Few-Shot Semantic Parsing with Language Models Trained On Code

Anonymous

Abstract: Large language models can perform semantic parsing with little training data, when prompted with in-context examples. It has been shown that this can be improved by formulating the problem as paraphrasing into canonical utterances, which casts the underlying meaning representation into a controlled natural language-like representation. Intuitively, such models can more easily output canonical utterances as they are closer to the natural language used for pre-training. More recently, models also pre-trained on code, like OpenAI Codex, have risen in prominence. Since semantic parsing requires translating natural language into code, such models may prove more adept at it. In this paper, we test this hypothesis and find that Codex performs better at semantic parsing than equivalent GPT-3 models. We find that unlike GPT-3, Codex performs similarly when targeting meaning representations directly, perhaps because meaning representations used in semantic parsing are structured similar to code.

Paper Type: short