Exploring AGI Benchmarks

Name: Exploring AGI Benchmarks
Start: 2025-05-16T18:30:00+02:00
End: 2025-05-16T21:30:00+02:00
Location: Kapellenstraße 22

Hosted By

Dibya C. and 3 others

Details

AI/ML research is driven by benchmarks.

In this meetup, we’ll explore the benchmarks that matter most for AGI. The goal is to collectively map where current SOTA (state-of-the-art) AI stands, and where it still falls short.

Format: Round table discussion, not a lecture

We believe in learning from each other. That’s why our events aren’t about speakers and audiences - they’re open round table discussions. Everyone is equal, and everyone contributes.

To contribute effectively to the discussion, we ask all attendees to come minimally prepared (details below). That way, we all bring one benchmark to the table and leave with ten.

How to prepare (one hour suffices)

Before the event, please:

Pick an AGI benchmark that interests you. You can choose one from this curated list (scroll down a bit after clicking the link). The list covers lot of ground, spanning benchmarks on reasoning, continual learning, open endedness, robotics and agentic systems. If you have a favorite benchmark that's not on the list, that's also great.

Explore your benchmark with a focus on:

Why is it relevant to AGI?
What’s the dataset like? What’s an example task?
How do SOTA systems perform? What architectures or algorithms work well?
What is the gap to human-level performance?

You’ll be asked about your chosen benchmark when you RSVP. We'll expect you to share your knowledge about the benchmark during the discussion.

Who should attend?

This event is for anyone seriously interested in the frontiers of AI and AGI, whether you’re a researcher, engineer, student, or thoughtful enthusiast. Expert level knowledge is not required. Curiosity and minimal preparation are enough.

Plan for the evening

We meet in a large space rented by one of our co-organizer. It's ideal for deep conversations.

Here is the rough plan for the evening: