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.
Ruan and Yaxing Cai and Ruihang Lai and Ziyi Xu and Yilong Zhao and Tianqi Chen , journal=
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.