International Conference on Mechanistic Interpretability and Neural Network Understanding

Important dates

1. 17 Dec

   Abstract / full paper deadline

   In 32 days

2. 13–14 Jan

   Conference

   Toronto, Ontario, Canada

All deadlines are as published by the organizer. Check the official website for the time zone.

About this conference

description 4: The conference will focus on the intricacies of interpretability in neural networks, targeted towards academics, researchers, and industry leaders interested in the systematic understanding of AI systems. Sessions will explore critical themes including algorithmic transparency, representation learning, and the societal implications of machine learning models. Participants are invited to submit various formats, including empirical research, theoretical papers, and posters that highlight innovative approaches to mechanistic interpretability. Through keynote speeches and discussion panels, attendees will engage with thought-provoking questions and contribute to shaping future research directions. The event encourages a multidisciplinary approach, bringing together expertise from computer science, cognitive psychology, and ethics. Opportunities for collaboration and networking will also be available, facilitating the exchange of ideas among attendees. Detailed registration information and submission deadlines will be provided on the official conference website.

Call for papers

Abstracts are due by 17 December 2026. Submission details are on the official website.