Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm · HackerPapers discussions

Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm — HN submission

Fri, 02 Dec 2022 18:38:41 GMT

Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm — mentioned by eru

Thu, 14 Jul 2022 04:06:25 GMT

Perhaps https://arxiv.org/pdf/2102.07350.pdf Also Gwern has done a lot on this.

Paper and discussion history. A mention is not an endorsement.

Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm — mentioned by gwern

Mon, 22 Feb 2021 18:18:18 GMT

https://arxiv.org/abs/2102.07350 already calls it "metaprompt" :) I gave it a quick stab a while ago but I think prompt programming is too new, and you can't easily cram demonstrations into an existing prompt, for it to really work well. It's more promising to train models on examples of tasks from instructions, or work on directly optimizing prompts for a goal (https://arxiv.org/abs/2101.00190) - it's a differentiable model and a whitebox, so use that power!

Paper and discussion history. A mention is not an endorsement.