Hacker News Reader

Comment by Dzugaru | Hacker News Reader

Outstanding work! I've participated in the challenge, but didn't get far. One of the questions I had at the time was - if I'm going to use ML to detect ink, could it invent hallucinated letters, or even parts of text, and how to prevent that?

View on HN

Yes, it's quite possible for ML to hallucinate ink, though it is on a much more local scale, like predicting a slightly longer stroke, filling in more of a character than is actually in the data, etc. Perhaps enough to change a reading of a character or show where ink isnt. It is difficult for ink detection to hallucinate grammatical and idiomatic greek and latin.

What is the input to the ML algorithm? Does it know the surrounding context so that it has a chance to deduce "if this stroke is slightly longer then the end result will be idiomatic greek and latin"?

The input is 3d chunks of reconstructed CT data from our scans. I can't remember the specifics but maybe enough voxels for .5mm^3 at a time or so? They're all available for free from https://registry.opendata.aws/vesuvius-challenge-herculaneum... . Our trained models are all available at https://huggingface.co/scrollprize

Not all machine learning is generative AI.

True but like regular document scanning software there can be errors in detection.

Just as with redacted documents (consistently blocked terms) or bad OCR jobs (wrong or missing characters), even if only a certain percentage comes out unmangled it is more readable than having no data at all.

A stable base corpus and some dynamic programming will allow you to clean up the remainder[0].

[0]: http://stackoverflow.com/a/11642687/2449774

The problem is when you can't tell which bits are unmangled. OCR systems will happily give you plausible but wrong readings, and even some scanners/copiers will change things: https://dkriesel.com/en/blog/2013/0802_xerox-workcentres_are...

Yeah. There was a weird Xerox printer bug that swapped digits (turning 6s into 8s) on scanned documents caused by the JBIG2 image format [1].

[1] https://www.dkriesel.com/en/blog/2013/0802_xerox-workcentres...

Reply on news.ycombinator.com