A detector I built was scoring 0.067 recall on temporal errors, meaning it caught about one in fifteen of the wrong dates it was supposed to find. The benchmark was the problem, and not in a way that showed up anywhere in the code.
A detector I built was scoring 0.067 recall on temporal errors, meaning it caught about one in fifteen of the wrong dates it was supposed to find.
The benchmark was the problem, and not in a way that showed up anywhere in the code.
I assumed the extraction was broken and went looking for the bug.
The page is ready to read now. The fuller skim-friendly version will appear here automatically.
A detector I built was scoring 0.067 recall on temporal errors, meaning it caught about one in fifteen of the wrong dates it was supposed to find. The benchmark was the problem, and not in a way that showed up anywhere in the code. I assumed the extraction was broken and went looking for the bug.
I assumed the extraction was broken and went looking for the bug. The contexts had been written in the wrong voice.
Open the app view to save this story, compare related coverage, and continue from the same source.