Multi-Task Transfer Matters During Instruction-Tuning

David Mueller, Mark Dredze, Nicholas Andrews


Abstract
Instruction-tuning trains a language model on hundreds of tasks jointly to improve a model’s ability to learn in-context;however, the mechanisms that drive in-context learning are poorly understood and, as a result, the role of instruction-tuning on in-context generalization is poorly understood as well.In this work, we study the impact of instruction-tuning on multi-task transfer: how well a model’s parameters adapt to an unseen task via fine-tuning.We find that instruction-tuning negatively impacts a model’s transfer to unseen tasks, and that model transfer and in-context generalization are highly correlated, suggesting that this catastrophic forgetting may impact in-context learning.We study methods to improve model transfer, finding that multi-task training—how well the training tasks are optimized—can significantly impact ICL generalization; additionally, we find that continual training on unsupervised pre-training data can mitigate forgetting and improve ICL generalization as well.Finally, we demonstrate that, early into training, the impact of instruction-tuning on model transfer to tasks impacts in-context generalization on that task.Overall, we provide significant evidence that multi-task transfer is deeply connected to a model’s ability to learn a task in-context.
Anthology ID:
2024.findings-acl.883
Volume:
Findings of the Association for Computational Linguistics: ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
14880–14891
Language:
URL:
https://aclanthology.org/2024.findings-acl.883
DOI:
10.18653/v1/2024.findings-acl.883
Bibkey:
Cite (ACL):
David Mueller, Mark Dredze, and Nicholas Andrews. 2024. Multi-Task Transfer Matters During Instruction-Tuning. In Findings of the Association for Computational Linguistics: ACL 2024, pages 14880–14891, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Multi-Task Transfer Matters During Instruction-Tuning (Mueller et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-acl.883.pdf