publications

For a complete list of my publications I refer to my GoogleScholar Profile

2026

  1. CVPR
    Better than Average: Spatially-Aware Aggregation of Segmentation Uncertainty Improves Downstream Performance
    Vanessa E Guarino, Claudia Winklmayr, Jannik Franzen, and 7 more authors
    In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
    Equal contribution: D Kainmueller*, C Karg*, CT Lüth*
    HTML
  2. TMLR
    Finally Outshining the Random Baseline: A Simple and Effective Solution for Active Learning in 3D Biomedical Imaging
    Carsten T Lüth, Jeremias Traub, Kim-Celine Kahl, and 6 more authors
    Transactions on Machine Learning Research (TMLR), 2026
    Equal contribution: CT Lüth*, J Traub*
    arXiv HTML

2025

  1. TMLR
    nnActive: A Framework for Evaluation of Active Learning in 3D Biomedical Segmentation
    Carsten T Lüth, Jeremias Traub, Kim-Celine Kahl, and 6 more authors
    Transactions on Machine Learning Research (TMLR), 2025
    Equal contribution: CT Lüth*, J Traub*. Compute: 150k A100 GPU hours
    HTML Code

2024

  1. ICLR
    ValUES: A Framework for Systematic Validation of Uncertainty Estimation in Semantic Segmentation
    Carsten T Lüth, Kim-Celine Kahl, Maximilian Zenk, and 2 more authors
    In International Conference on Learning Representations (ICLR), 2024
    Oral Presentation (Top 1% of submissions). Equal contribution: CT Lüth*, KC Kahl*
    Awarded HTML Code
  2. NeurIPS
    Overcoming Common Flaws in the Evaluation of Selective Classification Systems
    Jeremias Traub, Till J Bungert, Carsten T Lüth, and 4 more authors
    In Advances in Neural Information Processing Systems (NeurIPS), 2024
    Spotlight Presentation (Top 5% of submissions)
    Awarded arXiv HTML Code
  3. NeurIPS
    Navigating the maze of explainable ai: A systematic approach to evaluating methods and metrics
    Lukas Klein, Carsten Lüth, Udo Schlegel, and 3 more authors
    In Advances in Neural Information Processing Systems (NeurIPS), 2024
    arXiv HTML
  4. Prediction of disease severity in COPD: a deep learning approach for anomaly-based quantitative assessment of chest CT
    Silvia D Almeida, Tobias Norajitra, Carsten T Lüth, and 8 more authors
    European radiology, 2024
    HTML
  5. Capturing COPD heterogeneity: anomaly detection and parametric response mapping comparison for phenotyping on chest computed tomography
    Silvia D Almeida, Tobias Norajitra, Carsten T Lüth, and 8 more authors
    Frontiers in Medicine, 2024
    HTML
  6. How do deep-learning models generalize across populations? Cross-ethnicity generalization of COPD detection
    Silvia D Almeida, Tobias Norajitra, Carsten T Lüth, and 8 more authors
    Insights into Imaging, 2024
    HTML
  7. TMLR
    SURE-VQA: Systematic Understanding of Robustness Evaluation in Medical VQA Tasks
    Kim-Celine Kahl, Selen Erkan, Jeremias Traub, and 4 more authors
    Transactions on Machine Learning Research (TMLR), 2024
    arXiv HTML
  8. Why context matters in vqa and reasoning: Semantic interventions for vlm input modalities
    Kenza Amara, Lukas Klein, Carsten T Lüth, and 3 more authors
    arXiv, 2024
    arXiv HTML
  9. NeurIPS WS
    Interactive Semantic Interventions for VLMs: A Human-in-the-Loop Investigation of VLM Failure
    Carsten T Lüth, Lukas Klein, Kenza Amara, and 1 more author
    In NeurIPS Workshop on Safe Generative AI, 2024
    Equal contribution: L Klein*, K Amara*, CT Lüth*
    HTML

2023

  1. NeurIPS
    Navigating the Pitfalls of Active Learning Evaluation: A Systematic Framework for Meaningful Performance Assessment
    Carsten T Lüth, Till J Bungert, Lukas Klein, and 1 more author
    In Advances in Neural Information Processing Systems (NeurIPS), 2023
    HTML Code
  2. ICLR
    A Call to Reflect on Evaluation Practices for Failure Detection in Image Classification
    Paul F Jaeger, Carsten T Lüth, Lukas Klein, and 1 more author
    In International Conference on Learning Representations (ICLR), 2023
    Oral Presentation (Top 1% of submissions)
    Awarded arXiv HTML Code
  3. CRADL: Contrastive Representations for Unsupervised Anomaly Detection and Localization
    Carsten T Lüth, David Zimmerer, Gregor Koehler, and 3 more authors
    In BVM Workshop, 2023
    arXiv HTML
  4. MICCAI
    cOOpD: Reformulating COPD Classification on Chest CT Scans as Anomaly Detection Using Contrastive Representations
    Carsten T Lüth, Silvia D Almeida, Tobias Norajitra, and 7 more authors
    In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2023
    STAR Award. Equal contribution: CT Lüth*, SD Almeida*
    Awarded HTML Code

2022

  1. Unsupervised anomaly detection in the wild
    David Zimmerer, Daniel Paech, Carsten T Lüth, and 3 more authors
    In BVM Workshop, 2022
    HTML

2019

  1. arXiv
    Guided Image Generation with Conditional Invertible Neural Networks
    Lynton Ardizzone, Carsten T Lüth, Jakob Kruse, and 2 more authors
    arXiv preprint arXiv:1907.02392, 2019
    400+ citations
    arXiv HTML

2018

  1. ICLR
    Analyzing Inverse Problems with Invertible Neural Networks
    Lynton Ardizzone, Jakob Kruse, Sebastian Wirkert, and 6 more authors
    2018
    My research on INN training methods during my time as student researcher informed this work. 700+ citations
    arXiv HTML