You are viewing archived messages.
Go here to search the history.

Scott Anderson 2023-04-18 05:16:18

LLM prompting with a visual programming language arxiv.org/abs/2304.08103

Ivan Lugo 2023-04-18 23:53:09

I’m still writing my Mac app for this, dang it.. there’s always a faster paper.

Ivan Lugo 2023-04-18 23:55:07

I’m curious how much of this was generated too πŸ€“

Ivan Lugo 2023-04-19 00:00:39

The end sample mentions injection of expert domain knowledge, and calls out psychology and mock interviews.

Interviews? Sure. Psych I’m not sure.. I can hardly trust a human to understand my meanings and intent therapeutically, and the context build up is part of building session value up. That seems quite hard to stuff into a small sliding window.

Christian Gill 2023-04-19 09:30:58

I've seen the idea of using LLMs for psychotherapy (and the adjacent)

Christian Gill 2023-04-19 09:31:28

I personally like the idea of having the model be a conversation partner that helps you to think

Christian Gill 2023-04-19 09:32:15

But agree with you that when it comes to mental health there's a need for human comprehension

Thanh Dinh 2023-04-20 10:27:11

I don't know if I want to laugh of cry πŸ˜‚. This is done just by duct-taping a bunch of APIs and GPT-4 prompts together. Maybe the future of programming will be more English than code? chameleon-llm.github.io. Paper (arxiv.org/abs/2304.09842) and Code (github.com/lupantech/chameleon-llm)