If you don't have quantifiers in your statement (e.g. Sonnet > GPT-4o), I am assuming "for all tasks Sonnet > GPT-4o"). In my limited anecdotal experience, we need quantifiers in front "for task X. Sonnet > GPT-4o). In my opinion, these statements without quantifiers is sloppy

This is an amazing initiative. We need more of these! Thanks Perri.

Thank for the shoutout UW-Madison Computer Sciences and WPR

Our work on Pharmacovigilance using multi-agent LLM powered system. This was joint work with Duke medical and especially proud of the students for pulling this off. Blog post by prasad chalasani about the paper. Paper is going to appear in MLHC 2024. Paper: arxiv.org/abs/2408.01869

Impressive!

Look at the statistics for the ML-security papers! This is exactly why we need to grow SaTML Conference SaTML Conference for the win!

Great group! Especially Amrita Roy Chowdhury :-) Congrats to all.

Congrats Rajeev Alur Amazing news!

Excellent opportunity, and Konrad Rieck 🌈 is amazing!

Congrats Robust Intelligence and Cisco

Excellent opportunity! I spent one year at IMDEA Software and it was amazing!

Amazing Jonathan Ullman Happy to have you in the midwest. Congrats to you and Computer Science and Engineering at Michigan

Exciting competitions at SaTML Conference All of them look super interesting...

Excellent win for Computer Science and Engineering at Michigan

Really great venue! Nominate someone to be on the PC. My first paper in CAV was on mu-calculus. link.springer.com/chapter/10.100…

Excellent! Should be an interesting challenge.

Congrats to Jeannette! She has been great mentor for me since my grad school days at CMU Computer Science Department ColumbiaCompSci

I have used it and it is quite impressive.

Deadline for SaTML Conference has been extended to Sep 27. You are welcome:-)