ICML 2026, the 43rd International Conference on Machine Learning, will be held in Seoul, South Korea, from July 6 to 11, 2026.
This major global event brings together leading researchers, practitioners, and industry experts to advance the field of machine learning. As one of the world's premier AI conferences, ICML serves as a platform for presenting groundbreaking research and fostering collaboration across the international machine learning community.
ICML is renowned for showcasing cutting-edge work across all areas of machine learning, including deep learning, reinforcement learning, optimization, generative models, and AI for scientific discovery. The conference highlights both theoretical advances and real-world applications in fields such as robotics, healthcare, computer vision, speech recognition, and computational biology.
At ICML 2026, the ELLIS Institute Tübingen will present its latest research contributions, demonstrating the institute’s commitment to advancing the frontiers of machine learning and artificial intelligence.
Contributions
| Tuesday, Jul 7 | |
| Position: Safety Must Precede the Deployment of Open-Ended AI Agents | Ivaxi Sheth, Jan Wehner, Sahar Abdelnabi, Ruta Binkyte, Mario Fritz |
| Computational Arbitrage in AI Model Markets | Ricardo Dominguez-Olmedo, Bernhard Schölkopf, Moritz Hardt |
| Training with Honeypots: Reshaping How LLMs Fail | Samuel Simko, Punya Pandey, Zhijing Jin, Bernhard Schölkopf |
| Generation is Required for Data-Efficient Perception | Jack Brady, Bernhard Schölkopf, Thomas Kipf, Simon Buchholz, Wieland Brendel |
| When Does Sparsity Mitigate the Curse of Depth in LLMs | Dilxat Muhtar, Xinyuan Song, Sebastian Pokutta, Max Zimmer, Nico Pelleriti, Thomas Hofmann, Shiwei Liu |
| Search or Accelerate: Confidence-Switched Position Beam Search for Diffusion Language Models | Mingyu Cao, Alvaro Correia, Christos Louizos, Shiwei Liu, Lu Yin |
| Wednesday, Jul 8 | |
| Helpful to a Fault: Measuring Illicit Assistance in Multi-Turn, Multilingual LLM Agents | Nivya Talokar, Ayush Kumar Tarun, Murari Mandal, Maksym Andriushchenko, Antoine Bosselut |
| α-PFN: Fast Entropy Search via In-Context Learning | Herilalaina Rakotoarison, Steven Adriaensen, Tom Viering, Samuel Gabriel Müller, Carl Hvarfner, Frank Hutter, Eytan Bakshy |
| Use What You Know: Causal Foundation Models with Partial Graphs | Arik Reuter, Anish Dhir, Cristiana Diaconu, Jake Robertson, Ole Ossen, Frank Hutter, Adrian Weller, Mark van der Wilk, Bernhard Schölkopf |
| On the Interaction of Batch Noise, Adaptivity, and Compression, under (L₀,L₁)-Smoothness: An SDE Approach | Enea Monzio Compagnoni, Rustem Islamov, Frank Proske, Aurelien Lucchi, Antonio Orvieto, Eduard Gorbunov |
| Position: Safe Models Do Not Guarantee Safe Societies: The Case for Sociopolitical Risk | David Guzman Piedrahita, Dave Banerjee, Changling Li, Terry Zhang, Kevin Blin, Samuel Simko, Punya Pandey, Irene Strauss, Rada Mihalcea, Bernhard Schölkopf, Zhijing Jin |
| Position: LLM for Physics Research Requires Domain-Specialized Training and Tooling | Sirui Lu, Zhijing Jin, Terry Zhang, Pavel Kos, Juan Cirac, Bernhard Schölkopf |
