GRID: Grammar-Railed Decoding for Enterprise SQL Generation

Researchers introduced GRID, a grammar-constrained decoding engine for enterprise SQL generation that enforces syntactic validity, role-based access control, and provable guarantees via LALR(1) parser state keys and byte-level trie bridging. GRID achieves near-constant per-token cost (3.6-6.7 μs median) and improves execution accuracy by 13 points on Spider at 0.5B scale, with a hash-chained audit trail for compliance.

arXiv:2607.11951v1 Announce Type: new Abstract: Large language models can write SQL, but enterprise deployment demands more than plausible text: outputs must be syntactically valid, must respect per-role and per-schema policy, must carry provable not best-effort guarantees, must not slow down as generations grow, and must leave a compliance-grade record of every decision. We present GRID Grammar-Railed Decoding , a grammar-constrained decoding engine that keys exact next-token masks on parser configurations lexer scan state x LALR 1 stack rather than on token sequences, and uses the incrementally advanced LALR 1 parser itself as a viable-prefix oracle. LLM tokens are bridged to grammar terminals by a byte-level trie walk with a context-independent/context-dependent split that makes cache-key soundness hold by construction. Role-based access control is compiled into the language: role projections subset the grammar's productions and schema lexicons restrict identifier terminals, so forbidden verbs and identifiers are unreachable at mask level. Four guarantees soundness, completeness, termination, and near-constant per-token cost are stated with explicit preconditions and each paired with a test or benchmark. Rust kernels bring the per-token mask to a 3.6-6.7 us median, ahead of llguidance at p50 and p90 on two tokenizers with zero false rejects; per-token guard cost is position-flat at n=16,000. On Spider, constrained decoding is worth +13 execution-accuracy points at 0.5B, and one checker-guided repair pass over the provably mask-unenforceable residue column-level policy lifts a 7B model to 94.5% executable. A hash-chained per-token audit trail replays bit-identically with 100% tamper detection. We state plainly what the mask cannot do distribution faithfulness, column-level RBAC, non-LALR 1 languages and where measured cost remains.