SmallData Symposium 2024

Prof. Dr. 

Thomas Brox

Department of Computer Science, Faculty of Engineering, University of Freiburg
Georges-Köhler-Allee 052, 79110 Freiburg

Projects

Publications

Bratulić J, Mittal S, Hoffmann DT, Böhm S, Schirrmeister RT, Ball T, Rupprecht C, Brox T. Unlocking In-Context Learning for Natural Datasets Across Modalities, International Journal of Computer Vision. Jun 2026
Ging S, Walter S, Bratulić J, Dienert J, Bast H, Brox T. Using Knowledge Graphs to harvest datasets for efficient CLIP model training, 47th DAGM German Conference on Pattern Recognition (DAGM GCPR 2025). Jan 2026
Argus M, Bratulić J, Masnavi H, Velikanov M, Heppert N, Valada A, Brox T. cVLA: Towards Efficient Camera-Space VLAs, Conference on Robot Learning (CoRL) 2025. Dec 2025
Bratulić J, Mittal S, Brox T, Rupprecht C. On Geometric Understanding and Learned Data Priors in VGGT, arXiv. Dec 2025
Zhang B, Zhang Y, Zhu H, Yan S, Brox T, Boedecker J. Constrained Reinforcement Learning with Smoothed Log Barrier Function, Transactions on Machine Learning Research. Nov 2025
Kempf E, Schrodi S, Argus M, Brox T. When and How Does CLIP Enable Domain and Compositional Generalization? International Conference on Machine Learning. Oct 2025
Hindel J, Mohan R, Bratulić J, Cattaneo D, Brox T, Valada A. Label-Efficient LiDAR Semantic Segmentation with 2D-3D Vision Transformer Adapters, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Oct 2025
Gurung B, Hoffmann DT, Brox T. Common Data Properties Limit Object-Attribute Binding in CLIP, German Conference on Pattern Recognition. Aug 2025
Schrodi S, Schur J, Argus M, Brox T. Selective Concept Bottleneck Models Without Predefined Concepts, Transactions on Machine Learning Research. Jun 2025
Schrodi S, Hoffmann DT, Argus M, Fischer V, Brox T. Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Models, International Conference on Learning Representations (ICLR). May 2025
Hoffmann DT, Schrodi S, Bratulić J, Behrmann N, Fischer V, Brox T. Eureka-Moments in Transformers: Multi-Step Tasks Reveal Softmax Induced Optimization Problems, International Conference on Machine Learning. Jul 2024
Ging S, Bravo MA, Brox T. Open-ended VQA benchmarking of vision-language models by exploiting classification datasets and their semantic hierarchy, International Conference on Learning Representations 2024 (ICLR 2024). Feb 2024

Related Resources in the Small Data Companion

Argus M, et al. (2025) cVLA code repository
Model, algorithm or code
Brox T, et al. (2025) data-properties-clip-binding
Model, algorithm or code
Argus M, et al. (2025) Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Models (ICLR 2025 Oral)
Model, algorithm or code
Argus M, et al. (2025) MS COCO
Dataset
Argus M, et al. (2025) ImageNet
Dataset
Argus M, et al. (2025) MIT States
Dataset
Argus M, et al. (2025) UT Zappos
Dataset
Bast H, et al. (2026) EntityNet model
Model, algorithm or code
Bast H, et al. (2026) EntityNet dataset
Dataset
Bast H, et al. (2026) ImageNet
Dataset
Bast H, et al. (2026) iNaturalist 2019
Dataset
Bast H, et al. (2026) iNaturalist 2021
Dataset
Bast H, et al. (2026) Caltech-UCSD Birds-200-2011 (CUB-200-2011)
Dataset
Bast H, et al. (2026) Rare Species Dataset
Dataset
Bast H, et al. (2026) Common Objects in Context (COCO)
Dataset
Bast H, et al. (2026) Flickr30K
Dataset
Bast H, et al. (2026) Crossmodal-3600
Dataset
Bast H, et al. (2026) ImageNetV2
Dataset
Bast H, et al. (2026) ImageNet-R
Dataset
Bast H, et al. (2026) ObjectNet
Dataset
Bast H, et al. (2026) ImageNet-Sketch
Dataset
Bast H, et al. (2026) ImageNet-A
Dataset
Bast H, et al. (2026) Wikidata
Dataset
Bast H, et al. (2026) Project page
Website
Bast H, et al. (2026) Preprint
Original Preprint
Ball T, et al. (2026) Unlocking In-Context Learning for Natural Datasets Beyond Language Modelling (GCPR 2025 Best Paper Honorable Mention)
Model, algorithm or code
Ball T, et al. (2026) Unlocking In-Context Learning for Natural Datasets Beyond Language Modelling
Original Preprint
Boedecker J, et al. (2025) saferl-lib
Model, algorithm or code
Bratulić J, et al. (2025) VGGT Analysis github repository Contact for access
Model, algorithm or code
Behrmann N, et al. (2024) eurekaMoments
Model, algorithm or code
Bravo MA, et al. (2024) oVQA: Open-ended Visual Question Answering
Model, algorithm or code
Bravo MA, et al. (2024) Common Objects in Context (COCO)
Dataset
Bravo MA, et al. (2024) ImageNet
Dataset
Bravo MA, et al. (2024) Open-vocabulary Attribute Detection (OVAD)
Dataset
Bravo MA, et al. (2024) VGA: Visual Question Answering
Dataset
Bravo MA, et al. (2024) GQA: Question Answering on Image Scene Graphs
Dataset
Bravo MA, et al. (2024) Preprint
Original Preprint
Bravo MA, et al. (2024) Poster session website
Website
Bravo MA, et al. (2024) Slide deck
Slides from presentation
Bravo MA, et al. (2024) Poster
Poster from presentation
Argus M, et al. (2025) ImageNet
Dataset
Argus M, et al. (2025) CUB
Dataset
Argus M, et al. (2025) Places 365
Dataset
Argus M, et al. (2025) Official code
Model, algorithm or code
Bratulić J, et al. (2025) BALViT
Model, algorithm or code
Bratulić J, et al. (2025) BALVit
Website
Bratulić J, et al. (2025) Label-Efficient LiDAR Semantic Segmentation with 2D-3D Vision Transformer Adapters
Original Preprint
Argus M, et al. (2025) When and How Does CLIP Enable Domain and Compositional Generalization?
Model, algorithm or code
Argus M, et al. (2025) When and How Does CLIP Enable Domain and Compositional Generalization?
Original Preprint

Administrative Manager

Marc Schumacher

Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center –
University of Freiburg