Learning Dependency-Based Compositional Semantics

TLDR

Semantic parsing systems translate natural language questions into logical forms to query structured databases, but training such parsers typically requires costly annotated logical‑form examples. This work aims to learn a semantic parser directly from question–answer pairs by treating the logical form as a latent variable. The authors introduce dependency‑based compositional semantics (DCS) and train a log‑linear model over DCS logical forms using an alternating beam‑search and optimization procedure. On two standard semantic‑parsing benchmarks, the proposed system achieves accuracies comparable to state‑of‑the‑art models that rely on annotated logical forms.

Abstract

Suppose we want to build a system that answers a natural language question by representing its semantics as a logical forxm and computing the answer given a structured database of facts. The core part of such a system is the semantic parser that maps questions to logical forms. Semantic parsers are typically trained from examples of questions annotated with their target logical forms, but this type of annotation is expensive. Our goal is to instead learn a semantic parser from question–answer pairs, where the logical form is modeled as a latent variable. We develop a new semantic formalism, dependency-based compositional semantics (DCS) and define a log-linear distribution over DCS logical forms. The model parameters are estimated using a simple procedure that alternates between beam search and numerical optimization. On two standard semantic parsing benchmarks, we show that our system obtains comparable accuracies to even state-of-the-art systems that do require annotated logical forms.