Call for Participation (26/6/2026 online or in person): Workshop on Mechanistic Interpretability and Neuro-symbolic Models by-design

[fabiomassi...@gmail.com]

Site: MeMo Workshop https://sites.google.com/view/memo-workshop/

Date: 26/6/2026

Location: online or in person (Rome, Italy)

Registration link (free but necessary): Register here

 

Are you surprised about the ability of LLMs to perform complex reasoning tasks? Do you have a particular keenness for cracking the code of their inner workings? If the answer is yes, that's the case you are working in mechanistic interpretability or in neuro-symbolic reasoning of LLMs. Then, this is the workshop for you!  Observing properties and understanding the inner workings are two sides of the same coin.
Another way is possible (e.g., Position Paper: MeMo: Towards Language Models with Associative Memory Mechanisms - ACL Anthology)!

In the workshop, authors will discuss focused contributions on:

• Mechanistic Interpretability by-design
• Neuro-symbolic approaches by-design 

 

Apologies for multiple posting, but we really hope to see you there.