Menu

1 / 3

Beyond Closed-World Jev: Training-Free Choice, Yes/No, and Open-Word Extraction in One Tree Pass

DEV Community: nlp·Akintunde·about 15 hours ago
#S8MJUGBc
#devtonlp#devto#dev#l73#l59
Long article (7617 words) — no AI summary yet.
Reading 0:00
15s threshold

Using Tree Attention and the model's built-in vocabulary table ( lm_head ) to get multiple-choice, yes/no, and single-word answers without generating text token-by-token.

Code repository (MIT License): github.com/Akintunde102/pada

Every file, class, function, and benchmark mentioned below links directly to the exact file and line numbers in that repo.

Note on how this post is organized: I wrote this so someone new to machine learning can read it top-to-bottom without getting lost in jargon. If you already work with Transformers and PyTorch, each technical section starts with a short ML Summary block containing the tensor shapes, equations, and code paths so you can skip the plain-English analogy right below it.

1. Why I built this

On September 15, 2026, TypeSafe AI (founded by Diogo Almeida, Erik Gafni, and Sasha Sheng) launched Jev in early access.…

Continue reading — create a free account

Join HashtagPLUS to read full articles, follow hashtags, vote, and join the conversation.