naver/gdc
Code accompanying our papers on the "Generative Distributional Control" framework
This helps content creators and AI system designers ensure that text generated by large language models meets specific criteria. You input an existing language model and your desired rules (like "must be polite" or "50% of content should mention a specific topic"), and it outputs a fine-tuned model that follows these instructions. This is for anyone who uses AI to generate text and needs to control its characteristics and content.
118 stars. No commits in the last 6 months.
Use this if you need to guide the output of a pre-trained language model to align with specific content policies, ethical guidelines, or desired stylistic and factual distributions without losing its original knowledge.
Not ideal if you're looking for a tool to train a language model from scratch or if your control needs are simple enough to be handled by basic prompt engineering.
Stars
118
Forks
20
Language
Python
License
—
Category
Last pushed
Dec 07, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/generative-ai/naver/gdc"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
yxuansu/PlanGen
[EMNLP'21] Plan-then-Generate: Controlled Data-to-Text Generation via Planning
balhafni/personalized-gen
Code, models, and data for "Personalized Text Generation with Fine-Grained Linguistic Control"....
emmetsite342/visual-transformer-guide
Explain how Transformer AI models work with an interactive, beginner-friendly guide covering key...