220 citations · 788 across the 33 of their papers we have counts for
35 papers
Advancing Open and Reproducible Relational Learning: RelArena-, TabPFN-Rel and RPI
Adrian Hayler, Klemens Flöge, Alan Arazi +44
This first release of Prior Labs in relational learning shows our continued commitment to open science. We open-source three pieces of software that we expect to accelerate researc…
A Family of LLMs Liberated from Static Vocabularies
Aleph Alpha, :, Adnen Abdessaied +35
Tokenization is a central component of natural language processing in current large language models (LLMs), enabling models to convert raw text into processable units. Although lea…
SOMBRERO: Measuring and Steering Boundary Placement in End-to-End Hierarchical Sequence Models
Pit Neitemeier, Alessio Serra, Jiaze Li +3
Hierarchical sequence models replace fixed tokenization with learned segmentations that compress long byte sequences for efficient autoregressive modeling. While recent end-to-end…
Depth-Recurrent Attention Mixtures: Giving Latent Reasoning the Attention it Deserves
Jonas Knupp, Jan Hendrik Metzen, Jeremias Bohn +2
Depth-recurrence facilitates latent reasoning by sharing parameters across depths. However, prior work lacks combined FLOP-, parameter-, and memory-matched baselines, underutilizes…
Single-Pass Object-Focused Data Selection
Niclas Popp, Dan Zhang, Jan Hendrik Metzen +2
While unlabeled image data is often plentiful, the costs of high-quality labels pose an important practical challenge: Which images should one select for labeling to use the annota…
Retinex-Diffusion: On Controlling Illumination Conditions in Diffusion Models via Retinex Theory
Xiaoyan Xing, Vincent Tao Hu, Jan Hendrik Metzen +3
This paper introduces a novel approach to illumination manipulation in diffusion models, addressing the gap in conditional image generation with a focus on lighting conditions. We…
Label-free Neural Semantic Image Synthesis
Jiayi Wang, Kevin Alexander Laube, Yumeng Li +4
Recent work has shown great progress in integrating spatial conditioning to control large, pre-trained text-to-image diffusion models. Despite these advances, existing methods desc…
Zero-Shot Distillation for Image Encoders: How to Make Effective Use of Synthetic Data
Niclas Popp, Jan Hendrik Metzen, Matthias Hein
Multi-modal foundation models such as CLIP have showcased impressive zero-shot capabilities. However, their applicability in resource-constrained environments is limited due to the…
Identification of Fine-grained Systematic Errors via Controlled Scene Generation
Valentyn Boreiko, Matthias Hein, Jan Hendrik Metzen
Many safety-critical applications, especially in autonomous driving, require reliable object detectors. They can be very effectively assisted by a method to search for and identify…
Identifying Systematic Errors in Object Detectors with the SCROD Pipeline
Valentyn Boreiko, Matthias Hein, Jan Hendrik Metzen
The identification and removal of systematic errors in object detectors can be a prerequisite for their deployment in safety-critical applications like automated driving and roboti…
AutoCLIP: Auto-tuning Zero-Shot Classifiers for Vision-Language Models
Jan Hendrik Metzen, Piyapat Saranrittichai, Chaithanya Kumar Mummadi
Classifiers built upon vision-language models such as CLIP have shown remarkable zero-shot performance across a broad range of image classification tasks. Prior work has studied di…
Identification of Systematic Errors of Image Classifiers on Rare Subgroups
Jan Hendrik Metzen, Robin Hutmacher, N. Grace Hua +2
Despite excellent average-case performance of many image classifiers, their performance can substantially deteriorate on semantically coherent subgroups of the data that were under…
Certified Defences Against Adversarial Patch Attacks on Semantic Segmentation
Maksym Yatsura, Kaspar Sakmann, N. Grace Hua +2
Adversarial patch attacks are an emerging security threat for real world deep learning applications. We present Demasked Smoothing, the first approach (up to our knowledge) to cert…
Give Me Your Attention: Dot-Product Attention Considered Harmful for Adversarial Patch Robustness
Giulio Lovisotto, Nicole Finnie, Mauricio Munoz +2
Neural architectures based on attention such as vision transformers are revolutionizing image recognition. Their main benefit is that attention allows reasoning about all parts of…
Neural Architecture Search for Dense Prediction Tasks in Computer Vision
Thomas Elsken, Arber Zela, Jan Hendrik Metzen +4
The success of deep learning in recent years has lead to a rising demand for neural network architecture engineering. As a consequence, neural architecture search (NAS), which aims…
Meta-Learning the Search Distribution of Black-Box Random Search Based Adversarial Attacks
Maksym Yatsura, Jan Hendrik Metzen, Matthias Hein
Adversarial attacks based on randomized search schemes have obtained state-of-the-art results in black-box robustness evaluation recently. However, as we demonstrate in this work,…
Bag of Tricks for Neural Architecture Search
Thomas Elsken, Benedikt Staffler, Arber Zela +2
While neural architecture search methods have been successful in previous years and led to new state-of-the-art performance on various problems, they have also been criticized for…
Test-Time Adaptation to Distribution Shift by Confidence Maximization and Input Transformation
Chaithanya Kumar Mummadi, Robin Hutmacher, Kilian Rambach +3
Deep neural networks often exhibit poor performance on data that is unlikely under the train-time data distribution, for instance data affected by corruptions. Previous works demon…
Does enhanced shape bias improve neural network robustness to common corruptions?
Chaithanya Kumar Mummadi, Ranjitha Subramaniam, Robin Hutmacher +3
Convolutional neural networks (CNNs) learn to extract representations of complex features, such as object shapes and textures to solve image recognition tasks. Recent work indicate…
Efficient Certified Defenses Against Patch Attacks on Image Classifiers
Jan Hendrik Metzen, Maksym Yatsura
Adversarial patches pose a realistic threat model for physical world attacks on autonomous systems via their perception component. Autonomous systems in safety-critical domains suc…
Meta Adversarial Training against Universal Patches
Jan Hendrik Metzen, Nicole Finnie, Robin Hutmacher
Recently demonstrated physical-world adversarial attacks have exposed vulnerabilities in perception systems that pose severe risks for safety-critical applications such as autonomo…
Increasing the Robustness of Semantic Segmentation Models with Painting-by-Numbers
Christoph Kamann, Burkhard Güssefeld, Robin Hutmacher +2
For safety-critical applications such as autonomous driving, CNNs have to be robust with respect to unavoidable image corruptions, such as image noise. While previous works address…
Adversarial and Natural Perturbations for General Robustness
Sadaf Gulshad, Jan Hendrik Metzen, Arnold Smeulders
In this paper we aim to explore the general robustness of neural network classifiers by utilizing adversarial as well as natural perturbations. Different from previous works which…
Meta-Learning of Neural Architectures for Few-Shot Learning
Thomas Elsken, Benedikt Staffler, Jan Hendrik Metzen +1
The recent progress in neural architecture search (NAS) has allowed scaling the automated design of neural architectures to real-world domains, such as object detection and semanti…