| Position: Trustworthy AI Suffers from Invariance Conflicts and Causality is The Solution | Ruta Binkyte, Ivaxi Sheth, Zhijing Jin, Mohammad Havaei, Bernhard Schölkopf, Mario Fritz |
| Position: Causality is Key for Interpretability Claims to Generalise | Shruti Joshi, Aaron Mueller, David Klindt, Wieland Brendel, Dhanya Sridhar, Patrik Reizinger |
| Training AI Co-Scientists Using Rubric Rewards | Shashwat Goel, Rishi Hazra, Dulhan Jayalath, Timon Willi, Parag Jain, Shen, Ilias Leontiadis, Francesco Barbieri, Yoram Bachrach, Jonas Geiping, Chenxi Whitehouse |
| Rewiring Experts on the Fly: Continuous Rerouting for Better Online Adaptation in Mixture-of-Expert Models | Guinan Su, Yanwu Yang, Li Shen, Lu Yin, Shiwei Liu, Jonas Geiping |
| Efficient Parallel Samplers for Recurrent-Depth Models | Jonas Geiping, Xinyu Yang, Guinan Su |
| GradientStabilizer: Fix the Norm, Not the Gradient | Tianjin Huang, Zhangyang "Atlas" Wang, Haotian Hu, Zhenyu Zhang, Gaojie Jin, Xiang Li, Li Shen, Jiaxing Shang, Tianlong Chen, Ke Li, Lu Liu, Qingsong Wen, Shiwei Liu |
| Motion-Aware Caching for Efficient Autoregressive Video Generation | Jing Xu, Yuexiao Ma, Songwei Liu, Xuzhe Zheng, Shiwei Liu, Chenqian Yan, Xiawu Zheng, Rongrong Ji, Fei Chao, Wang |
| One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs | Di He, Songjun Tu, Keyu Wang, Lu Yin, Shiwei Liu |
| Thursday, Jul 9 | |
| PostTrainBench: Can LLM Agents Automate LLM Post-Training? | Ben Rank, Hardik Bhatnagar, Ameya Pandurang Prabhu, Shira Eisenberg, Karina Nguyen, Matthias Bethge, Maksym Andriushchenko |
| Neural Low-Discrepancy Sequences | Michael Van Huffel, Nathan Kirk, Makram Chahine, Daniela Rus, T. Konstantin Rusch |
| CauSciBench: Evaluating LLM Causal Inference for Scientific Research | Sawal Acharya, Terry Zhang, Andrew Kim, Anahita Haghighat, Xianlin Sun, Pepijn Cobben, Rahul Shrestha, Maximilian Mordig, Jacob Emmerson, Furkan Danisman, Yuen Chen, Clijo Jose, Andrei Muresanu, Justin Cui, Jiarui Liu, Yahang Qi, Punya Pandey, Yinya Huang, Bernhard Schölkopf, Zhijing Jin |
| MentisOculi: Revealing the Limits of Reasoning with Mental Imagery | Jana Zeller, Thaddäus Wiedemer, Fanfei Li, Thomas Klein, Prasanna Mayilvahanan, Matthias Bethge, Felix Wichmann, Ryan Cotterell, Wieland Brendel |
| Curating the Future: A Scalable Recipe for Training Open-Ended Forecasters | Nikhil Chandak, Shashwat Goel, Ameya Pandurang Prabhu, Moritz Hardt, Jonas Geiping |
| Date Pending | |
| Riemannian Networks over Full-Rank Correlation Matrices | Ziheng Chen, Xiaojun Wu, Bernhard Schölkopf, Nicu Sebe |
Workshops
| Thursday, Jul 9 | |
| Workshop on Weight-Space Symmetries: from Foundations to Practical Applications | Yani Ioannou, Boris Knyazev, Ekaterina Lobacheva, Adnan Mohammed, Antonio Orvieto, Alexander Theus |
| Friday, Jul 10 | |
| New Frontiers in Game-Theoretic Learning | Nicolò Cesa-Bianchi, Tatjana Chavdarova, Michael Jordan, Celestine Mendler-Dünner, Rene Vidal, Emmanouil-Vasileios Vlatakis-Gkaragkounis |
| Philosophy Meets Machine Learning: What Counts as Trustworthy? | Junhyung Park, Fanny Yang, Bernhard Schölkopf, Konstantin Genin, Thomas Icard, Vincent Fortuin, Jaesik Choi |