Functional Correspondences
w/ @Jiefang_Xiao @GaoMaolin @stevenygd Daniel Cremers
📄 xjffff.github.io/funcattn/
Simon Weber retweeted
DINOv2 has never seen a caption, and Qwen3 has never seen an image. We still aligned their embedding spaces without a single image-caption pair.
It even works when the images and the captions come from different datasets.
Project page: dominik-schnaus.github.io/un…
⬇️
Simon Weber retweeted
📍 Come see our #ICML2026 poster on Functional Attention!
🗓️ Tuesday, July 7th
⏰ 10:30–12:15
📌 Hall A #3011
If you want to chat about attention in function space vs. pairwise affinities, come find us!
🧵 Paper: xjffff.github.io/funcattn/
What if attention wasn't about matching tokens, but operating in function space?
Glad to share our #ICML2026 paper:
📄 Functional Attention: From Pairwise Affinities to Functional Correspondences
w/ @Jiefang_Xiao @GaoMaolin @stevenygd Daniel Cremers
📄 xjffff.github.io/funcattn/
Simon Weber retweeted
@Jiefang_Xiao will be at the #ICML2026 poster, awaiting for exciting discussions! Come and chat with us.
📍 Come see our #ICML2026 poster on Functional Attention!
🗓️ Tuesday, July 7th
⏰ 10:30–12:15
📌 Hall A #3011
If you want to chat about attention in function space vs. pairwise affinities, come find us!
🧵 Paper: xjffff.github.io/funcattn/
📍 Come see our #ICML2026 poster on Functional Attention!
🗓️ Tuesday, July 7th
⏰ 10:30–12:15
📌 Hall A #3011
If you want to chat about attention in function space vs. pairwise affinities, come find us!
🧵 Paper: xjffff.github.io/funcattn/
What if attention wasn't about matching tokens, but operating in function space?
Glad to share our #ICML2026 paper:
📄 Functional Attention: From Pairwise Affinities to Functional Correspondences
w/ @Jiefang_Xiao @GaoMaolin @stevenygd Daniel Cremers
📄 xjffff.github.io/funcattn/
Simon Weber retweeted
There are three reasons why it's called "Functional Attention."
1⃣We make functions, instead of tokens, as first-class citizens for attention module.
2⃣It was inspired by the geometry processing concepts of functional maps.
3⃣It works!
What if attention wasn't about matching tokens, but operating in function space?
Glad to share our #ICML2026 paper:
📄 Functional Attention: From Pairwise Affinities to Functional Correspondences
w/ @Jiefang_Xiao @GaoMaolin @stevenygd Daniel Cremers
📄 xjffff.github.io/funcattn/
Simon Weber retweeted
Really exciting work on Functional Attention! This feels especially useful for learning operators on PDEs, physical simulations, and geometric data where the underlying structure matters.
What if attention wasn't about matching tokens, but operating in function space?
Glad to share our #ICML2026 paper:
📄 Functional Attention: From Pairwise Affinities to Functional Correspondences
w/ @Jiefang_Xiao @GaoMaolin @stevenygd Daniel Cremers
📄 xjffff.github.io/funcattn/
Simon Weber retweeted
We introduce a novel functional attention mechanism, with the insight inspired by geometry. Collaboration of #TUM, #Oxford and #UTAustin.
What if attention wasn't about matching tokens, but operating in function space?
Glad to share our #ICML2026 paper:
📄 Functional Attention: From Pairwise Affinities to Functional Correspondences
w/ @Jiefang_Xiao @GaoMaolin @stevenygd Daniel Cremers
📄 xjffff.github.io/funcattn/
Why a functional view of attention?
✅ Linear-time complexity
✅ Better generalization across resolutions and discretizations
✅ State-of-the-art results on operator learning, PDE modeling, point cloud segmentation, and OOD generalization
Code: github.com/xjffff/FUNCATTN
Simon Weber retweeted
@tumcvg goes #ICCV2025 in Hawaii! 🛫🌋
We are very proud of our students who will present five papers (+ 1 workshop) during the conference!
In particular, check out Back-on-track, which is an award candidate. (Congrats @wrchen530!)
Simon Weber retweeted
Looking forward to presenting our paper "Finsler multi-dimensional scaling" at #CVPR2025 on Sunday 10:30am, Poster 462!
We investigate a largely uncharted research direction in computer vision, aka Finsler manifolds...
Simon Weber retweeted
Excited to be attending CVPR 2025 this week in Nashville!
I’ll be presenting our recent work:
“4Deform: Neural Surface Deformation for Robust Shape Interpolation”
📍 Poster session: [13th June 4pm-6pm poster #111]
#CVPR2025 #computervision
Looking forward to presenting our paper "Finsler multi-dimensional scaling" at #CVPR2025 on Sunday 10:30am, Poster 462!
We investigate a largely uncharted research direction in computer vision, aka Finsler manifolds...
Check out our paper📝 arxiv.org/abs/2503.18010 and our video📹 youtube.com/watch?v=gvah8ll7…
Very glad to announce that our "Finsler Multi-Dimensional Scaling" paper, accepted at #CVPR2025, is now on Arxiv! arxiv.org/abs/2503.18010
We are thrilled that our group has twelve papers accepted at #CVPR2025! 🚀
Congratulations to all of our students for this great achievement! 🎉
For more details, check out: cvg.cit.tum.de
After #CVPR2024, we further investigate beyond-Euclidean computer vision and extend the fundamental task of dimensionality reduction to the largely uncharted Finsler manifolds, with core applications in data analysis and visualisation!
4
172
Simon Weber retweeted
We are thrilled that our group has twelve papers accepted at #CVPR2025! 🚀
Congratulations to all of our students for this great achievement! 🎉
For more details, check out: cvg.cit.tum.de
Simon Weber retweeted
We are presenting DiffCD tomorrow morning at #ECCV2024!
Come by poster 214 to discuss how to fit neural SDFs to point clouds, from a geometric perspective
Very excited to announce our paper: "DiffCD: A Symmetric Differentiable Chamfer Distance for Neural Implicit Surface Fitting" at #ECCV2024!
Paper: arxiv.org/abs/2407.17058
Code/project: github.com/linusnie/diffcd
Looking forward to present "PoVar" at #ECCV2024 tomorrow at 4:30pm, Poster 097 !
Check out our paper 📝 : arxiv.org/abs/2405.05079
our code 💻 : github.com/tum-vision/povar and our video 📹: youtube.com/watch?v=PlFrfTaM…
Very glad to announce that our paper "Power Variable Projection for Initialization-Free Large-Scale Bundle Adjustment" has been accepted at #ECCV2024!
Check out arxiv.org/abs/2405.05079
Simon Weber retweeted
Feel free to drop by our poster stand tomorrow afternoon (Thursday 13:30-15:00, hall c 4-9, poster number #1402): icml.cc/virtual/2024/poster/… 😁
🤔 Variational learning is often thought to be impractical
🔥 Plot twist: it actually works better than Adam!
Meet IVON, a new optimizer that brings the best out of variational learning – 🧵 (1/9) #NLProc #ICML2024
📰 arxiv.org/abs/2402.17641
youtu.be/TRNYnRRJBRg