Bethge Lab

All

46 repositories

sober-reasoning
Public
A Sober Look at Language Model Reasoning
HTML
•4•88•5•0•Updated Oct 9, 2025Oct 9, 2025
onebench
Public
[ACL'25] The official code for "ONEBench to Test Them All: Sample-Level Benchmarking Over Open-Ended Capabilities"
Python
•
MIT License
•0•4•0•0•Updated Jul 29, 2025Jul 29, 2025
imagecorruptions
Public
Python package to corrupt arbitrary images.
python data-augmentation pip3 image-corruptions
Python
•
Apache License 2.0
•71•451•3•1•Updated May 6, 2025May 6, 2025
model-vs-human
Public
Benchmark your model on out-of-distribution datasets with carefully collected human comparison data (NeurIPS 2021 Oral)
benchmark tensorflow pytorch toolbox robustness
Python
•55•355•3•0•Updated Apr 17, 2025Apr 17, 2025
CiteME
Public
CiteME is a benchmark designed to test the abilities of language models in finding papers that are cited in scientific texts.
lm-evaluation citation-attribution citation-dataset
Python
•
Other
•5•48•1•0•Updated Nov 1, 2024Nov 1, 2024
sort-and-search
Public
Code for the paper: "Efficient Lifelong Model Evaluation in an Era of Rapid Progress" [NeurIPS'24]
Python
•1•9•0•0•Updated Oct 17, 2024Oct 17, 2024
frequency_determines_performance
Public
Code for the paper: "No Zero-Shot Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance" [NeurIPS'24]
Jupyter Notebook
•
MIT License
•4•91•0•0•Updated Apr 29, 2024Apr 29, 2024
foolbox
Public
A Python toolbox to create adversarial examples that fool neural networks in PyTorch, TensorFlow, and JAX
python machine-learning tensorflow keras pytorch adversarial-examples adversarial-attacks jax
Python
•
MIT License
•434•2.9k•23•6•Updated Apr 3, 2024Apr 3, 2024
bwki-weekly-tasks
Public
BWKI Task of the week
Jupyter Notebook
•3•1•0•2•Updated Mar 20, 2024Mar 20, 2024
DataTypeIdentification
Public
Code for the ICLR'24 paper: "Visual Data-Type Understanding does not emerge from Scaling Vision-Language Models"
MIT License
•0•13•0•0•Updated Jan 17, 2024Jan 17, 2024
robustness
Public
Robustness and adaptation of ImageNet scale models. Pre-Release, stay tuned for updates.
machine-learning domain-adaptation
Python
•
Apache License 2.0
•6•137•1•1•Updated Jul 5, 2023Jul 5, 2023
game-of-noise
Public
Trained model weights, training and evaluation code from the paper "A simple way to make neural networks robust against diverse image corruptions"
Python
•
MIT License
•12•62•0•0•Updated May 24, 2023May 24, 2023
tools
Public
Shell
•0•6•0•0•Updated Aug 3, 2022Aug 3, 2022
slurm-monitoring-public
Public
Monitor your high performance infrastructure configured over slurm using TIG stack
python influxdb grafana slurm telegraf
Python
•
MIT License
•8•24•0•0•Updated Apr 26, 2022Apr 26, 2022
AnalysisBySynthesis
Public
Adversarially Robust Neural Network on MNIST.
Python
•
Apache License 2.0
•18•63•5•0•Updated Feb 4, 2022Feb 4, 2022
inference_results_v1.1
Public
Python
•
Other
•24•0•0•0•Updated Jan 15, 2022Jan 15, 2022
slow_disentanglement
Public
Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding
Jupyter Notebook
•
MIT License
•12•73•2•0•Updated Nov 17, 2021Nov 17, 2021
notorious_difficulty_of_comparing_human_and_machine_perception
Public
Code for the three case studies: Closed Contour Detection, Synthetic Visual Reasoning Test, Recognition Gap
Jupyter Notebook
•
MIT License
•3•9•0•5•Updated Sep 14, 2021Sep 14, 2021
InDomainGeneralizationBenchmark
Public
Python
•
Apache License 2.0
•6•34•0•0•Updated Aug 30, 2021Aug 30, 2021
stylize-datasets
Public
A script that applies the AdaIN style transfer method to arbitrary datasets
Python
•
Other
•38•163•1•1•Updated Jul 5, 2021Jul 5, 2021
testing_visualizations
Public archive
Code for the paper " Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations"
machine-learning deep-learning cnn visualizations interpretiability
Python
•2•10•0•0•Updated Jun 24, 2021Jun 24, 2021
openimages2coco
Public
Convert Open Images annotations into MS Coco format to make it a drop in replacement
Jupyter Notebook
•
MIT License
•20•116•3•0•Updated Aug 4, 2020Aug 4, 2020
DeepLabCut
Public
Markerless tracking of user-defined features with deep learning
Python
•
GNU Lesser General Public License v3.0
•1.8k•1•0•0•Updated Jul 3, 2020Jul 3, 2020
magapi-wrapper
Public
Wrapper around Microsoft Academic Knowledge API to retrieve MAG data
api publications wrapper-script
Python
•
Apache License 2.0
•3•10•0•0•Updated Jun 20, 2020Jun 20, 2020
siamese-mask-rcnn
Public
Siamese Mask R-CNN model for one-shot instance segmentation
object-detection one-shot-learning instance-segmentation few-shot-learning one-shot-instance-segmentation
Jupyter Notebook
•
Other
•60•350•10•0•Updated May 19, 2020May 19, 2020
mmdetection
Public
Fork of the MMDetection Toolbox containing the Robustness Benchmark from the paper "Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming" (merged)
Python
•
Apache License 2.0
•9.8k•6•0•0•Updated Aug 5, 2019Aug 5, 2019
robust-detection-benchmark
Public
Code, data and benchmark from the paper "Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming" (NeurIPS 2019 ML4AD)
game-of-thrones object-detection autonomous-driving pascal-voc cityscapes robustness distortions ms-coco corruptions
Jupyter Notebook
•
MIT License
•24•187•0•0•Updated Jul 25, 2019Jul 25, 2019
texture-vs-shape
Public
Pre-trained models, data, code & materials from the paper "ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness" (ICLR 2019 Oral)
R
•105•0•0•0•Updated Jul 18, 2019Jul 18, 2019
foolbox-zoo-dummy
Public
Python
•0•0•0•0•Updated May 19, 2019May 19, 2019
docker-deeplearning
Public
Development of new unified docker container (WIP)
Python
•
Other
•5•9•4•0•Updated May 7, 2019May 7, 2019