Hugging Face introduces a decision-model interface
Hugging Face presented decision models as an alternative to generating text for classification-like tasks, returning structured decisions and probabilities through a typed interface. The article demonstrates the Transformers decision pipeline with GEV-26B-Decide and explains Choice, Score, and Noul primitives, while also describing custom training and a forthcoming TRL integration whose pull request remained open at publication.
The interface makes decisions and probabilities first-class software outputs, offering applications a typed alternative to asking generative language models to produce JSON answers.
Large Language Models (LLMs) are exceptionally good at generating text. We use them to answer questions, write code, reason about problems, and so much more. But not every problem requires generating text.
Consider a customer support system receiving the following message:
"I was charged twice for my subscription. I need a refund today."
The system needs to determine which team should handle the request, how urgent it is, and whether the customer is requesting a refund.
We could ask an LLM to return a JSON object containing these answers.
llama cli -hf ggml-org/gemma-4-e4b-it-GGUF:Q4_0
> We have the following customer support ticket, "I was charged twice for my subscription. I need a refund today.", which team should handle the request, how urgent it is, and whether the customer is requesting a refund. Give the answers in JSON.
[Start thinking]
...
[end thinking]
```json
{
"team_to_handle": "Billing",
"urgency": "High",
"request_refund": true
}
#Hugging Face#Transformers#GEV-26B-Decide#TRL#TypeSafe AI#Jev#Unsloth#Qwen