We explore the use of large pretrained language models as few-shot semantic parsers. The goal in semantic parsing is to generate a structured meaning representation given a natural language input. However, language models are trained to generate natural language. To bridge the gap, we use language models to paraphrase inputs into a controlled sublanguage resembling English that can be automatically mapped to a target meaning representation. With a small amount of data and very little code to convert into English-like representations, we provide a blueprint for rapidly bootstrapping semantic parsers and demonstrate good performance on multiple tasks.

Constrained Language Models Yield Few-Shot Semantic Parsers

It has long been known that the syntactic structure and the semantic representation of a sentence are closely associated [1, 2]. However, it is still a hard problem to exploit the syntactic-semantic correspondence in end-to-end neural semantic parsing, mainly due to the partial consistency between their structures. In this paper, we propose a neural dependency to semantics transformation model – Dep2Sem, which can effectively learn the structure correspondence between dependency trees and formal meaning representations. Based on Dep2Sem, a dependency-informed attention mechanism is proposed to exploit syntactic structure for neural semantic parsing. Experiments on Geo, Jobs, and Atis benchmarks show that our approach can significantly enhance the performance of neural semantic parsers.

Dependency to Semantics: Structure Transformation and Syntax-Guided Attention for Neural Semantic Parsing

“Semantic parsing aims to convert natural language utterances to logical forms. A critical challenge for constructing semantic parsers is the lack of labeled data. In this paper, we propose a data synthesis and iterative refinement framework for neural semantic parsing, which can build semantic parsers without annotated logical forms. We first generate a naive corpus by sampling logic forms from knowledge bases and synthesizing their canonical utterances. Then, we further propose a bootstrapping algorithm to iteratively refine data and model, via a denoising language model and knowledge-constrained decoding. Experimental results show that our approach achieves competitive performance on GEO, ATIS and OVERNIGHT datasets in both unsupervised and semi-supervised data settings.”

Data Synthesis and Iterative Refinement for Neural Semantic Parsing without Annotated Logical Forms

We explore the use of large pretrained language models as few-shot semantic parsers. The goal in semantic parsing is to generate a structured meaning representation given a natural language input. However, language models are trained to generate natural language. To bridge the gap, we use language models to paraphrase inputs into a controlled sublanguage resembling English that can be automatically mapped to a target meaning representation. Our results demonstrate that with only a small amount of data and very little code to convert into English-like representations, our blueprint for rapidly bootstrapping semantic parsers leads to surprisingly effective performance on multiple community tasks, greatly exceeding baseline methods also trained on the same limited data.

Semantic parsing maps natural language questions into logical forms, which can be executed against a knowledge base for answers. In real-world applications, the performance of a parser is often limited by the lack of training data. To facilitate zero-shot learning, data synthesis has been widely studied to automatically generate paired questions and logical forms. However, data synthesis methods can hardly cover the diverse structures in natural languages, leading to a large gap in sentence structure between synthetic and natural questions. In this paper, we propose a decomposition-based method to unify the sentence structures of questions, which benefits the generalization to natural questions. Experiments demonstrate that our method significantly improves the semantic parser trained on synthetic data (+7.9% on KQA and +8.9% on ComplexWebQuestions in terms of exact match accuracy). Extensive analysis demonstrates that our method can better generalize to natural questions with novel text expressions compared with baselines. Besides semantic parsing, our idea potentially benefits other semantic understanding tasks by mitigating the distracting structure features. To illustrate this, we extend our method to the task of sentence embedding learning, and observe substantial improvements on sentence retrieval (+13.1% for Hit@1).

/pdf/bridging-the-gap-between-synthetic-and-natural-questions-via-1nepl0m0.pdf

Bridging the Gap between Synthetic and Natural Questions via Sentence Decomposition for Semantic Parsing

Semantic parsing is challenging due to the structure gap and the semantic gap between utterances and logical forms. In this paper, we propose an unsupervised semantic parsing method - Synchronous Semantic Decoding (SSD), which can simultaneously resolve the semantic gap and the structure gap by jointly leveraging paraphrasing and grammar-constrained decoding. Specifically, we reformulate semantic parsing as a constrained paraphrasing problem: given an utterance, our model synchronously generates its canonical utterancel and meaning representation. During synchronously decoding: the utterance paraphrasing is constrained by the structure of the logical form, therefore the canonical utterance can be paraphrased controlledly; the semantic decoding is guided by the semantics of the canonical utterance, therefore its logical form can be generated unsupervisedly. Experimental results show that SSD is a promising approach and can achieve state-of-the-art unsupervised semantic parsing performance on multiple datasets.

/pdf/from-paraphrasing-to-semantic-parsing-unsupervised-semantic-nm44z0sxou.pdf

From Paraphrasing to Semantic Parsing: Unsupervised Semantic Parsing via Synchronous Semantic Decoding

Semantic parsing is challenging due to the structure gap and the semantic gap between utterances and logical forms. In this paper, we propose an unsupervised semantic parsing method - Synchronous Semantic Decoding (SSD), which can simultaneously resolve the semantic gap and the structure gap by jointly leveraging paraphrasing and grammar constrained decoding. Specifically, we reformulate semantic parsing as a constrained paraphrasing problem: given an utterance, our model synchronously generates its canonical utterance and meaning representation. During synchronous decoding: the utterance paraphrasing is constrained by the structure of the logical form, therefore the canonical utterance can be paraphrased controlledly; the semantic decoding is guided by the semantics of the canonical utterance, therefore its logical form can be generated unsupervisedly. Experimental results show that SSD is a promising approach and can achieve competitive unsupervised semantic parsing performance on multiple datasets.

Weipeng Zhang

Papers

From Paraphrasing to Semantic Parsing: Unsupervised Semantic Parsing via Synchronous Semantic Decoding

From Paraphrasing to Semantic Parsing: Unsupervised Semantic Parsing via Synchronous Semantic Decoding