ELLIS Institute Tübingen researchers are presenting 48 papers, including 2 orals, and co-organizing 3 workshops at NeurIPS 2026 in Sydney, Paris and Atlanta.
Researchers from the ELLIS Institute Tübingen are presenting 48 papers at NeurIPS 2026, one of the world’s leading conferences on machine learning and artificial intelligence.
This year the conference takes place across three locations, Sydney, Paris and Atlanta, and our groups are represented at all three. Our contributions include 2 oral presentations and 46 posters, on topics ranging from AI safety and LLM agents to tabular foundation models, optimization and causal inference. Members of our groups are also co-organizing 3 workshops.
The full list of contributions, sorted by location and session, is below.
Sydney
Orals (2)
| Title | Authors | Session | |
|---|---|---|---|
| Evaluating Neural Data Tokenizers: A Framework for Assessing Learned Representations of Spiking Activity | Federico D'Agostino, Alex Gilbert, Susanne Keller, Jaivardhan Kapoor, Nicolas Reategui, Vedang Lad, Matthias Kümmerer, Jai Bhagat, Emin Orhan, Hasan A Bedel, Nikos Karantzas, Fabian Sinz, Surya Ganguli, Jakob H Macke, Sophia Sanborn, Alexander Ecker, Katrin Franke, Matthias Bethge, Andreas Tolias, Konstantin Willeke | Orals 1D Computational Neuroscience & Brain Decoding Pyrmont Theatre |
Link |
| LittleLearner: Language Models Under Pedagogically-Controlled Knowledge Exposure | Fanfei Li, Jana Zeller, Manuel Prada-Corral, Thaddäus Wiedemer, Prasanna Mayilvahanan, Ryan Cotterell, Wieland Brendel | Orals 2A LLM Training, Scaling & Efficient Reasoning TikTok Entertainment Center |
Link |
Posters (32)
| Title | Authors | Session | |
|---|---|---|---|
| Causality can systematically address the monsters under the bench(marks) | Felix Leeb, Zhijing Jin, Bernhard Schölkopf | Sydney Poster Session 1 Hall 1-4 |
Link |
| Colosseum: Auditing Collusion in Cooperative Multi-Agent Systems | Mason Nakamura, Abhinav Kumar, Saswat Das, Sahar Abdelnabi, Saaduddin Mahmud, Ferdinando Fioretto, Shlomo Zilberstein, Eugene Bagdasarian | Sydney Poster Session 1 Hall 1-4 |
Link |
| Drifting Fields are not Conservative | Leonard T. Franz, Sebastian Hoffmann, Tim Weiland, Bernhard Schölkopf, Georg Martius | Sydney Poster Session 1 Hall 1-4 |
Link |
| Revisiting the Adam–SGD Gap Beyond Single Factors | Chenxiang Zhang, Rustem Islamov, Enea Monzio Compagnoni, Jun Pang, Aurelien Lucchi, Antonio Orvieto | Sydney Poster Session 1 Hall 1-4 |
Link |
| Strong Post-Training from Permissive, Reasoning-Dominant, Web-Scale Pretraining | Harsh Raj, Ali Elganzory, Marianna Nezhurina, Victor May, Van Khue Nguyen, David Salinas, Huu Nguyen, Jenia Jitsev | Sydney Poster Session 1 Hall 1-4 |
Link |
| Transferability for General Reasoning: An Automated Curriculum for Multi-Domain LLM RL | Yongjin Yang, Jiarui Liu, Yinghui He, Lechen Zhang, Bernhard Schölkopf, Zhijing Jin | Sydney Poster Session 1 Hall 1-4 |
Link |
| When is Warmstarting Effective for Scaling Language Models? | Neeratyoy Mallik, Maciej Janowski, Johannes Hog, Herilalaina Rakotoarison, Josif Grabocka, Frank Hutter, Aaron Klein | Sydney Poster Session 1 Hall 1-4 |
Link |
