10 citations · 10 across the 16 of their papers we have counts for
20 papers
The Self-Improvement Problem
Brian B. Moser, Stanislav Frolov, Tobias C. Nauen +2
With the rise of large language models, agentic systems are becoming increasingly autonomous, as they write code, call tools, revise prompts, and propose modifications to their own…
Layers Matter: Why Continual Learning Regularization Should Be Layer-Adaptive
Brian B. Moser, Ahmed Anwar, Tobias Christian Nauen +5
Continual learning regularizers like EWC fight forgetting by penalizing changes from previous-task parameters with per-parameter importance, typically diagonal Fisher values. Per-p…
LUMA: Benchmarking Segmentation via a Lightweight Universal Mask Adapter
Tobias Christian Nauen, Anosh Billimoria, Federico Raue +3
Comparing transformer backbones for image segmentation is confounded: each is paired with a different decoder, recipe, and pretraining, so reported differences rarely reflect the b…
The Gentle Collapse: Distributional Metrics for Continual Learning
Ahmed Anwar, Andreas Wagner, Federico Raue +2
Accuracy degradation is the standard metric for Catastrophic Forgetting (CF), however, it records only whether forgetting occurred or not. It saturates at the extremes and collapse…
OA-CutMix: Correcting the Label Bias of CutMix
Tobias Christian Nauen, Stanislav Frolov, Federico Raue +2
CutMix has become the de facto standard mixing augmentation, yet its label assignment rests on a flawed assumption: The area of the pasted patch faithfully reflects its semantic co…
TextTeacher: What Can Language Teach About Images?
Tobias Christian Nauen, Stanislav Frolov, Brian Bernhard Moser +3
The platonic representation hypothesis suggests that sufficiently large models converge to a shared representation geometry, even across modalities. Motivated by this, we ask: Can…
Hyperspherical Forward-Forward with Prototypical Representations
Shalini Sarode, Brian Moser, Joachim Folz +4
The Forward-Forward (FF) algorithm presents a compelling, bio-inspired alternative to backpropagation. However, while efficient in training, it has a computationally prohibitive in…
When Pretty Isn't Useful: Investigating Why Modern Text-to-Image Models Fail as Reliable Training Data Generators
Krzysztof Adamkiewicz, Brian Bernhard Moser, Stanislav Frolov +3
Recent text-to-image (T2I) diffusion models produce visually stunning images and demonstrate excellent prompt following. But do they perform well as synthetic vision data generator…
PRISM: Diversifying Dataset Distillation by Decoupling Architectural Priors
Brian B. Moser, Shalini Sarode, Federico Raue +6
Dataset distillation (DD) promises compact yet faithful synthetic data, but existing approaches often inherit the inductive bias of a single teacher model. As dataset size increase…
SubZeroCore: A Submodular Approach with Zero Training for Coreset Selection
Brian B. Moser, Tobias C. Nauen, Arundhati S. Shanbhag +4
The goal of coreset selection is to identify representative subsets of datasets for efficient model training. Yet, existing approaches paradoxically require expensive training-base…
HyperCore: Coreset Selection under Noise via Hypersphere Models
Brian B. Moser, Arundhati S. Shanbhag, Tobias C. Nauen +4
The goal of coreset selection methods is to identify representative subsets of datasets for efficient model training. Yet, existing methods often ignore the possibility of annotati…
ForAug: Mitigating Biases in Image Classification via Controlled Image Compositions
Tobias Christian Nauen, Brian Moser, Federico Raue +2
Large-scale image classification datasets exhibit strong compositional biases: objects tend to be centered, appear at characteristic scales, and co-occur with class-specific contex…
A Study in Dataset Distillation for Image Super-Resolution
Tobias Dietz, Brian B. Moser, Tobias Nauen +3
Dataset distillation aims to compress large datasets into compact yet highly informative subsets that preserve the training behavior of the original data. While this concept has ga…
Distill the Best, Ignore the Rest: Improving Dataset Distillation with Loss-Value-Based Pruning
Brian B. Moser, Federico Raue, Tobias C. Nauen +2
Dataset distillation has gained significant interest in recent years, yet existing approaches typically distill from the entire dataset, potentially including non-beneficial sample…
Zoomed In, Diffused Out: Towards Local Degradation-Aware Multi-Diffusion for Extreme Image Super-Resolution
Brian B. Moser, Stanislav Frolov, Tobias C. Nauen +2
Large-scale, pre-trained Text-to-Image (T2I) diffusion models have gained significant popularity in image generation tasks and have shown unexpected potential in image Super-Resolu…
Just Leaf It: Accelerating Diffusion Classifiers with Hierarchical Class Pruning
Arundhati S. Shanbhag, Brian B. Moser, Tobias C. Nauen +3
Diffusion models, celebrated for their generative capabilities, have recently demonstrated surprising effectiveness in image classification tasks by using Bayes' theorem. Yet, curr…
A Low-Resolution Image is Worth 1x1 Words: Enabling Fine Image Super-Resolution with Transformers and TaylorShift
Sanath Budakegowdanadoddi Nagaraju, Brian Bernhard Moser, Tobias Christian Nauen +3
Transformer-based architectures have recently advanced the image reconstruction quality of super-resolution (SR) models. Yet, their scalability remains limited by quadratic attenti…
Stochastic Control with Signatures
P. Bank, C. Bayer, P. P. Hager +2
This paper proposes to parameterize open loop controls in stochastic optimal control problems via suitable classes of functionals depending on the driver's path signature, a concep…
TaylorShift: Shifting the Complexity of Self-Attention from Squared to Linear (and Back) using Taylor-Softmax
Tobias Christian Nauen, Sebastian Palacio, Andreas Dengel
The quadratic complexity of the attention mechanism represents one of the biggest hurdles for processing long sequences using Transformers. Current methods, relying on sparse repre…
Which Transformer to Favor: A Comparative Analysis of Efficiency in Vision Transformers
Tobias Christian Nauen, Sebastian Palacio, Federico Raue +1
Self-attention in Transformers comes with a high computational cost because of their quadratic computational complexity, but their effectiveness in addressing problems in language…