Hacker Neue

Preferences
jiehong parent
The problem with LLMs for translation is when they refuse to do so if the topic being translated isn’t following their policies, even if the context shows it’s fine here.

It can be as simple as discuss one’s own religion

Alex-Programs
I made a tool which translates sentences as you browse, for immersion[0]. I solved this by giving the model a code (specifically, "483") to return in any refusal. Then, if I detect that in the output, I fail over to another model+provider.

I also have a few heuristics (e.g. "I can't translate" in many different languages) to detect if it deviates from that.

It works pretty well.

[0] https://nuenki.app

Jensson
You can just turn that off, at least on Googles model.

This item has no comments currently.