| FutureSim: Replaying World Events to Evaluate Adaptive Agents | Shashwat Goel, Nikhil Chandak, Arvindh Arun, Ameya Prabhu, Steffen Staab, Moritz Hardt, Maksym Andriushchenko, Jonas Geiping | Sydney Poster Session 2 Hall 1-4 |
Link |
| Improved State Mixing in Higher-order and Block Diagonal Linear Recurrent Networks | Igor Dubinin, Antonio Orvieto, Felix Effenberger | Sydney Poster Session 2 Hall 1-4 |
Link |
| Learning POMDP World Models from Observations with Language-Model Priors | Valentin Six, Frederik Panse, Mathis Fajeau, Lancelot Da Costa, Mridul Sharma, Alfonso Amayuelas, Tim Xiao, David Hyland, Philipp Hennig, Bernhard Schölkopf | Sydney Poster Session 2 Hall 1-4 |
Link |
| Stabilizing Extrapolation in Looped Transformers via Learned Stochastic Stopping | Hsun-Yu Kuo, El Mahdi Chayti, Patrik Reizinger, Wieland Brendel, Martin Jaggi | Sydney Poster Session 2 Hall 1-4 |
Link |
| Encoding RNA Topology into Synthetic Alignments for 3D Structure Prediction | Jörg Franke, Iris K Tennie, Michael Uhl, Ryan Koksal, Dominika Matus, Dominik Scheuer, Frank Hutter, Rolf Backofen, Frederic Runge | Sydney Poster Session 3 Hall 1-4 |
Link |
| Fixed-Point Reasoning: Stable and Adaptive Deep Looped Models | Sajad Movahedi, Shlomo Libo Feigin, Vera Milovanović, Alexander Theus, Thomas Hofmann, Valentina Boeva, T. Konstantin Rusch, Antonio Orvieto | Sydney Poster Session 3 Hall 1-4 |
Link |
| Instrumental Choices: Measuring the Propensity of LLM Agents to Pursue Instrumental Behaviors | Jonas Wiedermann-Möller, Leonard Dung, Maksym Andriushchenko | Sydney Poster Session 3 Hall 1-4 |
Link |
| Learning from the Self-future: On-policy Self-distillation for dLLMs | Yifu Luo, Zeyu Chen, Haoyu Wang, Xinhao Hu, Yuxuan Zhang, Zhizhou Sha, Shiwei Liu | Sydney Poster Session 3 Hall 1-4 |
Link |
| No More, No Less: Task Alignment in Terminal Agents | Sina Mavali, David Pape, Jonathan Evertz, Samira Abedini, Devansh Srivastav, Thorsten Eisenhofer, Sahar Abdelnabi, Lea Schönherr | Sydney Poster Session 3 Hall 1-4 |
Link |
| The Alien Space of Science: Sampling Coherent but Cognitively Unavailable Research Directions | Alejandro H. Artiles, Martin Weiss, Levin Brinkmann, Iyad Rahwan, Bernhard Schölkopf, Chris Pal, Hugo Larochelle, Anirudh Goyal, Nasim Rahaman | Sydney Poster Session 3 Hall 1-4 |
Link |
| Beyond IID: How General Are Tabular Foundation Models, Really? | Lennart Purucker, Andrej Tschalzev, Nick Erickson, David Holzmüller, Alan Arazi, Alexander Pfefferle, Mustafa Tajjar, Gioia Blayer, Gael Varoquaux, Frank Hutter | Sydney Poster Session 4 Hall 1-4 |
Link |
| DiffRatio: Training One-Step Diffusion Models Without Teacher Supervision | Wenlin Chen, Mingtian Zhang, Jiajun He, Zijing Ou, José Miguel Hernández-Lobato, Bernhard Schölkopf, David Barber | Sydney Poster Session 4 Hall 1-4 |
Link |
| Partition, Prompt, Aggregate: Statistical Self-Consistency in Language Models | Patrik Wolf, Thomas Kleine Buening, Andreas Krause, Celestine Mendler-Dünner | Sydney Poster Session 5 Hall 1-4 |
Link |
| Intrinsically Interpretable Attention via Sparse Post-Training | Florent Draye, Anson Lei, Hsiao-Ru Pan, Ingmar Posner, Bernhard Schölkopf | Sydney Poster Session 5 Hall 1-4 |
Link |
