Accepted Papers

Assessing Visual Privacy Risks in Multimodal AI: A Novel Taxonomy-Grounded Evaluation of Vision-Language Models 

• Efthymios Tsaprazlis (University of Southern California)

• Tiantian Feng (University of Southern California)

• Anil Ramakrishna (Meta)

• Rahul Gupta (Amazon)

• Shrikanth Narayanan (University of Southern California)

Monte Carlo Stochastic Depth for Uncertainty Estimation in Deep Learning

• Adam T. Müller (Heilbronn University of Applied Sciences)

• Tobias Rögelein (Heilbronn University of Applied Sciences)

• Nicolaj C. Stache (Heilbronn University of Applied Sciences)

Lost in Fog: Sensor Perturbations Expose Reasoning Fragility in Driving VLAs

• Abhinaw Priyadershi (NVIDIA)

• Jelena Frtunikj (NVIDIA)

Generative Texture Diversification of 3D Pedestrians for Robust Autonomous Driving Perception

• Arka Bhowmick (BIT Technology Solutions)

• Enes Özeren (Ludwig-Maximilians-Universität München)

• Ahmed Abdullah (Technische Hochschule Mannheim)

• Oliver Wasenmüller (Technische Hochschule Mannheim)

Long-Tailed Backdoor Attack Using Dynamic Data Augmentation Operations

• Lu Pang (State University of New York at Stony Brook)

• Tao Sun (Amazon)

• Weimin Lyu (Amazon)

• Haibin Ling (Westlake University)

• Chao Chen (State University of New York at Stony Brook)

Forecasting the Past: Gradient-Based Distribution Shift Detection in Trajectory Prediction

• Michele De Vita (Friedrich-Alexander Universität Erlangen-Nürnberg)

• Julian Wiederer (Mercedes-Benz)

• Vasileios Belagiannis (Friedrich-Alexander Universität Erlangen-Nürnberg)

Does Knowledge About Perceptual Uncertainty Help an Agent in Automated Driving?

• Natalie Grabowsky (Technische Universität Berlin)

• Annika Mütze (Universität Osnabrück)

• Matthias Rottmann (Universität Osnabrück)

Design and Behavior of Sparse Mixture-of-Experts Layers in CNN-based Semantic Segmentation

• Svetlana Pavlitska (FZI Research Center for Information Technology)

• Haixi Fan (FZI Research Center for Information Technology)

• Konstantin Ditschuneit (Scenarium AI)

• J. Marius Zöllner (Karlsruhe Institute of Technology)

RetroMotion: Retrocausal Motion Forecasting Models are Instructable

• Royden Wagner (Karlsruhe Institute of Technology)

• Omer Sahin Tas (FZI Research Center for Information Technology)

• Felix Hauser (FZI Research Center for Information Technology)

• Marlon Steiner (Karlsruhe Institute of Technology)

• Dominik Strutz (Karlsruhe Institute of Technology)

• Abhishek Vivekanandan (Karlsruhe Institute of Technology)

• Jaime Villa (Universidad Carlos III de Madrid)

• Yinzhe Shen (Karlsruhe Institute of Technology)

• Carlos Fernandez (Karlsruhe Institute of Technology)

• Christoph Stiller (Karlsruhe Institute of Technology)

Knowledge-Guided Failure Prediction: Detecting When Object Detectors Miss Safety-Critical Objects

• Jakob Paul Zimmermann (Freie Universität Berlin)

• Gerrit Holzbach (Frauenhofer)

• David Jonathan Lerch (Karlsruhe Institute of Technology)

Rethinking Skip Connections in Diffusion Models for Replication Mitigation

• Chenghao Li (University of Southern California)

• Yuke Zhang (University of Southern California)

• Dake Chen (Meta)

• Jingqi Xu (University of Southern California)

• Peter Anthony Beerel (University of Southern California)

Formal Verification of Biomedical Vision Transformers: A Framework for Safety-Critical Healthcare Applications

• Prasanth (Birla Institute of Technology and Science)

Robust Detection of Directional Adversarial Attacks in Deep Neural Networks for Radiological Imaging

• Tristan Malatyński (AGH University of Science and Technology)

• Joanna Jaworek-Korjakowska (AGH University of Science and Technology)