I always thought that the point of instruction tuning and ability to use prompts to get the model to do 0 shot tasks was that you don't have to collect tons of example data/samples. The method proposed here requires you to have tons of data. If you have that, why not just fine tune the underlying model?
Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting
31–33 of 33 posts
Re: Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting
#32Here is a basic implementation of this in Ruby:
https://gist.github.com/schappim/ad8b4953486617c7f813751c8ee...
Re: Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting
#33Holy unnecessary use of terminology to explain a reverse graph traversal. “Loss”, “gradients”, “differentiating”— no! stop! This must be what AI hype actually is. Complete incoherent language to explain a very straight forward concept. This is just: LLMs judging intermediate node outputs, and reverse traversing the graph while doing so until it modifies the original prompt.
Interestingly, backpropagation is the natural way I'd describe this process, not reverse graph traversal.
Background difference I suppose.
> This must be what AI hype actually is. Complete incoherent language to explain a very straightforward concept.
True, a lot of papers overdo the jargon just for hype purposes. My favorite funniest example is this one from Google Research (and universities) (have linked the paper review video below)
See the YouTube chapter about "Multidiffusion" (around 38minutes)
They spent multiple paragraphs formulating an "optimisation problem" which when peeled down amounts to taking the mean, just to be able to superficially cite their own previous paper.
Quite the sorry state of things.