Generating Data for Symbolic Language with Large Language Models
Submission Type: Regular Long Paper
Submission Track: Language Modeling and Analysis of Language Models
Submission Track 2: NLP Applications
Keywords: Large Language Model, Data Generation, Symbolic Language, Code Generation
TL;DR: We show that with appropriate prompt and verification, LLMs can serve as viable data generators even for complex symbolic languages, and a small model trained on such synthetic data can outperform much larger LLM data generator.
Abstract: While large language models (LLMs) bring not only performance but also complexity, recent work has started to turn LLMs into data generators rather than task inferencers, where another affordable task model is trained for efficient deployment and inference.
However, such an approach has primarily been applied to natural language tasks, and has not yet been explored for symbolic language tasks with complex structured outputs (e.g., semantic parsing and code generation).
In this paper, we propose SymGen which utilizes LLMs
for generating various annotation-expensive symbolic language data.
SymGen consists of an informative prompt to steer generation and an agreement-based verifier to improve data correctness.
We conduct extensive experiments on six symbolic language tasks across various settings.
Compared with the LLMs, we demonstrate the 1%-sized task model can achieve comparable or better performance, largely cutting inference and deployment costs.
We also show that generated data with only a few human demonstrations can be as effective as over 10 times the amount of human-annotated data when training the task model, saving a considerable amount of annotation effort.
SymGen takes a step toward data generation for annotation-expensive complex tasks, and we release the code at URL.
Submission Number: 2382
Loading