# Handling Noisy Labels for Robustly Learning from Self-Training Data for Low-Resource Sequence Labeling

## TL;DR

This paper introduces a novel neural network approach that explicitly models clean and noisy labels separately to improve the robustness of self-training in low-resource sequence labeling tasks like Chunking and NER.

## Contribution

It proposes the Clean and Noisy Label Neural Network that effectively handles noisy self-labeled data by explicit label modeling, enhancing low-resource sequence labeling performance.

## Key findings

- Outperforms baseline methods on Chunking and NER tasks.
- Robustly handles noisy self-labeled data in low-resource settings.
- Combines explicit and implicit noise handling for improved results.

## Abstract

In this paper, we address the problem of effectively self-training neural networks in a low-resource setting. Self-training is frequently used to automatically increase the amount of training data. However, in a low-resource scenario, it is less effective due to unreliable annotations created using self-labeling of unlabeled data. We propose to combine self-training with noise handling on the self-labeled data. Directly estimating noise on the combined clean training set and self-labeled data can lead to corruption of the clean data and hence, performs worse. Thus, we propose the Clean and Noisy Label Neural Network which trains on clean and noisy self-labeled data simultaneously by explicitly modelling clean and noisy labels separately. In our experiments on Chunking and NER, this approach performs more robustly than the baselines. Complementary to this explicit approach, noise can also be handled implicitly with the help of an auxiliary learning task. To such a complementary approach, our method is more beneficial than other baseline methods and together provides the best performance overall.