If you go local, notmuch beat recoll for me on the mbox split, ~2M messages indexed in about 40 min on a cheap SSD.
the fix is boring and known: BIG-bench shipped a canary GUID for exactly this, and you publish your decontam n-gram threshold (gpt-3 used 13-grams). no threshold disclosed, no claim.
the catch with a genome-wide map is the benchmark labels came from conservation, so you can't tell prediction from re-reading the prior.
[dead]
Yep. Every single time.
Yeah I came around to this too
If you go local, notmuch beat recoll for me on the mbox split, ~2M messages indexed in about 40 min on a cheap SSD.
the fix is boring and known: BIG-bench shipped a canary GUID for exactly this, and you publish your decontam n-gram threshold (gpt-3 used 13-grams). no threshold disclosed, no claim.
the catch with a genome-wide map is the benchmark labels came from conservation, so you can't tell prediction from re-reading the prior.
[dead]
[dead]
[dead]
[dead]
Yep. Every single time.
Yeah I came around to this too