Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Check out integratability with LLMs / LMQL / JSONFORMER #229

Open
nielstron opened this issue May 24, 2023 · 1 comment
Open

Check out integratability with LLMs / LMQL / JSONFORMER #229

nielstron opened this issue May 24, 2023 · 1 comment
Labels
enhancement New feature or request

Comments

@nielstron
Copy link
Owner

Is your feature request related to a problem? Please describe.
The ambiguity resolver of quantulum is not very good.

Describe the solution you'd like
This may improve by using a significantly larger language model. Either it can be queried directly or programmatically using lmql to decide among the options of ambiguous quantities.

In particular this function could be improved:

def disambiguate_unit(unit, text, lang="en_US", classifier_path=None):

Describe alternatives you've considered
None

Additional context
LLMs are increasingly showing impressive performance in general natural language processing. The classifier in this tool is quite outdated and weak in comparison.

@nielstron nielstron added the enhancement New feature or request label May 24, 2023
@nielstron nielstron changed the title Check out integratability with LLMs / LMQL Check out integratability with LLMs / LMQL / JSONFORMER Jul 5, 2023
@nielstron
Copy link
Owner Author

another promising alternative is https://github.com/1rgs/jsonformer to directly extract units in json format

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

1 participant