mlsanityData to 7 Oct 2026

ConTinTin: Continual Learning from Task Instructions

arXiv 2203.08512 ↗First posted 16 Mar 2022cs.CLcited by 26

Not on the map. Why?

Abstract (arXiv)

The mainstream machine learning paradigms for NLP often work with two underlying presumptions. First, the target task is predefined and static; a system merely needs to learn to solve it exclusively. Second, the supervision of a task mainly comes from a set of labeled examples. A question arises: how to build a system that can keep learning new tasks from their instructions? This work defines a new learning paradigm ConTinTin (Continual Learning from Task Instructions), in which a system should learn a sequence of new tasks one by one, each task is explained by a piece of textual instruction. The system is required to (i) generate the expected outputs of a new task by learning from its instruction, (ii) transfer the knowledge acquired from upstream tasks to help solve downstream tasks (i.e., forward-transfer), and (iii) retain or even improve the performance on earlier tasks after learning new tasks (i.e., backward-transfer). This new problem is studied on a stream of more than 60 tasks, each equipped with an instruction. Technically, our method InstructionSpeak contains two strategies that make full use of task instructions to improve forward-transfer and backward-transfer: one is to learn from negative outputs, the other is to re-visit instructions of previous tasks. To our knowledge, this is the first time to study ConTinTin in NLP. In addition to the problem formulation and our promising approach, this work also contributes to providing rich analyses for the community to better understand this novel learning problem.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). Builds on 23 of its 37 references that are in the map. First posted on arXiv 16 Mar 2022.

Builds on

23 of 37 references.

References · 23

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
  3. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
  4. 2020Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inferencecited by 934
  5. 2021Cross-Task Generalization via Natural Language Crowdsourcing Instructionscited by 696
  6. 2018Overcoming catastrophic forgetting with hard attention to the taskcited by 672
  7. 2021DER: Dynamically Expandable Representation for Class Incremental Learningcited by 355
  8. 2019Cosmos QA: Machine Reading Comprehension with Contextual Commonsense Reasoningcited by 281
  9. 2018The Natural Language Decathlon: Multitask Learning as Question Answeringcited by 235
  10. 2019Episodic Memory in Lifelong Language Learningcited by 188
  11. 2019LAMOL: LAnguage MOdeling for Lifelong Language Learningcited by 156
  12. 2020Continual Lifelong Learning in Natural Language Processing: A Surveycited by 152
11 more
  1. 2019Quoref: A Reading Comprehension Dataset with Questions Requiring Coreferential Reasoningcited by 100
  2. 2020The Turking Test: Can Language Models Understand Instructions?cited by 82
  3. 2019Sentence Embedding Alignment for Lifelong Relation Extractioncited by 70
  4. 2021Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Taskscited by 46
  5. 2021Learn Continually, Generalize Rapidly: Lifelong Knowledge Accumulation for Few-shot Learningcited by 35
  6. 2021Incremental Few-shot Text Classification with Multi-round New Classes: Formulation, Dataset and Systemcited by 30
  7. 2019Negative Training for Neural Dialogue Response Generationcited by 24
  8. 2018Lifelong Learning for Sentiment Classificationcited by 21
  9. 2017Using Task Descriptions in Lifelong Machine Learning for Improved Performance and Zero-Shot Transfercited by 19
  10. 2017Toward Continual Learning for Conversational Agentscited by 14
  11. 2020Dynamic memory to alleviate catastrophic forgetting in continuous learning settingscited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 6; 2023: 7; 2024: 7; 2025: 3; 2026: 3.

26 papers.

Sort:

By area:

  1. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  2. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 860
  3. 2022OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalizationcited by 235
  4. 2024Continual Learning of Large Language Models: A Comprehensive Surveycited by 228
  5. 2022Fine-tuned Language Models are Continual Learnerscited by 152
  6. 2024Continual Learning for Large Language Models: A Surveycited by 144
  7. 2023LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructionscited by 124
  8. 2022Continual Learning of Natural Language Processing Tasks: A Surveycited by 83
  9. 2023Style Over Substance: Evaluation Biases for Large Language Modelscited by 68
  10. 2024Towards Lifelong Learning of Large Language Models: A Surveycited by 60
  11. 2024Recent Advances of Foundation Language Models-based Continual Learning: A Surveycited by 46
  12. 2022Robustness of Learning from Task Instructionscited by 34
14 more
  1. 2022Continual Training of Language Models for Few-Shot Learningcited by 33
  2. 2023Large Language Model Instruction Following: A Survey of Progresses and Challengescited by 31
  3. 2023MUFFIN: Curating Multi-Faceted Instructions for Improving Instruction-Followingcited by 30
  4. 2025MLLM-CL: Continual Learning for Multimodal Large Language Modelscited by 21
  5. 2023Class-Incremental Learning based on Label Generationcited by 11
  6. 2025When Continue Learning Meets Multimodal Large Language Model: A Surveycited by 8
  7. 2023Sub-network Discovery and Soft-masking for Continual Learning of Mixed Taskscited by 7
  8. 2024CLoG: Benchmarking Continual Learning of Image Generation Modelscited by 5
  9. 2024Boosting LLM Translation Skills without General Ability Loss via Rationale Distillationcited by 5
  10. 2026Beyond Static Models: An Evolving Framework for Continual Learning in Large Language Models across Training Stagescited by 4
  11. 2024Benchmarking Generative Models on Computational Thinking Tests in Elementary Visual Programmingcited by 3
  12. 2026φ-DPO: Fairness Direct Preference Optimization Approach to Continual Learning in Large Multimodal Modelscited by 2
  13. 2026Turning Back Without Forgetting: Selective Backward Refinement for Parameter-Efficient Continual Learningcited by 1
  14. 2025Enhancing Memory Recall in LLMs with Gauss-Tin: A Hybrid Instructional and Gaussian Replay Approachcited by 0