We are delighted to announce that four research papers from L3S researchers have been accepted at NeurIPS 2025 – The Thirty-Ninth Annual Conference on Neural Information Processing Systems (neurips.cc), one of the world’s leading conferences in artificial intelligence and machine learning.
The accepted works showcase the breadth of AI research at L3S, spanning emotion recognition, video understanding, explainable and interpretable AI, and neural architecture optimization. Together, they highlight our ongoing efforts to advance trustworthy, efficient, and human-centered AI.
Accepted papers:
- EmoNet-Face: An Expert-Annotated Benchmark for Synthetic Emotion Recognition
(Christoph Schuhmann, Robert Kaczmarczyk, Gollam Rabby, Felix Friedrich, Maurice Kraus, Krishna Kalyan, Kourosh Nadi, Huu Nguyen, Kristian Kersting, Sören Auer) - Enhancing Temporal Understanding in Video-LLMs through Stacked Temporal Attention in Vision Encoders
(Ali Rasekh, Erfan Bagheri Soula, Omid Daliran, Simon Gottschalk, Mohsen Fayyaz) - CHiQPM: Calibrated Hierarchical Interpretable Image Classification
(Thomas Norrenbrock, Timo Kaiser, Sovan Biswas, Neslihan Kose, Ramesh Manuvinakurike, Bodo Rosenhahn) - Neural Attention Search
(Difan Deng, Marius Lindauer)
Congratulations to all authors on this outstanding achievement and their contribution to advancing cutting-edge AI research at L3S!
