MeshUp #13: Evaluating and Comparing AI Models
This week, Changkyu Choi will present on the topic How AI conferences evaluate and compare models.
Abstract
This talk provides an overview of the current landscape of how AI models are evaluated and compared at major AI conferences. We analyze the historical growth of paper submissions and the institutional shift toward specialized tracks, such as the datasets and benchmarks track, to understand how the community is standardizing model performance assessment. Finally, the talk discusses opportunities for interdisciplinary research groups, such as MishMash, to help expand these evaluation frameworks to include the intersection of AI and creativity.
Bio
Changkyu Choi is an AI researcher focused on understanding the learning and reasoning mechanisms of neural networks. He is an associate professor at the University of Inland Norway and holds a part-time position at the University of Oslo. He has been actively contributing to major AI conferences since 2024.
Access
MishMash MeetUps are short, informal meetings in the consortium where both early career and established researchers present ongoing projects. The events are open for everyone, but, for security reasons, Zoom links are only provided to people that are affiliated with a MishMash Work Package. If not, please ask for access. The MeshUps are not recorded since we want these short events to also serve as a social meeting point for the centre.
Subscribe to all events (webcal)
Download calendar (ICS)
If the webcal link doesn't open in your calendar (e.g. Outlook), use "Download calendar (ICS)" and open the file in your calendar app.