A trie automaton with Aho-Corasick precomputed token masks makes finite-set constrained decoding near-constant per step, giving 7x faster masking and 29x higher vLLM batch throughput than XGrammar.
ICML , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Trie Automata for Constrained Decoding over Large Finite Sets
A trie automaton with Aho-Corasick precomputed token masks makes finite-set constrained decoding near-constant per step, giving 7x faster masking and 29x higher vLLM batch throughput than XGrammar.