MLR@Penn Workshop on Mentorship in Foundation Models Research

This workshop is organized by MLR@Penn, a student organization dedicated to machine learning research at the University of Pennsylvania, and is co-located with the Conference on Language Modeling (COLM) 2024. The goal of this workshop is to provide a forum to discuss recent advances in foundation models, where junior researchers (undergraduates, Master's students, PhD candidates, early career industry) can seek mentorship and learn from prominent contributors to the field. 

While our workshop (in particular, the mentorship tables session) addresses a broad audience across the various language modeling topics listed in the COLM 2024 Call for Papers, the main focus of our panel and invited talk will be on a smaller set of topics which presently hold widespread interest, including:

• Agents / Agentic Frameworks (Tools, Grounding, Multimodality)

• Benchmarking (Development, Reliability)

• Model Attribution and Interpretability (Emergent Abilities, Scaling Laws, (Mechanistic) Interventions, Probing)

We aim to promote meaningful dialogue and foster inclusivity by enabling researchers from diverse backgrounds to attend our program, who may otherwise not be able to receive mentorship or join networking opportunities to propel their careers.