News for hackers. Stuff that matters.
Re:Translating Akkadian clay tablets with ChatGPT? by flenserboy on Monday May 15, 2023 @06:32PM (#35955408)
Not only double-check, but probably triple- or quadruple-. There's too much chance of hallucination to trust one or two instances; I think this could be done, but would require a number of disconnected instances doing the translations, with the resulting documents compared with each other to see a) which ones are more prone to making things up, b) if they hallucinate in similar ways, and c) where they agree, which would allow further study to find out what may have been in their training that allowed similar results to be output.
Re:Translating Akkadian clay tablets with ChatGPT? by sp332 on Monday May 15, 2023 @06:35PM (#35955435)
I would try round-tripping it at least. But I would have to know a little more Akkadian to tell if it's coming up with synonyms or completely different words.
Re:Translating Akkadian clay tablets with ChatGPT? by earthboundkid on Monday May 15, 2023 @11:28PM (#35957454)
Last week, a story in praise of ChatGPT ended with a haiku that has the wrong syllable count. Why check when it sounds good?
Re:Translating Akkadian clay tablets with ChatGPT? by gattilorenz on Tuesday May 16, 2023 @02:06AM (#35958222)
Do you have a link?
Re:Translating Akkadian clay tablets with ChatGPT? by earthboundkid on Tuesday May 16, 2023 @11:03PM (#35970793)
https://nadh.in/blog/this-time-it-feels-different/ No shade on the author. It just seems like an apt metaphor for how we get blinded by LLMs.
Re:Translating Akkadian clay tablets with ChatGPT? by sp332 on Tuesday May 16, 2023 @10:27AM (#35962364)
ChatGPT is the only LLM I have seen that can count syllables at all. It seems to be a gap in the way most of them are trained, since they only get text and never “hear” words pronounced.