acalvino4
Joined 8 months ago
Karma
-3
Submissions
4
Member since
Mar 12
Recent submissions
Comment · 17 days ago
"suspect" meaning speculate, as far as I can tell. not really anything to argue with on that front. I never denied there was a mistake. Cheating implies intent, which no one here can prove, so seems useless to debate. Open source development is awesome in that it keeps people to a high bar, let's just celebrate that the process clearly worked here, and not muck about in speculation and judgement of others intentions that we have essentially no evidence for.
Comment · 17 days ago
submitting corrections is an acknowledgment in action
Comment · 17 days ago
Even before the ai age, people published inaccurate/misleading benchmarks all the time. Sometimes those were actual cheating, and other times an honest mistake. What matters is that the author acknowledges the issue and attempts to correct the mistake, which clearly seems to have happened here: https://github.com/ClickHouse/ClickBench/pull/2751
Most of this thread doesn't seem to acknowledge that, and feels more like anti-ai psychosis using this incident as an anecdote to justify the pre-existing hatred of ai-generated projects.