undefined | Better HN

0 pointsKuinox2mo ago0 comments

Yet translation was the main application for applied language machine learning.

0 comments

When would've thought that to solve natural language translation, one would first need to solve... natural language.

All those arguments about agents and hallucinations kind of distracted people from noticing we've accodentally built a universal translator.

KuinoxOP2mo ago

It's been at least 10 years that google translate had hallucinations. Some translation simply change depending of a ponctuation mark. But peoples complain only now that they heard about AI.

Of course it's not perfect, but I agree that we didn't had a machine translation as good before.

krige2mo ago

As someone both exposed to this new wave of LLM style translation in various media, and someone who has background in translation, no we didn't.

TeMPOraL2mo ago

Could you please explain briefly then why my statement is wrong? What are the fundamental challenges not addressed by LLMs today? Do you think the whole approach has insurmountable roadblocks ahead, or is it more of a matter of refinement?

krige2mo ago

Context dependant phrases, from simple pronouns to whole domain specific terms, are still randomly wrong, sometimes appallingly so. Hallucinations still happen. Auto-AI translation youtube uses is, bluntly, horrid. Any jokes, even obvious ones, are still fumbled frequently.

LLM based translation looks more convincing but requires the same level of scrutiny that previous tools did. From a workflow POV they only added higher compute costs for very questionable gains.

2 more replies

j / k navigate · click thread line to collapse

0 comments

TeMPOraL2mo ago

When would've thought that to solve natural language translation, one would first need to solve... natural language.

All those arguments about agents and hallucinations kind of distracted people from noticing we've accodentally built a universal translator.

KuinoxOP2mo ago

It's been at least 10 years that google translate had hallucinations. Some translation simply change depending of a ponctuation mark. But peoples complain only now that they heard about AI.

Of course it's not perfect, but I agree that we didn't had a machine translation as good before.

krige2mo ago

As someone both exposed to this new wave of LLM style translation in various media, and someone who has background in translation, no we didn't.

TeMPOraL2mo ago

krige2mo ago

LLM based translation looks more convincing but requires the same level of scrutiny that previous tools did. From a workflow POV they only added higher compute costs for very questionable gains.

2 more replies

j / k navigate · click thread line to collapse