Distantly-Supervised Joint Extraction with Noise-Robust Learning

TL;DR

This paper introduces DENRL, a noise-robust framework for joint entity and relation extraction from distantly-labeled data, effectively handling noisy labels and outperforming large language models on benchmark datasets.

Contribution

The paper proposes a novel noise-robust joint extraction framework, DENRL, that utilizes a transformer backbone and iterative self-adaptation to improve extraction accuracy in noisy distantly-labeled data.

Findings

DENRL outperforms large language model baselines on benchmark datasets.

It effectively handles noisy labels using relation patterns and entity-relation dependencies.

The framework achieves better interpretability with simple heuristics.

Abstract

Joint entity and relation extraction is a process that identifies entity pairs and their relations using a single model. We focus on the problem of joint extraction in distantly-labeled data, whose labels are generated by aligning entity mentions with the corresponding entity and relation tags using a knowledge base (KB). One key challenge is the presence of noisy labels arising from both incorrect entity and relation annotations, which significantly impairs the quality of supervised learning. Existing approaches, either considering only one source of noise or making decisions using external knowledge, cannot well-utilize significant information in the training data. We propose DENRL, a generalizable framework that 1) incorporates a lightweight transformer backbone into a sequence labeling scheme for joint tagging, and 2) employs a noise-robust framework that regularizes the tagging model with significant relation patterns and entity-relation dependencies, then iteratively self-adapts to instances with less noise from both sources. Surprisingly, experiments on two benchmark datasets show that DENRL, using merely its own parametric distribution and simple data-driven heuristics, outperforms large language model-based baselines by a large margin with better interpretability.