Researchers extract visibly pushdown grammars from neural language models
A new arXiv paper presents a fully black-box method for extracting visibly pushdown grammars (VPGs) from neural language models, building on the VPL* learning framework for tree automata.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv cs.CL
Learning of Tree Automata applied to Neural Language Acceptors