Keuper Labs
@keuper-labs.bsky.social
40 followers 66 following 27 posts
Research in #ComputerVision , #MachineLearning, #HPC and #Geophysics https://www.keuper-labs.org/
Posts Media Videos Starter Packs
Pinned
keuper-labs.bsky.social
Our #NeurIPS ORAL "MaxSup: Overcoming Representation Collapse in Label Smoothing", joint work with @cispa.de, is now on Arxiv: arxiv.org/pdf/2502.15798
Reposted by Keuper Labs
janiskeuper.bsky.social
Our new paper (accepted at #NeurIPS UrbanAI Workshop): "Real-time Prediction of Urban Sound Propagation
with Conditioned Normalizing Flows" arxiv.org/pdf/2510.04510 is part of our work using generative model to predict complex physics www.physics-gen.org
Reposted by Keuper Labs
janiskeuper.bsky.social
Our GCPR @gcpr-by-dagm.bsky.social paper (Oral) "Assessing Foundation Models for Mold Colony Detection with Limited Training Data" is now on Arxiv: arxiv.org/pdf/2510.00561
keuper-labs.bsky.social
Our #NeurIPS ORAL "MaxSup: Overcoming Representation Collapse in Label Smoothing", joint work with @cispa.de, is now on Arxiv: arxiv.org/pdf/2502.15798
Reposted by Keuper Labs
Reposted by Keuper Labs
janiskeuper.bsky.social
Following up on the discussions about prompt injections which were found in Arxiv paper sources - indicating that some authors try to manipulate LLM generate reviews: beyond the ethical implications, I was wondering if this actually works and did some experiments...

arxiv.org/abs/2509.10248
Prompt Injection Attacks on LLM Generated Reviews of Scientific Publications
The ongoing intense discussion on rising LLM usage in the scientific peer-review process has recently been mingled by reports of authors using hidden prompt injections to manipulate review scores. Sin...
arxiv.org
Reposted by Keuper Labs
shashank-agnihotri.bsky.social
🎉Our TMLR paper FlowBench is out!

Optical flow keeps getting better on benchmarks, but reliability and generalization remain a challenge for safety-critical tasks.

FlowBench provides the most comprehensive robustness benchmark to date.

📄 openreview.net/forum?id=Kh4...
💻 github.com/shashankskag...
Reposted by Keuper Labs
shashank-agnihotri.bsky.social
🚨 10 days left! Submit to L2S: Learning to Sense @ NeurIPS 2025 📡🤖

📅 Dec 6/7, San Diego, USA

🎤 Invited speakers from academia & industry

🌐 Details & CFP: sites.google.com/view/l2s-wor...

Let’s shape the future of how machines sense!
#L2S #NeurIPS2025 #NeurIPSWorkshop
keuper-labs.bsky.social
🚀 Calling all ML, CV and Sensor Enthusiasts!

Join us at the L2S: Learning to Sense Workshop at #NeurIPS2025 !

🗓 Date: December 6/7, 2025
📍 Location: San Diego, USA

sites.google.com/view/l2s-wor...
keuper-labs.bsky.social
Congratulations to @juliagrabinski.bsky.social for successfully defending her PhD thesis on "Multifaceted Analysis of Deep Convolutional Neural Networks and Novel Fourier Modules" 🥳🎓
juliagrabinski.bsky.social
PhD: DONE! ✅ If you are wondering how I feel, just check out these smiling pics. Incredibly thankful and happy! 😊 #PhDLife #PhDConpleted
Reposted by Keuper Labs
katharinaprasse.bsky.social
Poster is up and we are looking forward to the #ICML2025 poster session. Come join @patrickknab.bsky.social and me at Poster #W-214 presenting our work with @smarton.bsky.social, Christian Bartelt, and @margretkeuper.bsky.social @margretkeuper.bsky.social #UniMa
Reposted by Keuper Labs
katharinaprasse.bsky.social
Join us on Thursday 11-13 in poster hall West #214 to discuss image segments as concepts. #ICML2025 @patrickknab.bsky.social @smarton.bsky.social Christian bartelt @margretkeuper.bsky.social @keuper-labs.bsky.social
keuper-labs.bsky.social
Attending #ICML2025? Watch out for our paper "DCBM: Data-Efficient Visual Concept Bottleneck Models" presented by @katharinaprasse.bsky.social -> github.com/KathPra/DCBM
keuper-labs.bsky.social
Attending #ICML2025? Watch out for our paper "DCBM: Data-Efficient Visual Concept Bottleneck Models" presented by @katharinaprasse.bsky.social -> github.com/KathPra/DCBM
keuper-labs.bsky.social
Congratulations to @paulgavrikov.bsky.social for an excellent PhD defense today!
keuper-labs.bsky.social
2) AIM improves interpretability by learning to mask out irrelevant features in a self-supervised way (stay tuned for the pre-print)
keuper-labs.bsky.social
Two papers with @keuper-labs.bsky.social participation accepted at #ICCV2025:

1) Scientific figure generation with TikZero, it generates scientific figures from text as high-level, human-interpretable, and editable graphics programs -> arxiv.org/pdf/2503.11509
Reposted by Keuper Labs
janiskeuper.bsky.social
Presenting our poster #45 today at #CVPR in poster session 3: "PhysicsGen: Can Generative Models Learn from Images to Predict Complex Physical Relations?"

cvpr.thecvf.com/virtual/2025...
Reposted by Keuper Labs
tejaswinimedi.bsky.social
Excited to share our work at the WiCV Workshop, CVPR 2025! Catch our presentation today from 3:00–4:30 PM (Nashville time)!
keuper-labs.bsky.social
Now presenting at poster #377
keuper-labs.bsky.social
Now presenting at poster #452
keuper-labs.bsky.social
Presenting now (12 to 2pm) at poster #212
keuper-labs.bsky.social
Presenting now (12-2pm) at poster #226
keuper-labs.bsky.social
Corner Cases - How Size and Position of Objects Challenge ImageNet-Trained Models openreview.net/forum?id=B6l...

M. Fatima, S. Jung, @margretkeuper.bsky.social