mlsanityData to 7 Oct 2026

Towards Continual Knowledge Learning of Language Models

arXiv 2110.03215 ↗First posted 7 Oct 2021cs.CL cs.LGcited by 172

Not on the map. Why?

Abstract (arXiv)

Large Language Models (LMs) are known to encode world knowledge in their parameters as they pretrain on a vast amount of web corpus, which is often utilized for performing knowledge-dependent downstream tasks such as question answering, fact-checking, and open dialogue. In real-world scenarios, the world knowledge stored in the LMs can quickly become outdated as the world changes, but it is non-trivial to avoid catastrophic forgetting and reliably acquire new knowledge while preserving invariant knowledge. To push the community towards better maintenance of ever-changing LMs, we formulate a new continual learning (CL) problem called Continual Knowledge Learning (CKL). We construct a new benchmark and metric to quantify the retention of time-invariant world knowledge, the update of outdated knowledge, and the acquisition of new knowledge. We adopt applicable recent methods from literature to create several strong baselines. Through extensive experiments, we find that CKL exhibits unique challenges that are not addressed in previous CL setups, where parameter expansion is necessary to reliably retain and learn knowledge simultaneously. By highlighting the critical causes of knowledge forgetting, we show that CKL is a challenging and important problem that helps us better understand and train ever-changing LMs. The benchmark datasets, evaluation script, and baseline code to reproduce our results are available at https://github.com/joeljang/continual-knowledge-learning.

Key facts. Cited by 172 papers in the map (data through 7 Oct 2026). Builds on 45 of its 62 references that are in the map. First posted on arXiv 7 Oct 2021.

Builds on

45 of 62 references.

References · 45

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  4. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  5. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  6. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  7. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  8. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  9. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  10. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  11. 2019Language Models as Knowledge Bases?cited by 1,922
  12. 2017Gradient Episodic Memory for Continual Learningcited by 1,871
33 more
  1. 2020Don't Stop Pretraining: Adapt Language Models to Domains and Taskscited by 1,548
  2. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,362
  3. 2016Progressive Neural Networkscited by 1,332
  4. 2020How Much Knowledge Can You Pack Into the Parameters of a Language Model?cited by 713
  5. 2021Deduplicating Training Data Makes Language Models Bettercited by 705
  6. 2017Lifelong Learning with Dynamically Expandable Networkscited by 689
  7. 2021Knowledge Neurons in Pretrained Transformerscited by 590
  8. 2020KILT: a Benchmark for Knowledge Intensive Language Taskscited by 571
  9. 2019ELI5: Long Form Question Answeringcited by 567
  10. 2021Carbon Emissions and Large Neural Network Trainingcited by 561
  11. 2019Defending Against Neural Fake Newscited by 553
  12. 2021Editing Factual Knowledge in Language Modelscited by 548
  13. 2018Wizard of Wikipedia: Knowledge-Powered Conversational agentscited by 530
  14. 2017Zero-Shot Relation Extraction via Reading Comprehensioncited by 425
  15. 2021Entity-Based Knowledge Conflicts in Question Answeringcited by 349
  16. 2020K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapterscited by 318
  17. 2021Time-Aware Language Models as Temporal Knowledge Basescited by 290
  18. 2021Rainbow Memory: Continual Learning with a Memory of Diverse Samplescited by 273
  19. 2021Beyond Goldfish Memory: Long-Term Open-Domain Conversationcited by 266
  20. 2020Modifying Memories in Transformer Modelscited by 222
  21. 2021Internet-Augmented Dialogue Generationcited by 217
  22. 2021PAQ: 65 Million Probably-Asked Questions and What You Can Do With Themcited by 198
  23. 2019Episodic Memory in Lifelong Language Learningcited by 188
  24. 2020Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Biascited by 183
  25. 2020Recall and Learn: Fine-tuning Deep Pretrained Language Models with Less Forgettingcited by 169
  26. 2019LAMOL: LAnguage MOdeling for Lifelong Language Learningcited by 156
  27. 2021SituatedQA: Incorporating Extra-Linguistic Contexts into QAcited by 145
  28. 2020Question and Answer Test-Train Overlap in Open-Domain Question Answering Datasetscited by 137
  29. 2021Decontextualization: Making Sentences Stand-Alonecited by 101
  30. 2019E-BERT: Efficient-Yet-Effective Entity Embeddings for BERTcited by 96
  31. 2021Can Generative Pre-trained Language Models Serve as Knowledge Bases for Closed-book QA?cited by 62
  32. 2020Facts as Experts: Adaptable and Interpretable Neural Memory over Symbolic Knowledgecited by 37
  33. 2021An Efficient Transformer Decoder with Compressed Sub-layerscited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 15; 2023: 47; 2024: 57; 2025: 29; 2026: 23.

