# Structural Automata for Neural Semantic Parsing

arXiv: https://arxiv.org/abs/2610.13896 · HTML: https://mlsanity.com/paper/2610.13896/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 30 references that are in the map.
- First posted on arXiv 9 Oct 2026.

## Builds on (most cited first, 23 of 23)

- [Neural Text Generation with Unlikelihood Training](https://mlsanity.com/paper/1908.04319/) (2019)
- [PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models](https://mlsanity.com/paper/2109.05093/) (2021)
- [Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning](https://mlsanity.com/paper/2305.13971/) (2023)
- [Guiding LLMs The Right Way: Fast, Non-Invasive Constrained Generation](https://mlsanity.com/paper/2403.06988/) (2024)
- [Leveraging Grammar and Reinforcement Learning for Neural Program Synthesis](https://mlsanity.com/paper/1805.04276/) (2018)
- [Automata-based constraints for language model decoding](https://mlsanity.com/paper/2407.08103/) (2024)
- [The Parallel Meaning Bank: Towards a Multilingual Corpus of Translations Annotated with Compositional Meaning Representations](https://mlsanity.com/paper/1702.03964/) (2017)
- [Type-Constrained Code Generation with Language Models](https://mlsanity.com/paper/2504.09246/) (2025)
- [IterGen: Iterative Semantic-aware Structured LLM Generation with Backtracking](https://mlsanity.com/paper/2410.07295/) (2024)
- [A Pseudo-Semantic Loss for Autoregressive Models with Logical Constraints](https://mlsanity.com/paper/2312.03905/) (2023)
- [SMATCH++: Standardized and Extended Evaluation of Semantic Graphs](https://mlsanity.com/paper/2305.06993/) (2023)
- [Exploring Neural Methods for Parsing Discourse Representation Structures](https://mlsanity.com/paper/1810.12579/) (2018)
- [ChopChop: a Programmable Framework for Semantically Constraining the Output of Language Models](https://mlsanity.com/paper/2509.00360/) (2025)
- [BenchCLAMP: A Benchmark for Evaluating Language Models on Syntactic and Semantic Parsing](https://mlsanity.com/paper/2206.10668/) (2022)
- [Evaluating Scoped Meaning Representations](https://mlsanity.com/paper/1802.08599/) (2018)
- [Pre-Trained Language-Meaning Models for Multilingual Parsing and Generation](https://mlsanity.com/paper/2306.00124/) (2023)
- [Grammar-Forced Translation of Natural Language to Temporal Logic using LLMs](https://mlsanity.com/paper/2512.16814/) (2025)
- [An Analysis of the Utility of Explicit Negative Examples to Improve the Syntactic Abilities of Neural Language Models](https://mlsanity.com/paper/2004.02451/) (2020)
- [Neural Semantic Parsing with Extremely Rich Symbolic Meaning Representations](https://mlsanity.com/paper/2404.12698/) (2024)
- [Repair, Not Improvement: Decomposing Constrained Decoding in Tool-Call Abstention](https://mlsanity.com/paper/2608.13959/) (2026)
- [Retrieval-Augmented Semantic Parsing: Improving Generalization with Lexical Knowledge](https://mlsanity.com/paper/2412.10207/) (2024)
- [Verifier-Guided Code Translation via Meta-Step Decoding](https://mlsanity.com/paper/2605.17626/) (2026)
- [Stay Within Your Bounds: Distance-Guided Decoding for Guaranteed Context-Free Grammar Compliance](https://mlsanity.com/paper/2608.28229/) (2026)

## Abstract (arXiv)

Neural semantic parsers generate meaning representations token by token, yet well-formedness depends on references, scope, and graph relations that unfold during generation. Constrained decoding prevents locally invalid tokens but does not guarantee that a valid prefix can be completed within a fixed output-length limit. We investigate structural automata at three stages of DRS parsing (decoding, post hoc processing, and training) and additionally evaluate decoding on AMR. For decoding, we introduce accepting-path decoding (APD), which maintains a verified completion within the remaining length limit and follows the model's preferred token whenever that token can also be completed. APD guarantees an accepted output without retracting tokens. Under tight length limits, it improves Smatch++ over local masking by 5.1 and 2.4 points on DRS and by 6.4 and 3.2 points on AMR, comparable to rollback under the same limit, which retracts tokens. For post hoc processing, invalid outputs concentrate on the same inputs: eight unconstrained samples yield no valid output for 10.2% of inputs with limited training data, compared with 0.4% with full data, where filtering comes within 0.2 Smatch++ points of APD. For training, penalizing automaton-rejected tokens improves development Smatch++ by 3.6-11.0 points after 10% of training, but this advantage diminishes as cross-entropy alone reduces the probability assigned to rejected tokens from 14% to below 2% within the first sixth of training with limited data.