Top Tweets for #MLinPL

Last day of the Conference! After two packed days of talks and conversations, today we dive into some of the most exciting questions in machine learning. Here’s what’s happening today at the Copernicus Science Centre: 🕗 9:30 – 11:00 | Morning Sessions Main Hall • Out of Context Generalization in LLMs — Anna Sztyber-Betley (Warsaw University of Technology) • Emergent Misalignment: Narrow Finetuning Can Produce Broadly Misaligned LLMs — Jan Betley (TruthfulAI) • Eliciting Secret Knowledge from Language Models — Bartosz Cywiński (Warsaw University of Technology; IDEAS Research Institute) Hall A • Conformal Object Detection by Sequential Risk Control — Léo Andéol (Institute of Mathematics of Toulouse / SNCF) • A Generalized Approach to Label Shift: The Conditional Probability Shift Model — Paweł Teisseyre, PhD (Warsaw University of Technology) • Joint Empirical Risk Minimization for Instance-Dependent Positive-Unlabeled Data — Jan Mielniczuk (Polish Academy of Sciences) Hall B • Limits in Causal Discovery and the Path Forward — Mateusz Gajewski& Mateusz Olko • Inference on Local Variable Importance Measures for Heterogeneous Treatment Effects — Paweł Morzywołek (University of Washington) • Learning Personalized Treatment Decisions in Precision Medicine — Michael Vollenweider (ETH Zurich) 11:15 – 12:15 | Midday Sessions Main Hall: Panel Discussion – AI in Security Hall A: Open Foundation Models: Scaling Laws and Generalisation — Jenia Jitsev (LAION / Juelich Supercomputer Center / ELLIS) Hall B: From Pixels to Nucleotides — Alexey Dosovitskiy (Inceptive) 12:15 – 13:45 | Poster Session 2 +🍴☕ Continue the conversations over lunch and explore the second round of posters. 14:45 – 15:45 | Sponsor Talks Hall A: Measuring What Matters: The Challenges of Evaluating and Optimizing LLMs — Patrycja Cieplicka (Tooploox) Hall B: All Things Metal 🤘 — Maciek Draguła & Artem Yerofieiev (Tenstorrent) 16:00 – 17:00 | Closing Invited Talks What’s the Next Wave of Disruption in Science and Engineering? — Johannes Brandstetter (Johannes Kepler University / Emmi AI) Equity in Healthcare AI: The Interplay of Data, Biases, and Explanation — Bartlomiej Papiez (University of Oxford) Modular Learning for Improving AI Assistants — Herke van Hoof (University of Amsterdam) 17:10 – 17:40 | Closing Remarks And we’re not done yet! Tomorrow, you can still join us for Tutorials https://t.co/4DsVz1FQrm