Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting

arxiv.org

Post reply on HN
Print view View original

1–10 of 33 posts

Re: Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting

#3

Post by hnuser123456 »

meame2010 wrote: post #2

Implemented in AdalFlow: https://github.com/SylphAI-Inc/AdalFlow

Congrats on the paper! I read through some of the github docs and read through the paper, this sounds very impressive, but I'm trying to think of how to best use this in practice... is the idea that I could give some kind of high-level task/project description (like a Python project), and this framework would intelligently update its own prompting to avoid getting stuck and to continue "gaining skill" throughout the process of working on a task? Could this be used to build such a system? Very curious to learn more.

Re: Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting

#4

Post by meame2010 »

hnuser123456 wrote: post #3
meame2010 wrote: post #2

Implemented in AdalFlow: https://github.com/SylphAI-Inc/AdalFlow

Congrats on the paper! I read through some of the github docs and read through the paper, this sounds very impressive, but I'm trying to think of how to best use this in practice... is the idea that I could give some kind of high-level task/project description (like a Python project), and this framework would intelligently update its own prompting to avoid getting stuck and to continue "gaining skill" throughout the…

you need a training dataset, and a task pipeline that works. You can refer to this doc: https://adalflow.sylph.ai/use_cases/question_answering.html

Re: Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting

#5

Post by hnuser123456 »

meame2010 wrote: post #4
hnuser123456 wrote: post #3

Earlier quoted context omitted.

Congrats on the paper! I read through some of the github docs and read through the paper, this sounds very impressive, but I'm trying to think of how to best use this in practice... is the idea that I could give some kind of high-level task/project description (like a Python project), and this framework would intelligently update its own prompting to avoid getting stuck and to continue "gaining skill" throughout the…

you need a training dataset, and a task pipeline that works. You can refer to this doc: https://adalflow.sylph.ai/use_cases/question_answering.html

Thank you, I missed the use cases section, that explains a lot. Nice documentation. Might play with this when I get home.

Re: Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting

#8

Post by xg15 »

Just read the abstract so far. Sounds amazing, but just for the sake of understanding, what would be the inputs and outputs of such a system? If the prompt is generated, how do you tell the system what you'd like to have? And what is the ground truth that is trained against? Examples of the desired text?

Re: Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting

#9

Post by seanhunter »

nullc wrote: post #6

requires a backwards trained llm no? I don't think anyone has pretrained a remotely-close-to-SOTA sized backwards model.

Haven’t read the paper yet just the abstract, but it sounds like it uses a backwards trained llm itself to generate prompts and examples but can do the autodiff on any llm.
Post reply on HN