Konrad Hinsen 2023-04-03 09:29:10
There are two technologies that I'd expect to be useful in the context of LLMs:
Has anyone seen such connections pointed out or explored yet?
Jason Morris 2023-04-03 14:21:05
I'm looking at a text version of my own tool, isomorphic with the visual language, to make it possible to get LLM to do Codex-style code suggestions. Haven't actually done it, yet, and haven't seen anyone doing it, but I'm thinking precisely the same thing.
Nilesh Trivedi 2023-04-04 05:54:08
I built this command-line bot that automates complex online tasks using both the browser and the APIs of cloud-services: github.com/aicombinator/bot
It's built with GPT-3.5-Turbo, Puppeteer and ActivePieces.
Ivan Lugo 2023-04-05 20:00:48
So is it the case that
agents are specifically built prompt templates that are backed with some fun interprocess coms, passing the results to various piped tools like a python / JS script to get the next ‘prompt piece’, and then looping until something ‘correct’ comes up? I feel github.com/mpaepper/llm_agents simplified the abstractions a bit to help me understand what was going on.
Nilesh Trivedi 2023-04-06 12:23:14
Thinking about multi-agent / multi-tool systems where humans and bots collaborate:
Devin Walters 2023-04-06 23:33:02
DoD calls this “human-machine teaming”
Devin Walters 2023-04-07 00:32:41
(Though it is assumed that there’s a human in the loop, so it’s not the same thing)