News for hackers. Stuff that matters.
Re:Show HN: I extracted the safety filters from Apple Intell... by heavyset_go on Sunday July 06, 2025 @11:12PM (#44486410)
As long as we have to rely on pre trained networks and curated training sets, normal people will not be able to surpass this issue.
Re:Show HN: I extracted the safety filters from Apple Intell... by ghxst on Monday July 07, 2025 @03:39AM (#44487673)
If the training data was "censored" by leaving out certain information, is there any practical way to inject that missing data after the model has already been trained?
Re:Show HN: I extracted the safety filters from Apple Intell... by heavyset_go on Monday July 07, 2025 @03:57AM (#44487774)
You can fine tune a model with new information, but it is not the same thing as training it from scratch, and can only get you so far.

You might even be able to poison a model against being fine-tuned on certain information, but that's just a conjecture.

Re:Show HN: I extracted the safety filters from Apple Intell... by calaphos on Monday July 07, 2025 @05:26AM (#44488372)
If it's just filtered out in the training sets, adding the information as context should work out fine - after all this is exactly how o3, Gemini 2.5 and co deal with information that is newer than their training data cutoff.
Re:Show HN: I extracted the safety filters from Apple Intell... by selfhoster11 on Monday July 07, 2025 @05:30AM (#44488395)
Yes, RAG is one way to do that.