mlsanityData to 7 Oct 2026

STAD: Self-Training with Ambiguous Data for Low-Resource Relation Extraction

arXiv 2209.01431 ↗First posted 3 Sep 2022cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

We present a simple yet effective self-training approach, named as STAD, for low-resource relation extraction. The approach first classifies the auto-annotated instances into two groups: confident instances and uncertain instances, according to the probabilities predicted by a teacher model. In contrast to most previous studies, which mainly only use the confident instances for self-training, we make use of the uncertain instances. To this end, we propose a method to identify ambiguous but useful instances from the uncertain instances and then divide the relations into candidate-label set and negative-label set for each ambiguous instance. Next, we propose a set-negative training method on the negative-label sets for the ambiguous instances and a positive training method for the confident instances. Finally, a joint-training method is proposed to build the final relation extraction system on all data. Experimental results on two widely used datasets SemEval2010 Task-8 and Re-TACRED with low-resource settings demonstrate that this new self-training approach indeed achieves significant and consistent improvements when comparing to several competitive self-training systems. Code is publicly available at https://github.com/jjyunlp/STAD

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 18 of its 37 references that are in the map. First posted on arXiv 3 Sep 2022.

Builds on

18 of 37 references.

References · 18

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  3. 2017Zero-Shot Relation Extraction via Reading Comprehensioncited by 425
  4. 2019Matching the Blanks: Distributional Similarity for Relation Learningcited by 271
  5. 2018FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluationcited by 254
  6. 2019NLNL: Negative Learning for Noisy Labelscited by 151
  7. 2020Self-training Improves Pre-training for Natural Language Understandingcited by 96
  8. 2021Re-TACRED: Addressing Shortcomings of the TACRED Datasetcited by 73
  9. 2019Long-tail Relation Extraction via Knowledge Graph Embeddings and Graph Convolution Networkscited by 61
  10. 2019Partial Label Learning with Self-Guided Retrainingcited by 47
  11. 2020Semi-supervised Relation Extraction via Incremental Meta Self-Trainingcited by 35
  12. 2021Gradient Imitation Reinforcement Learning for Low Resource Relation Extractioncited by 33
6 more
  1. 2016Improved Relation Classification by Deep Recurrent Neural Networks with Data Augmentationcited by 26
  2. 2019Deep Contextualized Self-training for Low Resource Dependency Parsingcited by 24
  3. 2020Data Rejuvenation: Exploiting Inactive Training Examples for Neural Machine Translationcited by 17
  4. 2021Self-Training Sampling with Monolingual Data Uncertainty for Neural Machine Translationcited by 17
  5. 2021MapRE: An Effective Semantic Mapping Approach for Low-resource Relation Extractioncited by 16
  6. 2022Information Extraction in Low-Resource Scenarios: Survey and Perspectivecited by 10

Built on by

Not cited yet.