| LAION-BVD: A 10-Million-Hour Open Video Dataset for Multimodal Pre-training | Andreas Hochlehnert, Marianna Nezhurina, Mehdi Cherti, Andrej Radonjic, Thaddäus Wiedemer, Christoph Schuhmann, Romain Beaumont, Wieland Brendel, Bernhard Schölkopf, A. Sophia Koepke, Jenia Jitsev, Matthias Bethge | Sydney Poster Session 5 Hall 1-4 |
Link |
| AlphaQ: Calibration-Free Bit Allocation for Mixture-of-Experts Quantization | Wanqi Yang, Yuexiao Ma, Alexander Conzelmann, Xiawu Zheng, Michael Mahoney, T. Konstantin Rusch, Shiwei Liu | Sydney Poster Session 6 Hall 1-4 |
Link |
| Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs | Alexander Panfilov, Peter Romov, Igor Shilov, Yves-Alexandre de Montjoye, Jonas Geiping, Maksym Andriushchenko | Sydney Poster Session 6 Hall 1-4 |
Link |
| Deriving Hyperparameter Scaling Laws via Modern Optimization Theory | Egor Shulgin, Jörg Franke, Dimitri von Rütte, Tianyue Zhang, Niccolò Ajroldi, Korbinian Pöppel, Bernhard Schölkopf, Aaron Klein, Peter Richtarik, Antonio Orvieto | Sydney Poster Session 6 Hall 1-4 |
Link |
| KiteNorm: Variance Regularisation for Stable and Scalable Post-LN Transformers | Leon Aviv Trochelmann, Sajad Movahedi, Shiwei Liu, Antonio Orvieto | Sydney Poster Session 6 Hall 1-4 |
Link |
| PhyTS: A Benchmark for Scientific Time Series | Benedict Armstrong, Jeroen Audenaert, Hannah P Binney, Alice Cheng, Maarten De Vos, Luis F Domingues, Allison Eto, Mary Feliz, Joseph Formaggio, Jessica Fry, Philip Harris, Ilay Kamai, Erik Katsavounidis, Mykyta Kliapets, Konstantinos Kontras, Aobo Li, Paul Liang, Daniel Muthukrishna, Christina Reissel, Daniela Rus, T. Konstantin Rusch, Avi Shporer, Penny L Slocum, Talia E Weiss, Lindley Winslow, Kyungseop Yoon | Sydney Poster Session 6 Hall 1-4 |
Link |
| STRABLE: Benchmarking Tabular Machine Learning with Strings | Gioia Blayer, Myung Jun Kim, Félix Lefebvre, Lennart Purucker, Alan Arazi, Eilam Shapira, Roi Reichart, Frank Hutter, Marine Le Morvan, David Holzmüller, Gael Varoquaux | Sydney Poster Session 6 Hall 1-4 |
Link |
| Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks | David Schmotz, Luca Beurer-Kellner, Sahar Abdelnabi, Maksym Andriushchenko | Sydney Poster Session 6 Hall 1-4 |
Link |
| SuperSycophantic: Stress-Testing Frontier LLMs from Single- to Multi-Turn Sycophancy | Terry J. C. Zhang, Oscar S Yasunaga, Wenyuan Jiang, Jessica Bo, Florent Draye, Aydin Javadov, Arth Singh, Lucille Muir, VICTORIA OLDEMBURGO DE MELLO, Bernhard Schölkopf, Zhijing Jin | Sydney Poster Session 6 Hall 1-4 |
Link |
| Understanding the Surprising Generalization Properties of Tabular Foundation Models | Nour Shaheen, Junwei Ma, Alex Labach, Frank Hutter, Anthony L Caterini, Valentin Thomas | Sydney Poster Session 6 Hall 1-4 |
Link |
| Tio: Language Models with Parallel Streams of Thoughts, Inputs and Outputs | Guinan Su, Yanwu Yang, Xueyan Li, Jonas Geiping | TBC | Link |
Workshops (1)
| Title | Organizers | ||
|---|---|---|---|
| On-Device Intelligence: Foundation Models under Real-World Constraints | Niao He, Bingcong Li, Shiwei Liu, Hao Ma, Michael Muehlebach, Daniela Rus, Marko Zaric, Melanie Zeilinger, Zhuxiu Xu, Sergii Kryvoblotskyi | Workshop | Link |
Paris
Posters (13)
| Title | Authors | Session | |
|---|---|---|---|
