WHEN: June 30, 2026, 10:00-13:00
WHERE: Building C7.4 at Saarland University Campus, Conference Room 1.17
This event is part of the ELLIS Sites ICML Fest 2026: Celebrate, Connect, Collaborate, associated with our ELLIS unit.
| Title | Authors | Presenter(s) |
|---|---|---|
| DAVE: Distribution-aware Attribution via ViT Gradient Decomposition | Adam Wróbel, Siddhartha Gairola, Jacek Tabor, Bernt Schiele, Bartosz Zieliński, Dawid Rymarczyk | Siddhartha Gairola |
| Temporal Context Reinstatement Drives Episodic-Like Order Memory in Long-Context Language Models | Mathis Pink, Vy A. Vo, Qinyuan Wu, Jianing Mu, Javier S. Turek, Uri Hasson, Kenneth A. Norman, Sebastian Michelmann, Alexander Huth, Mariya Toneva | Mathis Pink |
| On the Ability of Transformers to Verify Plans | Yash Sarrof, Yupei Du, Katharina Stein, Alexander Koller, Sylvie Thiébaux, Michael Hahn | Yash Sarrof |
| Fixed Aggregation Features Can Rival GNNs | Celia Rubio-Madrigal, Rebekka Burkholz | Celia Rubio-Madrigal |
| Robustness of Mixtures of Experts to Feature Noise | Dong Sun, Rahul Nittala, Rebekka Burkholz | Rahul Nittala |
| Off-Policy Evaluation with Strategic Agents via Local Disclosure | Kiet Q. H. Vo, Abbavaram Gowtham Reddy, Julian Rodemann, Siu Lun Chau, Krikamol Muandet | Kiet Q. H. Vo |
| SparseOpt: Addressing Normalization-induced Gradient Skew in Sparse Training | Mohammed Adnan, Rohan Jain, Tom Jacobs, Ekansh Sharma, Rahul G. Krishnan, Rebekka Burkholz, Yani Ioannou | Tom Jacobs |
| Position: Safety Must Precede the Deployment of Open-Ended AI Agents | Ivaxi Sheth, Jan Wehner, Sahar Abdelnabi, Ruta Binkyte, Mario Fritz | Ivaxi Sheth |
| MaMa: A Game-Theoretic Approach for Designing Safe Agentic Systems | Jonathan Nöther, Adish Singla, Goran Radanovic | Jonathan Nöther |
| How Few-Shot Examples Add Up: A Causal Decomposition of Function Vectors in In-Context Learning | Entang Wang, Yiwei Wang, Aleksandra Bakalova, Michael Hahn | Entang Wang |
| ExPLAIND: Unifying Model, Data, and Training Attribution to Study Model Behavior | Florian Eichin, Yupei Du, Philipp Mondorf, Maria Matveev, Barbara Plank, Michael A. Hedderich | Yupei Du |
| Robust In-Context Reinforcement Learning Under Reward Poisoning Attacks | Paulius Sasnauskas, Yiğit Yalın, Goran Radanović | Yiğit Yalın |
| Certified Circuits: Stability Guarantees for Mechanistic Circuits | Alaa Anani, Tobias Lorenz, Bernt Schiele, Mario Fritz, Jonas Fischer | Alaa Anani & Tobias Lorenz |
| Fast Reconstruction of Mixture of Bernoulli Distributions | Sanyam Agarwal, Pranjal Dutta, Markus Bläser | Sanyam Agarwal |
| Discovering Interpretable Algorithms by Decompiling Transformers to RASP | Xinting Huang, Aleksandra Bakalova, Satwik Bhattamishra, William Merrill, Michael Hahn | Michael Hahn |
| Title | Authors | Presenter(s) | Workshop |
|---|---|---|---|
| Linear GCNs Need Better Bias, Not More Expressive Power | Celia Rubio-Madrigal, Rebekka Burkholz | Celia Rubio-Madrigal | Workshop on Graph Foundation Models: A New Era for Graph Machine Learning |
| Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer? | Franka Bause, Jonas Niederle, Martin Pawelczyk, Rebekka Burkholz | Jonas Niederle | 2nd Workshop on Compositional Learning: Safety, Interpretability, and Agents |