Overview

Semantic parsing studies how to translate language into explicit structures—such as logical forms, meaning graphs, database queries, and event frames—so systems can answer questions, access data, or support reasoning. Approaches range from grammar- and type-based methods to neural parsers and large language models, often using schemas, intermediate representations, or constrained decoding to guide outputs. Deterministic validation and execution can make generated queries or programs more reliable, while multilingual data and annotation methods help extend parsing across languages and domains. A central challenge is producing structures that are semantically correct and generalize to unfamiliar compositions: valid syntax or strong performance in one direction does not guarantee reliable interpretation.

Latest papers

CardsList

34 papers6 papers in the last four weeks, up 50% on the four weeks before. 0.0% of all new papers.

  1. A Method for Learning Large-Scale Computational Construction Grammars from Semantically Annotated Corpora

    Mar 13, 2026Paul Van Eecke, Katrien BeulsSemantic Parsing

  2. Logics-Parsing-Omni Technical Report

    Mar 10, 2026Xin An, Jingyi Cai, Xiangyang Chen +22Multimodal UnderstandingSemantic Parsing

  3. Effective QA-driven Annotation of Predicate-Argument Relations Across Languages

    Feb 26, 2026Jonathan Davidov, Aviv Slobodkin, Shmuel Tomi Klein +3Semantic ParsingCross-Lingual Transfer

  4. On Robustness of Prompt-based Semantic Parsing with Large Pre-trained Language Model: An Empirical Study on Codex

    Jan 30, 2023Terry Yue Zhuo, Zhuang Li, Yujin Huang +4Adversarial Robustness of Language ModelsSemantic Parsing