Bag of Lies: Robustness in Continuous Pre-training BERT

• URL: http://arxiv.org/abs/2406.09967v1
• Date: Fri, 14 Jun 2024 12:16:08 GMT
• Title: Bag of Lies: Robustness in Continuous Pre-training BERT
• Authors: Ine Gevers, Walter Daelemans,
• Abstract summary: This study aims to acquire more insights into the continuous pre-training phase of BERT regarding entity knowledge. Since the last update of BERT's pre-training data, the model has little to no entity knowledge about COVID-19. We compare the baseline BERT model with the further pre-trained variants on the fact-checking benchmark Check-COVID.
• Score: 2.4850657856181946
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This study aims to acquire more insights into the continuous pre-training phase of BERT regarding entity knowledge, using the COVID-19 pandemic as a case study. Since the pandemic emerged after the last update of BERT's pre-training data, the model has little to no entity knowledge about COVID-19. Using continuous pre-training, we control what entity knowledge is available to the model. We compare the baseline BERT model with the further pre-trained variants on the fact-checking benchmark Check-COVID. To test the robustness of continuous pre-training, we experiment with several adversarial methods to manipulate the input data, such as training on misinformation and shuffling the word order until the input becomes nonsensical. Surprisingly, our findings reveal that these methods do not degrade, and sometimes even improve, the model's downstream performance. This suggests that continuous pre-training of BERT is robust against misinformation. Furthermore, we are releasing a new dataset, consisting of original texts from academic publications in the LitCovid repository and their AI-generated false counterparts.

Related papers

• Pre-training and Diagnosing Knowledge Base Completion Models [58.07183284468881]
  We introduce and analyze an approach to knowledge transfer from one collection of facts to another without the need for entity or relation matching. The main contribution is a method that can make use of large-scale pre-training on facts, which were collected from unstructured text. To understand the obtained pre-trained models better, we then introduce a novel dataset for the analysis of pre-trained models for Open Knowledge Base Completion.
  arXiv  Detail & Related papers  (2024-01-27T15:20:43Z)
• Self-Distillation for Further Pre-training of Transformers [83.84227016847096]
  We propose self-distillation as a regularization for a further pre-training stage. We empirically validate the efficacy of self-distillation on a variety of benchmark datasets for image and text classification tasks.
  arXiv  Detail & Related papers  (2022-09-30T02:25:12Z)
• Continual Pre-Training Mitigates Forgetting in Language and Vision [43.80547864450793]
  We show that continually pre-trained models are robust against catastrophic forgetting. We provide empirical evidence supporting the fact that self-supervised pre-training is more effective in retaining previous knowledge than supervised protocols.
  arXiv  Detail & Related papers  (2022-05-19T07:27:12Z)
• Knowledge Distillation as Efficient Pre-training: Faster Convergence, Higher Data-efficiency, and Better Transferability [53.27240222619834]
  Knowledge Distillation as Efficient Pre-training aims to efficiently transfer the learned feature representation from pre-trained models to new student models for future downstream tasks. Our method performs comparably with supervised pre-training counterparts in 3 downstream tasks and 9 downstream datasets requiring 10x less data and 5x less pre-training time.
  arXiv  Detail & Related papers  (2022-03-10T06:23:41Z)
• On the Transferability of Pre-trained Language Models: A Study from Artificial Datasets [74.11825654535895]
  Pre-training language models (LMs) on large-scale unlabeled text data makes the model much easier to achieve exceptional downstream performance. We study what specific traits in the pre-training data, other than the semantics, make a pre-trained LM superior to their counterparts trained from scratch on downstream tasks.
  arXiv  Detail & Related papers  (2021-09-08T10:39:57Z)
• The MultiBERTs: BERT Reproductions for Robustness Analysis [86.29162676103385]
  Re-running pretraining can lead to substantially different conclusions about performance. We introduce MultiBERTs: a set of 25 BERT-base checkpoints. The aim is to enable researchers to draw robust and statistically justified conclusions about pretraining procedures.
  arXiv  Detail & Related papers  (2021-06-30T15:56:44Z)
• Is BERT a Cross-Disciplinary Knowledge Learner? A Surprising Finding of Pre-trained Models' Transferability [74.11825654535895]
  We investigate whether the power of the models pre-trained on text data, such as BERT, can be transferred to general token sequence classification applications. We find that even on non-text data, the models pre-trained on text converge faster than the randomly models.
  arXiv  Detail & Related papers  (2021-03-12T09:19:14Z)
• A Systematic Evaluation of Transfer Learning and Pseudo-labeling with BERT-based Ranking Models [2.0498977512661267]
  We evaluate transferability of BERT-based neural ranking models across five English datasets. Each of our collections has a substantial number of queries, which enables a full-shot evaluation mode. We find that training on pseudo-labels can produce a competitive or better model compared to transfer learning.
  arXiv  Detail & Related papers  (2021-03-04T21:08:06Z)
• GiBERT: Introducing Linguistic Knowledge into BERT through a Lightweight Gated Injection Method [29.352569563032056]
  We propose a novel method to explicitly inject linguistic knowledge in the form of word embeddings into a pre-trained BERT. Our performance improvements on multiple semantic similarity datasets when injecting dependency-based and counter-fitted embeddings indicate that such information is beneficial and currently missing from the original model.
  arXiv  Detail & Related papers  (2020-10-23T17:00:26Z)
• Cost-effective Selection of Pretraining Data: A Case Study of Pretraining BERT on Social Media [18.21146856681127]
  Recent studies on domain-specific BERT models show that effectiveness on downstream tasks can be improved when models are pretrained on in-domain data. We pretrain two models on tweets and forum text respectively, and empirically demonstrate the effectiveness of these two resources. In addition, we investigate how similarity measures can be used to nominate in-domain pretraining data.
  arXiv  Detail & Related papers  (2020-10-02T18:06:31Z)

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.