ZEROTOP: Zero-Shot Task-Oriented Semantic Parsing using Large Language Models
Submission Type: Regular Short Paper
Submission Track: Semantics: Lexical, Sentence level, Document Level, Textual Inference, etc.
Submission Track 2: NLP Applications
Keywords: zero-shot, semantic parsing, mtop, large language models, llm, QA datasets
TL;DR: LLMs fail in zero-shot semantic parsing because they cannot detect unanswerable questions and we improve it by training a LLM that can abstain.
Abstract: We explore the use of large language models (LLMs) for zero-shot semantic parsing. Semantic parsing involves mapping natural language utterances to task-specific meaning representations. LLMs are generally trained on publicly available text and code and cannot be expected to directly generalize to domain-specific parsing tasks in a zero-shot setting. In this work, we propose ZEROTOP, a zero-shot task-oriented parsing method that decomposes semantic parsing problem into a set of abstractive and extractive question-answering (QA) problems. For each utterance, we prompt the LLM with questions corresponding to its top-level intent and a set of slots and use the LLM generations to construct the target meaning representation. We observe that current LLMs fail to detect unanswerable questions; and as a result, cannot handle questions corresponding to missing slots. We address this by fine-tuning a language model on public QA datasets using synthetic negative samples. Experimental results show that our QA-based decomposition paired with the fine-tuned LLM can zero-shot parse ≈ 16% of utterances in the MTOP dataset.
Submission Number: 830
Loading