Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2403.02718 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913254598508544 |
|---|---|
| author | Huang, Mengyi Xiao, Meng Wang, Ludi Du, Yi |
| author_facet | Huang, Mengyi Xiao, Meng Wang, Ludi Du, Yi |
| contents | Continuous Relation Extraction (CRE) aims to incrementally learn relation knowledge from a non-stationary stream of data. Since the introduction of new relational tasks can overshadow previously learned information, catastrophic forgetting becomes a significant challenge in this domain. Current replay-based training paradigms prioritize all data uniformly and train memory samples through multiple rounds, which would result in overfitting old tasks and pronounced bias towards new tasks because of the imbalances of the replay set. To handle the problem, we introduce the DecouPled CRE (DP-CRE) framework that decouples the process of prior information preservation and new knowledge acquisition. This framework examines alterations in the embedding space as new relation classes emerge, distinctly managing the preservation and acquisition of knowledge. Extensive experiments show that DP-CRE significantly outperforms other CRE baselines across two datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_02718 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DP-CRE: Continual Relation Extraction via Decoupled Contrastive Learning and Memory Structure Preservation Huang, Mengyi Xiao, Meng Wang, Ludi Du, Yi Computation and Language Continuous Relation Extraction (CRE) aims to incrementally learn relation knowledge from a non-stationary stream of data. Since the introduction of new relational tasks can overshadow previously learned information, catastrophic forgetting becomes a significant challenge in this domain. Current replay-based training paradigms prioritize all data uniformly and train memory samples through multiple rounds, which would result in overfitting old tasks and pronounced bias towards new tasks because of the imbalances of the replay set. To handle the problem, we introduce the DecouPled CRE (DP-CRE) framework that decouples the process of prior information preservation and new knowledge acquisition. This framework examines alterations in the embedding space as new relation classes emerge, distinctly managing the preservation and acquisition of knowledge. Extensive experiments show that DP-CRE significantly outperforms other CRE baselines across two datasets. |
| title | DP-CRE: Continual Relation Extraction via Decoupled Contrastive Learning and Memory Structure Preservation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2403.02718 |