CodeIE: Large Code Generation Models are Better Few-Shot Information Extractors
This addresses the problem of structured output generation in IE for NLP practitioners, offering a novel approach that improves performance in few-shot scenarios.
The paper tackles the challenge of performing information extraction (IE) tasks like named entity recognition and relation extraction with large language models by recasting structured outputs as code instead of natural language, using code generation models (Code-LLMs) like Codex, and shows that this method consistently outperforms fine-tuned IE-specific models and prompting natural language LLMs in few-shot settings across seven benchmarks.
Large language models (LLMs) pre-trained on massive corpora have demonstrated impressive few-shot learning ability on many NLP tasks. A common practice is to recast the task into a text-to-text format such that generative LLMs of natural language (NL-LLMs) like GPT-3 can be prompted to solve it. However, it is nontrivial to perform information extraction (IE) tasks with NL-LLMs since the output of the IE task is usually structured and therefore is hard to be converted into plain text. In this paper, we propose to recast the structured output in the form of code instead of natural language and utilize generative LLMs of code (Code-LLMs) such as Codex to perform IE tasks, in particular, named entity recognition and relation extraction. In contrast to NL-LLMs, we show that Code-LLMs can be well-aligned with these IE tasks by designing code-style prompts and formulating these IE tasks as code generation tasks. Experiment results on seven benchmarks show that our method consistently outperforms fine-tuning moderate-size pre-trained models specially designed for IE tasks (e.g., UIE) and prompting NL-LLMs under few-shot settings. We further conduct a series of in-depth analyses to demonstrate the merits of leveraging Code-LLMs for IE tasks.