mint

joined 1 year ago
[–] [email protected] 4 points 11 months ago (1 children)

@nmtake @fediverse >Akkoma is currently leading in this regard with their MRF feature, which allows administrators to write arbitrary polices to rewrite or delete messages
Yeah, bub, about that: https://akkoma.dev/AkkomaGang/akkoma/src/branch/develop/lib/pleroma/web/activity_pub/mrf.ex#L67
You're not allowed to reject or mangle activities of specific types in your own policies, and said change was introduced for no reason other than spite. Pleroma, which Akkoma rips off, has no such restrictions.

[–] [email protected] 3 points 1 year ago (1 children)

@pomstan @0x2d Well, what else you'd expect when gene splicing redditors and mastodon users?

[–] [email protected] 3 points 1 year ago

@p @ceo_of_monoeye_dating @Nerd02 @bmygsbvur @db0 Yes, but it should be able to count two concepts together even if there were no overlap between the two in training data.

[–] [email protected] 4 points 1 year ago (12 children)

@p @ceo_of_monoeye_dating @Nerd02 @bmygsbvur @db0 Yeah, it's using local CLIP model, something I've suggested both to gr*f and jakparty.soy admin. The problem is that it requires a lot of clock cycles, preferably on GPU, so it isn't something people with $5 VPSes can afford. Not fully sure about effectiveness, either, malicious actors can keep scrambling the image so that it passes the filter yet is still recognizable by human brain.

[–] [email protected] 1 points 1 year ago (1 children)

@bmygsbvur Pleroma is exactly the same and no one cared in six years.