06-28-2024, 03:30 PM
Putting this in a separate post since it's independent of the previous matter:
A little while ago Selden suggested possibly using an LLM or other AI to comb through the EG text and report back what it thinks are errors. While false positives would no doubt exist to some extent, this would be a very useful tool to help us manage the large and growing EG. If such a thing is possible or reasonably easy to make I do not know, but it also might be worth a chat with Trond.
A little while ago Selden suggested possibly using an LLM or other AI to comb through the EG text and report back what it thinks are errors. While false positives would no doubt exist to some extent, this would be a very useful tool to help us manage the large and growing EG. If such a thing is possible or reasonably easy to make I do not know, but it also might be worth a chat with Trond.