| AM-Bench: A Unified Taxonomy and Evaluation Suite for Agentic Misalignment | Eric Zhang, Terry J. C. Zhang, Chijioke Ugwuanyi, Jerick Shi, Bernhard Schölkopf, Zhijing Jin | Paris Poster Session 1 Paris Poster Hall |
Link |
| TempoPFN: Synthetic Pre-training of Linear RNNs for Zero-shot Time Series Forecasting | Vladyslav Moroshan, Julien Siems, Arber Zela, Timur Carstensen, Frank Hutter | Paris Poster Session 1 Paris Poster Hall |
Link |
| AI Agents May Always Fall for Prompt Injections | Sahar Abdelnabi, Eugene Bagdasarian | Paris Poster Session 3 Paris Poster Hall |
Link |
| An Open-Source Training Dataset for Foundation Models for Black-box Optimization | Aaron Klein, Herilalaina Rakotoarison, Luca Thale-Bombien, David Salinas | Paris Poster Session 3 Paris Poster Hall |
Link |
| From Uncertain Judgments to Calibrated Rankings: Conformal Elo Estimation for LLM Evaluation | Bora Kargi, David Salinas | Paris Poster Session 3 Paris Poster Hall |
Link |
| PENEX: AdaBoost-Inspired Neural Network Regularization | Klaus-Rudolf Kladny, Bernhard Schölkopf, Michael Muehlebach | Paris Poster Session 4 Paris Poster Hall |
Link |
| Perception for Action in Latent World Models | Petr Ivashkov, Randall Balestriero, Bernhard Schölkopf | Paris Poster Session 4 Paris Poster Hall |
Link |
| GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory | Pepijn Cobben, Xuanqiang A Huang, Thao Pham, Isabel Dahlgren, Bernhard Schölkopf, Terry J. C. Zhang, Zhijing Jin | Paris Poster Session 5 Paris Poster Hall |
Link |
| Models That Know How Evaluations Are Designed Score Safer | Katharina Deckenbach, Haritz Puerto, Jonas Geiping, Sahar Abdelnabi | Paris Poster Session 5 Paris Poster Hall |
Link |
| MulTaBench: Benchmarking Multimodal Tabular Learning with Text and Image | Alan Arazi, Eilam Shapira, Shoham Grunblat, Mor Ventura, Elad Hoffer, Gioia Blayer, David Holzmüller, Lennart Purucker, Gael Varoquaux, Frank Hutter, Roi Reichart | Paris Poster Session 5 Paris Poster Hall |
Link |
| PVFormer: Proper Velocity Transformer for Stable and Scalable Hyperbolic Representation Learning | Xianglong Shi, Nicu Sebe, Bernhard Schölkopf, Ziheng Chen | Paris Poster Session 5 Paris Poster Hall |
Link |
| Towards Understanding Self-Pretraining for Sequence Classification | Omar Coser, Loredana Zollo, Paolo Soda, Antonio Orvieto | Paris Poster Session 5 Paris Poster Hall |
Link |
| Umbilic Multinomial Logistic Regression | Zihan Su, Nicu Sebe, Bernhard Schölkopf, Ziheng Chen | Paris Poster Session 5 Paris Poster Hall |
Link |
Workshops (2)
| Title | Organizers | ||
|---|---|---|---|
| Personalized, Aligned, Long-Term Memory for AI Systems (PALM) Workshop | Mario Fritz, Seong Joon Oh, Sahar Abdelnabi, Junxiao Shen, Hugo Lopes, Haritz Puerto, Ivaxi Sheth, Seokwon Jung | Workshop | Link |
| Sim2Science: ML with Imperfect Scientific Models | Georgia Channing, Noémi Éltetö, Richard Gao, Daniel Gedon, Magdalena Lederbauer; scientific advisors: Cecilia Clementi, Jakob Macke, Max Welling | Workshop | Link |
Atlanta
Posters (1)
| Title | Authors | Session | |
|---|---|---|---|
| Reaching a Consensus in Predictive Loops | Jiduan Wu, Rediet Abebe, Celestine Mendler-Dünner | Atlanta Poster Session 5 Hall C1 |
Link |