Finetuning BERT on Partially Annotated NER Corpora

• URL: http://arxiv.org/abs/2211.14360v1
• Date: Fri, 25 Nov 2022 19:54:30 GMT
• Title: Finetuning BERT on Partially Annotated NER Corpora
• Abstract summary: This work presents an approach to finetuning BERT on partially labelled datasets using self-supervision and label preprocessing. Our approach outperforms the previous LSTM-based label preprocessing baseline, significantly improving the performance on poorly labelled datasets.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Most Named Entity Recognition (NER) models operate under the assumption that training datasets are fully labelled. While it is valid for established datasets like CoNLL 2003 and OntoNotes, sometimes it is not feasible to obtain the complete dataset annotation. These situations may occur, for instance, after selective annotation of entities for cost reduction. This work presents an approach to finetuning BERT on such partially labelled datasets using self-supervision and label preprocessing. Our approach outperforms the previous LSTM-based label preprocessing baseline, significantly improving the performance on poorly labelled datasets. We demonstrate that following our approach while finetuning RoBERTa on CoNLL 2003 dataset with only 10% of total entities labelled is enough to reach the performance of the baseline trained on the same dataset with 50% of the entities labelled.

Related papers

• Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning [88.48555005545694]
  We propose a Controllable Pseudo-label Generation (CPG) framework to expand the labeled dataset with reliable pseudo-labels from the unlabeled dataset.<n>CPG operates through a controllable self-reinforcing optimization cycle.<n>CPG achieves consistent improvements, surpassing state-of-the-art methods by up to $textbf15.97%$ in accuracy.
  arXiv  Detail & Related papers  (2025-10-05T01:52:19Z)
• Label Unification for Cross-Dataset Generalization in Cybersecurity NER [0.0]
  The field of cybersecurity NER lacks standardized labels, making it challenging to combine datasets.<n>We perform a coarse-grained label unification and conduct pairwise cross-dataset evaluations using BiLSTM models.<n>Results show that models trained on unified datasets generalize poorly across datasets.
  arXiv  Detail & Related papers  (2025-07-18T12:47:20Z)
• Wiki-TabNER: Integrating Named Entity Recognition into Wikipedia Tables [18.330753799139845]
  A new dataset, Wiki-TabNER, is proposed to enrich the existing benchmark datasets.<n>This paper describes the distinguishing features of the Wiki-TabNER dataset and the labeling process.<n>In addition, we propose a prompting framework for evaluating the new large language models on the within tables NER task.
  arXiv  Detail & Related papers  (2024-03-07T15:22:07Z)
• Training-Free Generalization on Heterogeneous Tabular Data via Meta-Representation [67.30538142519067]
  We propose Tabular data Pre-Training via Meta-representation (TabPTM) A deep neural network is then trained to associate these meta-representations with dataset-specific classification confidences. Experiments validate that TabPTM achieves promising performance in new datasets, even under few-shot scenarios.
  arXiv  Detail & Related papers  (2023-10-31T18:03:54Z)
• Soft Curriculum for Learning Conditional GANs with Noisy-Labeled and Uncurated Unlabeled Data [70.25049762295193]
  We introduce a novel conditional image generation framework that accepts noisy-labeled and uncurated data during training. We propose soft curriculum learning, which assigns instance-wise weights for adversarial training while assigning new labels for unlabeled data. Our experiments show that our approach outperforms existing semi-supervised and label-noise robust methods in terms of both quantitative and qualitative performance.
  arXiv  Detail & Related papers  (2023-07-17T08:31:59Z)
• Aligning benchmark datasets for table structure recognition [10.323063834827416]
  Benchmark datasets for table structure recognition (TSR) must be carefully processed to ensure they are consistently. In this work, we show that aligning these benchmarks$x$removing errors and inconsistency between them$x$improves performance significantly. Overall we believe our work has significant implications for benchmark processing performance and potentially other tasks as well.
  arXiv  Detail & Related papers  (2023-03-01T18:20:24Z)
• Ground Truth Inference for Weakly Supervised Entity Matching [76.6732856489872]
  We propose a simple but powerful labeling model for weak supervision tasks. We then tailor the labeling model specifically to the task of entity matching. We show that our labeling model results in a 9% higher F1 score on average than the best existing method.
  arXiv  Detail & Related papers  (2022-11-13T17:57:07Z)
• Recognizing Nested Entities from Flat Supervision: A New NER Subtask, Feasibility and Challenges [3.614392310669357]
  This study proposes a new subtask, nested-from-flat NER, which corresponds to a realistic application scenario. We train span-based models and deliberately ignore the spans nested inside labeled entities, since these spans are possibly unlabeled entities. With nested entities removed from the training data, our model achieves 54.8%, 54.2% and 41.1% F1 scores on the subset of spans within entities on ACE 2004, ACE 2005 and GENIA, respectively.
  arXiv  Detail & Related papers  (2022-11-01T06:41:42Z)
• Generating Data to Mitigate Spurious Correlations in Natural Language Inference Datasets [27.562256973255728]
  Natural language processing models often exploit spurious correlations between task-independent features and labels in datasets to perform well only within the distributions they are trained on. We propose to tackle this problem by generating a debiased version of a dataset, which can then be used to train a debiased, off-the-shelf model. Our approach consists of 1) a method for training data generators to generate high-quality, label-consistent data samples; and 2) a filtering mechanism for removing data points that contribute to spurious correlations.
  arXiv  Detail & Related papers  (2022-03-24T09:08:05Z)
• LiDAR dataset distillation within bayesian active learning framework: Understanding the effect of data augmentation [63.20765930558542]
  Active learning (AL) has re-gained attention recently to address reduction of annotation costs and dataset size. This paper performs a principled evaluation of AL based dataset distillation on (1/4th) of the large Semantic-KITTI dataset. We observe that data augmentation achieves full dataset accuracy using only 60% of samples from the selected dataset configuration.
  arXiv  Detail & Related papers  (2022-02-06T00:04:21Z)
• QUEACO: Borrowing Treasures from Weakly-labeled Behavior Data for Query Attribute Value Extraction [57.56700153507383]
  This paper proposes a unified query attribute value extraction system in e-commerce search named QUEACO. For the NER phase, QUEACO adopts a novel teacher-student network, where a teacher network that is trained on the strongly-labeled data generates pseudo-labels. For the AVN phase, we also leverage the weakly-labeled query-to-attribute behavior data to normalize surface form attribute values from queries into canonical forms from products.
  arXiv  Detail & Related papers  (2021-08-19T03:24:23Z)
• Low-Budget Label Query through Domain Alignment Enforcement [48.06803561387064]
  We tackle a new problem named low-budget label query. We first improve an Unsupervised Domain Adaptation (UDA) method to better align source and target domains. We then propose a simple yet effective selection method based on uniform sampling of the prediction consistency distribution.
  arXiv  Detail & Related papers  (2020-01-01T16:52:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.