rustastra

5 ms·

1. 1.▲
   Hallucination is a problem we'll have to live with for a long time(theregister.com)

   9 points by rustastra 2y ago|11 comments
2. 2.▲
   by rustastra 2y ago

   Same with symbolic systems!
3. 3.▲
   by rustastra 2y ago

   aiui a natural language question e.g. "What is the refund policy?" gets matched against formalized contracts, and the relevant bit of the contract gets translated into natural language deterministically. At least this is the way I
4. 4.▲
   Automated reasoning to remove LLM hallucinations(aws.amazon.com)

   57 points by rustastra 2y ago|38 comments
5. 5.▲
   by rustastra 3y ago

   I couldn't agree more with this take: > Our demand for stimulating content is being overtaken by supply. Analogously, with AI, we might be in a world where scientific progress is accelerated beyond our wildest dreams, where we have
6. 6.▲
   by rustastra 4y ago

   Curious to learn more about prompt engineering takeaways here. Was feeding more context (or chapters of textbooks, bits of papers, documentation) helpful? It does seem like layering information and being very precise helps a lot. Eerily lik
7. 7.▲
   by rustastra 6y ago

   As a user, am I expected to write the labeling algorithms myself or do you offer some in-built ones?
8. 8.▲
   by rustastra 6y ago

   Could feature extraction be helpful here?
9. 9.▲
   by rustastra 6y ago

   I am very curious to know which pre-trained models work better for this task and whether it's possible at all to do without a neural net...