Enhancing Lifelong Language Learning by Improving Pseudo-Sample Generation
To achieve lifelong language learning, pseudo-rehearsal methods leverage samples generated from a language model to refresh the knowledge of previously learned tasks. Without proper controls, however, these methods could fail to retain the knowledge of complex tasks with longer texts since most of t...
Main Authors: | Kasidis Kanwatchara, Thanapapas Horsuwan, Piyawat Lertvittayakumjorn, Boonserm Kijsirikul, Peerapon Vateekul |
---|---|
Format: | Article |
Language: | English |
Published: |
The MIT Press
2022-06-01
|
Series: | Computational Linguistics |
Online Access: | http://dx.doi.org/10.1162/coli_a_00449 |
Similar Items
-
Meta Lifelong-Learning With Selective and Task-Aware Adaptation
by: Thanapapas Horsuwan, et al.
Published: (2024-01-01) -
Prompt-Based Label-Aware Framework for Few-Shot Multi-Label Text Classification
by: Thanakorn Thaminkaew, et al.
Published: (2024-01-01) -
Enhance Text-to-Text Transfer Transformer with Generated Questions for Thai Question Answering
by: Puri Phakmongkol, et al.
Published: (2021-11-01) -
Enhancing Network Visibility and Security with Advanced Port Scanning Techniques
by: Rana Abu Bakar, et al.
Published: (2023-08-01) -
Argumentative explanations for pattern-based text classifiers
by: Piyawat Lertvittayakumjorn, et al.
Published: (2023-01-01)