What about a 4x accept on initial reviews?

... as it requires a deeper understanding of the physical world, embodiment, and real-world interactions beyond textual abstraction. Her perspective on how these two forms of intelligence complement each other in AI advancements was truly fascinating!

I found her insights on the comparison between spatial intelligence and language intelligence especially thought-provoking. While language intelligence has made remarkable progress with large language models, spatial intelligence presents unique challenges, as ...

What about CVPR2030?

Just to clarify, sometime a disrespectful email from the AC to the reviewer can be considered as a form on punishment! 😉

Also to further clarify, I have AC/SAC roles in conferences these days, but have seen such emails sent to my colleagues!

I agree with your note. I am just trying to put myself in the shoes of someone who is seeing this unfolding, from outside!!

Indeed. At this scale, we are in dire need of better tooling.

But, Christian, I want to also argue: if this is all volentary work, how do we justify punishing late/no-show reviewers? So are we saying late-reviewers are bad, but it is ok for PCs/ACs/SACs to be late because this volentary?

And if you professionalize this, there is the risk of incentivising revenue and money-making, and losing the entire community, like how some open access and predatory journals are acting now! (outsourcing is even worse, because they don't understand the value of the science being published!)

This was the team that created recommendations on how to properly evaluate AI models to get software as a medical device approvals.