guidance VS lmql

[1]: https://github.com/guidance-ai/guidance/tree/main

> The most obvious usage of this is forcing a model to output valid JSON

Isn't this something that Outlines [0], Guidance [1] and others [2] already solve much more elegantly?

0. https://github.com/outlines-dev/outlines

1. https://github.com/guidance-ai/guidance

2. https://github.com/sgl-project/sglang

Why do you have Guidance in caps?

https://github.com/guidance-ai/guidance

or ...

https://huggingface.co/docs/text-generation-inference/concep...

or ... ?

A quick glance through these, they don't seem yet to call json_object on OpenAI with the word JSON in the prompt, which works wonders with the 0125 models.

have been experimenting with guidance and lmql. a bit too early to give any well formed opinions but really do like the idea of constraining llm output.

AutoGen (https://github.com/microsoft/autogen) is orthogonal: it's designed for agents to converse with each other.

The original comparison to LangChain from Microsoft was Guidance (https://github.com/guidance-ai/guidance) which appears to have shifted development a bit. I haven't had much experience with it but from the examples it still seems like needless overhead.

Thanks for your comment.

I did not know about "Betteridge's law of headlines", quite interesting. Thanks for sharing :)

You raise some interesting points.

1) Safety: It is true that LVMs and LLMs have unknown biases and could potentially create unsafe content. However, this is not necessarily unique to them, for example, Google had the same problem with their supervised learning model https://www.theverge.com/2018/1/12/16882408/google-racist-go.... It all depends on the original data. I believe we need systems on top of our models to ensure safety. It is also possible to restrict the output domain of our models (https://github.com/guidance-ai/guidance). Instead of allowing our LVMs to output any words, we could restrict it to only being able to answer "red, green, blue..." when giving the color of a car.

2) Cost: You are right right now LVMs are quite expensive to run. As you said are a great way to go to market faster but they cannot run on low-cost hardware for the moment. However, they could help with training those smaller models. Indeed, with see in the NLP domain that a lot of smaller models are trained on data created with GPT models. You can still distill the knowledge of your LVMs into a custom smaller model that can run on embedded devices. The advantage is that you can use your LVMs to generate data when it is scarce and use it as a fallback when your smaller device is uncertain of the answer.

3) Labelling data: I don't think labeling data is necessarily cheap. First, you have to collect the data, depending on the frequency of your events could take months of monitoring if you want to build a large-scale dataset. Lastly, not all labeling is necessarily cheap. I worked at a semiconductor company and labeled data was scarce as it required expert knowledge and could only be done by experienced employees. Indeed not all labelling can be done externally.

However, both approaches are indeed complementary and I think systems that will work the best will rely on both.

Thanks again for the thought-provoking discussion. I hope this answer some of the concerns you raised

I've had a bit of trouble getting function calling to work with cases that aren't just extracting some data from the input. The format is correct but it was harder to get the correct data if it wasn't a simple extraction.

Hopefully OpenAI and others will offer something like https://github.com/guidance-ai/guidance at some point to guarantee overall output structure.

Failed validations will retry, but from what I've seen JSONSchema + generated JSON examples are decently reliable in practice for gpt-3.5-turbo and extremely reliable on gpt-4.

have been experimenting with guidance and lmql. a bit too early to give any well formed opinions but really do like the idea of constraining llm output.

lqml

This is also similar in spirit to LMQL

https://github.com/eth-sri/lmql

LLM calls are just function calls, so most functional composition is already afforded by any general-purpose language out there. If you need fancy stuff, use something like Python‘s functools.

Working on https://github.com/eth-sri/lmql (shameless plug, sorry), we have always found that compositional abstractions on top of LMQL are mostly there already, once you internalize prompts being functions.

LMQL is another.

if you want guidance control for LLMs, you should check this out: https://lmql.ai/