With a 10% acceptance (which can happen especially if you exclude MR), rate 2/20 is expected and 0/20 happens more often than 1/10. With 15%, happens 3% of the time, so you will see it with a large PC. Given non-randomizes paper selection, it could be even more frequent.
-
-
-
Let's stick with 15-20% please. But I see your point. Then how likely will this happen 3, 4, 5 times in a row?
- Još 3 druga odgovora
Novi razgovor -
-
-
HotCRP does allow review feedback, at least from other reviewers.
-
HotCRP allows many things, the question is if these features are used (spoiler alert: they are not) :/
Kraj razgovora
Novi razgovor -
-
-
I think reviewers are much better at ranking the papers in their stacks than they are at scoring them individually. This would also be a form of normalization. I believe John Douceur experimented with this about 15yrs ago..?
-
Yeah, I had the same thought. After submitting all their reviews, PC members could submit a relative ranking of their reviewed papers
- Još 1 odgovor
Novi razgovor -
-
-
This is amazing. Why don't now chairs use it?
- Još 5 drugih odgovora
-
-
-
Normalizing reviewer scores sounds really good!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
The normalization of scores is one of the primary tasks for the PC chairs. It is imperative to factor in the reviews from the PC members who are consistently negative or positive. HotCRP provides a nice interface to rank the reviewers.
-
Tweet je nedostupan.
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.
group. Interests: system/software security, mitigation, sanitization, fuzzing, exploitation