Mirror: A Universal Framework for Various Information Extraction Tasks

TL;DR

Mirror introduces a universal, multi-span cyclic graph framework for various information extraction tasks, enabling versatile, efficient, and competitive performance across multiple datasets and settings.

Contribution

It reformulates IE as a multi-span cyclic graph problem and proposes a non-autoregressive decoding algorithm, supporting complex IE, MRC, and classification tasks.

Findings

Outperforms or matches SOTA in few-shot and zero-shot scenarios

Supports multi-span, n-ary extraction, and diverse IE tasks

Demonstrates high versatility across 30 datasets and 8 tasks

Abstract

Sharing knowledge between information extraction tasks has always been a challenge due to the diverse data formats and task variations. Meanwhile, this divergence leads to information waste and increases difficulties in building complex applications in real scenarios. Recent studies often formulate IE tasks as a triplet extraction problem. However, such a paradigm does not support multi-span and n-ary extraction, leading to weak versatility. To this end, we reorganize IE problems into unified multi-slot tuples and propose a universal framework for various IE tasks, namely Mirror. Specifically, we recast existing IE tasks as a multi-span cyclic graph extraction problem and devise a non-autoregressive graph decoding algorithm to extract all spans in a single step. It is worth noting that this graph structure is incredibly versatile, and it supports not only complex IE tasks, but also machine reading comprehension and classification tasks. We manually construct a corpus containing 57 datasets for model pretraining, and conduct experiments on 30 datasets across 8 downstream tasks. The experimental results demonstrate that our model has decent compatibility and outperforms or reaches competitive performance with SOTA systems under few-shot and zero-shot settings. The code, model weights, and pretraining corpus are available at https://github.com/Spico197/Mirror .