Publications (43)
ActMAD: Activation Matching to Align Distributions for Test-Time-Training
Muhammad Jehanzeb Mirza, Pol Jané Soneira, Wei Lin +3
Test-Time-Training (TTT) is an approach to cope with out-of-distribution (OOD) data by adapting a trained model to distribution shifts occurring at test-time. We propose to perform…
STSBench: A Spatio-temporal Scenario Benchmark for Multi-modal Large Language Models in Autonomous Driving
Christian Fruhwirth-Reisinger, DuÅ¡an MaliÄ, Wei Lin +3
We introduce STSBench, a scenario-based framework to benchmark the holistic understanding of vision-language models (VLMs) for autonomous driving. The framework automatically mines…
MAtch, eXpand and Improve: Unsupervised Finetuning for Zero-Shot Action Recognition with Language Knowledge
Wei Lin, Leonid Karlinsky, Nina Shvetsova +6
Large scale Vision-Language (VL) models have shown tremendous success in aligning representations between visual and text modalities. This enables remarkable progress in zero-shot…
GBlobs: Local LiDAR Geometry for Improved Sensor Placement Generalization
DuÅ¡an MaliÄ, Christian Fruhwirth-Reisinger, Alexander Prutsch +3
This technical report outlines the top-ranking solution for RoboSense 2025: Track 3, achieving state-of-the-art performance on 3D object detection under various sensor placements.…
Learn to Rank: Visual Attribution by Learning Importance Ranking
David Schinagl, Christian Fruhwirth-Reisinger, Alexander Prutsch +2
Interpreting the decisions of complex computer vision models is crucial to establish trust and accountability, especially in safety-critical domains. An established approach to int…
Video Test-Time Adaptation for Action Recognition
Wei Lin, Muhammad Jehanzeb Mirza, Mateusz Kozinski +3
Although action recognition systems can achieve top performance when evaluated on in-distribution test points, they are vulnerable to unanticipated distribution shifts in test data…