# Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study

arXiv: https://arxiv.org/abs/2303.01081 · HTML: https://mlsanity.com/paper/2303.01081/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 49 references that are in the map.
- First posted on arXiv 2 Mar 2023.

## Builds on (most cited first, 28 of 28)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning](https://mlsanity.com/paper/1711.05769/) (2017)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Designing and Interpreting Probes with Control Tasks](https://mlsanity.com/paper/1909.03368/) (2019)
- [Learning multiple visual domains with residual adapters](https://mlsanity.com/paper/1705.08045/) (2017)
- [AdapterHub: A Framework for Adapting Transformers](https://mlsanity.com/paper/2007.07779/) (2020)
- [What do you learn from context? Probing for sentence structure in contextualized word representations](https://mlsanity.com/paper/1905.06316/) (2019)
- [Linguistic Knowledge and Transferability of Contextual Representations](https://mlsanity.com/paper/1903.08855/) (2019)
- [QuAC : Question Answering in Context](https://mlsanity.com/paper/1808.07036/) (2018)
- [Representation Degeneration Problem in Training Natural Language Generation Models](https://mlsanity.com/paper/1907.12009/) (2019)
- [Episodic Memory in Lifelong Language Learning](https://mlsanity.com/paper/1906.01076/) (2019)
- [Continual Lifelong Learning in Natural Language Processing: A Survey](https://mlsanity.com/paper/2012.09823/) (2020)
- [An Empirical Investigation of the Role of Pre-training in Lifelong Learning](https://mlsanity.com/paper/2112.09153/) (2021)
- [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://mlsanity.com/paper/1908.08962/) (2019)
- [Lifelong Pretraining: Continually Adapting Language Models to Emerging Corpora](https://mlsanity.com/paper/2110.08534/) (2021)
- [ELLE: Efficient Lifelong Pre-training for Emerging Data](https://mlsanity.com/paper/2203.06311/) (2022)
- [Efficient Meta Lifelong-Learning with Limited Memory](https://mlsanity.com/paper/2010.02500/) (2020)
- [Multi-granularity hierarchical attention fusion networks for reading comprehension and question answering](https://mlsanity.com/paper/1811.11934/) (2018)
- [Iterative Network Pruning with Uncertainty Regularization for Lifelong Sentiment Classification](https://mlsanity.com/paper/2106.11197/) (2021)

## Built on by (most cited first, 11 of 11)

- [Continual Learning of Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2404.16789/) (2024)
- [Lawyer LLaMA Technical Report](https://mlsanity.com/paper/2305.15062/) (2023)
- [Unlocking the Potential of Model Merging for Low-Resource Languages](https://mlsanity.com/paper/2407.03994/) (2024)
- [Learn or Recall? Revisiting Incremental Learning with Pre-trained Language Models](https://mlsanity.com/paper/2312.07887/) (2023)
- [What Will My Model Forget? Forecasting Forgotten Examples in Language Model Refinement](https://mlsanity.com/paper/2402.01865/) (2024)
- [Demystifying Language Model Forgetting with Low-rank Example Associations](https://mlsanity.com/paper/2406.14026/) (2024)
- [Can LLMs Learn New Concepts Incrementally without Forgetting?](https://mlsanity.com/paper/2402.08526/) (2024)
- [Balancing the Causal Effects in Class-Incremental Learning](https://mlsanity.com/paper/2402.10063/) (2024)
- [MLUBench: A Benchmark for Lifelong Unlearning Evaluation in MLLMs](https://mlsanity.com/paper/2606.12809/) (2026)
- [Revisiting Continual Semantic Segmentation with Pre-trained Vision Models](https://mlsanity.com/paper/2508.04267/) (2025)
- [Mi:dm K 2.5 Pro](https://mlsanity.com/paper/2603.18788/) (2026)

## Abstract (arXiv)

Large pre-trained language models help to achieve state of the art on a variety of natural language processing (NLP) tasks, nevertheless, they still suffer from forgetting when incrementally learning a sequence of tasks. To alleviate this problem, recent works enhance existing models by sparse experience replay and local adaption, which yield satisfactory performance. However, in this paper we find that pre-trained language models like BERT have a potential ability to learn sequentially, even without any sparse memory replay. To verify the ability of BERT to maintain old knowledge, we adopt and re-finetune single-layer probe networks with the parameters of BERT fixed. We investigate the models on two types of NLP tasks, text classification and extractive question answering. Our experiments reveal that BERT can actually generate high quality representations for previously learned tasks in a long term, under extremely sparse replay or even no replay. We further introduce a series of novel methods to interpret the mechanism of forgetting and how memory rehearsal plays a significant role in task incremental learning, which bridges the gap between our new discovery and previous studies about catastrophic forgetting.