Re:Show HN: Using LLama2 to Correct OCR Errors
by Reubend on Wednesday August 02, 2023 @07:23PM
(#36979277)
Is it possible that with different hyperparameters, LLAMA 2 would be more faithful to the original? I'm pretty surprised that it would hallucinate anything majorly different than the source material, since that task seems pretty "easy" for LLMs these days.
[ Parent
| on HN ]
Re:Show HN: Using LLama2 to Correct OCR Errors
by eigenvalue on Wednesday August 02, 2023 @07:28PM
(#36979321)
Yes, I'm sure there are ways to reduce the hallucination, though I don't know that these would all be accessible via the Python bindings. I think I might also be sampling too much from the model which is leading to it making up more stuff than normal. It's certainly an area I want to focus on more.