Cuckoo: An IE Free Rider Hatched by Massive Nutrition in LLM's Nest

TL;DR

This paper introduces Cuckoo, a versatile information extraction model that leverages existing LLM data, acting as a free rider to improve IE tasks efficiently without additional data collection.

Contribution

Cuckoo is the first IE model to utilize LLM pre-training and post-training data, enabling effective few-shot adaptation and continuous evolution with LLM advancements.

Findings

Cuckoo outperforms existing pre-trained IE models in few-shot settings.

It leverages 102.6M extractive data from LLM resources.

Cuckoo adapts effectively to traditional and complex IE tasks.

Abstract

Massive high-quality data, both pre-training raw texts and post-training annotations, have been carefully prepared to incubate advanced large language models (LLMs). In contrast, for information extraction (IE), pre-training data, such as BIO-tagged sequences, are hard to scale up. We show that IE models can act as free riders on LLM resources by reframing next-token \emph{prediction} into \emph{extraction} for tokens already present in the context. Specifically, our proposed next tokens extraction (NTE) paradigm learns a versatile IE model, \emph{Cuckoo}, with 102.6M extractive data converted from LLM's pre-training and post-training data. Under the few-shot setting, Cuckoo adapts effectively to traditional and complex instruction-following IE with better performance than existing pre-trained IE models. As a free rider, Cuckoo can naturally evolve with the ongoing advancements in LLM data preparation, benefiting from improvements in LLM training pipelines without additional manual effort.