HackerTrans
TopNewTrendsCommentsPastAskShowJobs

jmeaden

no profile record

Submissions

Why most AI coding benchmarks are misleading (COMPASS paper)

arxiv.org
13 points·by jmeaden·قبل 10 أشهر·2 comments

comments

jmeaden
·قبل 10 أشهر·discuss
Hi, I’m one of the authors. Happy to answer questions about the dataset (LLM coding performance compared to 390k+ human submissions), the scoring approach, or the methodology behind COMPASS. Feedback and critique are welcome.
jmeaden
·قبل 10 أشهر·discuss
agree
jmeaden
·قبل 10 أشهر·discuss
This "taste" concept has overlap with conscientiousness, a trait that companies look for when hiring. But, it is hard to develop.