Tutorials

Morning: 9:00–12:30

Failure Modes in Agentic AI: From Reproducible Triggers and Diagnostics to Reliable Fixes

Organizers: Sha Li, Zihan Wang, Hejie Cui and Manling Li
Room: TBA

Description: This will be a 3.5 hour tutorial with slides presentation and real-time demonstrations.

Language Models as Windows into Human Language Processing: Neural Encoding, Decoding, and Brain Alignment with NLP

Organizers: Jingyuan Sun, Yunhao Zhang, Yang Cui and Yifan Wang
Room: TBA

Description: The tutors will deliver this mainly in the form of slides presentation, and will potentially include some short live demos of the presented systems.

Tokenization for Large Language Models

Organizers: Valentin Hofmann, Julie Kallini, Sachin Kumar, Jindrich Libovický, Tomasz Limisiewicz and Elizabeth Salesky
Room: TBA

Description: Direct lecturing with hands-on exercises in Python Notebooks.

Afternoon: 14:00–17:30

Autoformalization & Mathematical Reasoning

Organizers: Lan Zhang, Marco Valentino, Jordan Meadows and Andre Freitas
Room: TBA

Description: This tutorial consists of presentations, live demonstrations, and interactive discussions introducing autoformalization with large language models (LLMs). Participants will learn the fundamentals of formal mathematical reasoning, theorem provers (Lean4 and Isabelle), retrieval-augmented autoformalization, refinement using theorem-prover feedback, evaluation methods, and applications in scientific domains. The session includes Q&A and audience interaction throughout.

Causal Foundations for Interpretability: What Your Methods Actually Identify

Organizers: Shruti Joshi, Aaron Mueller, Patrik Reizinger and Dhanya Sridhar
Room: TBA

Description: Lectures, case studies, and demonstration of concepts with interactive demos.

Multilingual Multicultural LLMs

Organizers: Wei Xu, David Ifeoluwa Adelani, Vinodkumar Prabhakaran, Sara Hooker and Alan Ritter
Room: TBA

Description: Lectures, Q&A.