Choose timezone
Your profile timezone:
Natural Language Processing
[lamarr-nlp] Guest Talk by Prof. Lukas Galke Poech from University of Southern Denmark | Cooperation, Culture, and Coordination in Neural Ecosystems
→
Europe/Berlin
2.122 (Friedrich-Hirzebruch-Allee 6, Bonn)
2.122
Friedrich-Hirzebruch-Allee 6, Bonn
Description
As part of the Lamarr NLP Colloquium, we have the pleasure to host Prof. Lukas Galke Poech from the Southern University of Denmark. Lukas will give a talk on Cooperation, Culture, and Coordination in Neural Ecosystems.
Abstract:
We provide language models with increasing levels of autonomy, let them use tools deliberately, and have them interact with other agents and with humans at scale. How do we make sure that such neural ecosystems remain cooperative, culturally aligned, and robust against misuse? In this talk, I present three recent results. Cooperation: We show that repeated interactions and inter-group competition together shape cooperative propensities of language model agents. Culture: We localize culture-specific neurons in multilingual language models and show that they can be modulated independently of language. Coordination: We introduce a benchmark for guarded query routing that tests robustness against out-of-distribution and unsafe inputs. These perspectives show how research on interpretability and resilience can advance AI safety.
Bio:
Lukas Galke Poech is an Assistant Professor in the Department of Mathematics and Computer Science at the University of Southern Denmark. His research focuses on advancing AI safety through interpretability. He is the principal investigator of the MIST Project on Scalable Mechanistic Interpretability for Safe and Trustworthy Language Model Agents, supported by the Novo Nordisk Foundation. Lukas received his PhD in representation learning from Kiel University, Germany. He then joined the Max Planck Institute for Psycholinguistics as a postdoc, where he drew inspiration from research on how language works in the human brain to understand learning, behavior, and cognition in language models. His work has been published in venues including Neural Networks and Nature Communications.
Lukas Galke Poech is an Assistant Professor in the Department of Mathematics and Computer Science at the University of Southern Denmark. His research focuses on advancing AI safety through interpretability. He is the principal investigator of the MIST Project on Scalable Mechanistic Interpretability for Safe and Trustworthy Language Model Agents, supported by the Novo Nordisk Foundation. Lukas received his PhD in representation learning from Kiel University, Germany. He then joined the Max Planck Institute for Psycholinguistics as a postdoc, where he drew inspiration from research on how language works in the human brain to understand learning, behavior, and cognition in language models. His work has been published in venues including Neural Networks and Nature Communications.
Looking forward to your participation.
Date: Wednesday, Feb 11, 2026
Time: 10:15 - 11:15 pm (CET).
Location: Friedrich-Hirzebruch-Allee 6, 53115 Bonn, Germany
Room: 2.122 + Zoom
Zoom: https://uni-bonn.zoom-x.de/j/63819604806?pwd=64PSGa9HyTym9j1bjy6jhcJF3eHebi.1
Date: Wednesday, Feb 11, 2026
Time: 10:15 - 11:15 pm (CET).
Location: Friedrich-Hirzebruch-Allee 6, 53115 Bonn, Germany
Room: 2.122 + Zoom
Zoom: https://uni-bonn.zoom-x.de/j/63819604806?pwd=64PSGa9HyTym9j1bjy6jhcJF3eHebi.1