172 papers.

Sort:

By area:

  1. 2023Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!cited by 1,106
  2. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 860
  3. 2024A Survey on Large Language Models for Code Generationcited by 646
  4. 2022Knowledge Unlearning for Mitigating Privacy Risks in Language Modelscited by 424
  5. 2023A Survey on Large Language Models for Recommendationcited by 378
  6. 2024Knowledge Conflicts for LLMs: A Surveycited by 274
  7. 2022RealTime QA: What's the Answer Right Now?cited by 239
  8. 2023Evaluating the Ripple Effects of Knowledge Editing in Language Modelscited by 238
  9. 2024Continual Learning of Large Language Models: A Comprehensive Surveycited by 228
  10. 2023Class-Incremental Learning: A Surveycited by 209
  11. 2022Factuality Enhanced Language Models for Open-Ended Text Generationcited by 200
  12. 2023Continual Pre-training of Language Modelscited by 194
38 more
  1. 2022Patching open-vocabulary models by interpolating weightscited by 187
  2. 2022Large Language Models with Controllable Working Memorycited by 169
  3. 2024Continual Learning for Large Language Models: A Surveycited by 144
  4. 2023Continual Pre-Training of Large Language Models: How to (re)warm your model?cited by 135
  5. 2022TimeLMs: Diachronic Language Models from Twittercited by 129
  6. 2022StreamingQA: A Benchmark for Adaptation to New Knowledge over Time in Question Answering Modelscited by 123
  7. 2024Simple and Scalable Strategies to Continually Pre-train Large Language Modelscited by 122
  8. 2022TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Modelscited by 117
  9. 2021Lifelong Pretraining: Continually Adapting Language Models to Emerging Corporacited by 114
  10. 2023The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-Tuningcited by 106
  11. 2024From Matching to Generation: A Survey on Generative Information Retrievalcited by 89
  12. 2024The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Modelscited by 89
  13. 2023Red teaming ChatGPT via Jailbreaking: Bias, Robustness, Reliability and Toxicitycited by 86
  14. 2022Continual Pre-Training Mitigates Forgetting in Language and Visioncited by 82
  15. 2025Lifelong Learning of Large Language Model based Agents: A Roadmapcited by 78
  16. 2022ELLE: Efficient Lifelong Pre-training for Emerging Datacited by 70
  17. 2024Reuse, Don't Retrain: A Recipe for Continued Pretraining of Language Modelscited by 69
  18. 2023Can LMs Learn New Entities from Descriptions? Challenges in Propagating Injected Knowledgecited by 66
  19. 2023Evaluating GPT-4 and ChatGPT on Japanese Medical Licensing Examinationscited by 63
  20. 2022Continual Learning with Foundation Models: An Empirical Study of Latent Replaycited by 61
  21. 2022Entity Cloze By Date: What LMs Know About Unseen Entitiescited by 55
  22. 2024Instruction-tuned Language Models are Better Knowledge Learnerscited by 54
  23. 2023Knowledge Card: Filling LLMs' Knowledge Gaps with Plug-in Specialized Language Modelscited by 46
  24. 2024Recent Advances of Foundation Language Models-based Continual Learning: A Surveycited by 46
  25. 2023Meta-Learning Online Adaptation of Language Modelscited by 43
  26. 2023TiC-CLIP: Continual Training of CLIP Modelscited by 38
  27. 2024IRCAN: Mitigating Knowledge Conflicts in LLM Generation via Identifying and Reweighting Context-Aware Neuronscited by 36
  28. 2024AKEW: Assessing Knowledge Editing in the Wildcited by 35
  29. 2024Online Adaptation of Language Models with a Memory of Amortized Contextscited by 35
  30. 2023Audio-Visual Class-Incremental Learningcited by 33
  31. 2023How Do Large Language Models Capture the Ever-changing World Knowledge? A Review of Recent Advancescited by 33
  32. 2023Efficient Continual Pre-training for Building Domain Specific Large Language Modelscited by 33
  33. 2023chatClimate: Grounding Conversational AI in Climate Sciencecited by 32
  34. 2023Give Me the Facts! A Survey on Factual Knowledge Probing in Pre-trained Language Modelscited by 32
  35. 2023Emptying the Ocean with a Spoon: Should We Edit Models?cited by 27
  36. 2022Language Anisotropic Cross-Lingual Model Editingcited by 26
  37. 2024Learnable Privacy Neurons Localization in Language Modelscited by 26
  38. 2023Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Modelscited by 25

50 most cited of 172 All 172 (JSON)