Users are enthusiastic about a proposed cybersecurity benchmark because they see it as a killer idea that could use evals based on priced gig work.
Based on 1 visible X reactions from 2 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@xeophon Would be a killer bench. Aren't there some evals based on priced gig work at least? It doesn't need to be frontier difficulty, the frontier is the margin.
@xlr8harder I saw this at least once for CVE/cybersec benches For economic benches like GDPVal it’s hard to say as it’s hard to put a number on the thing itself. I guess RLI could do this
@xlr8harder thats RLI, yes.
@xlr8harder I saw this at least once for CVE/cybersec benches For economic benches like GDPVal it’s hard to say as it’s hard to put a number on the thing itself. I guess RLI could do this
Users are enthusiastic about a proposed cybersecurity benchmark because they see it as a killer idea that could use evals based on priced gig work.
Based on 1 visible X reactions from 2 accounts; directional sample.
Ask a question below.
Published answers will appear here.