Best of Both Worlds: A Pliable and Generalizable Neuro-Symbolic Approach for Relation Classification
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917605635260416 |
|---|---|
| author | Vacareanu, Robert Alam, Fahmida Islam, Md Asiful Riaz, Haris Surdeanu, Mihai |
| author_facet | Vacareanu, Robert Alam, Fahmida Islam, Md Asiful Riaz, Haris Surdeanu, Mihai |
| contents | This paper introduces a novel neuro-symbolic architecture for relation classification (RC) that combines rule-based methods with contemporary deep learning techniques. This approach capitalizes on the strengths of both paradigms: the adaptability of rule-based systems and the generalization power of neural networks. Our architecture consists of two components: a declarative rule-based model for transparent classification and a neural component to enhance rule generalizability through semantic text matching. Notably, our semantic matcher is trained in an unsupervised domain-agnostic way, solely with synthetic data. Further, these components are loosely coupled, allowing for rule modifications without retraining the semantic matcher. In our evaluation, we focused on two few-shot relation classification datasets: Few-Shot TACRED and a Few-Shot version of NYT29. We show that our proposed method outperforms previous state-of-the-art models in three out of four settings, despite not seeing any human-annotated training data. Further, we show that our approach remains modular and pliable, i.e., the corresponding rules can be locally modified to improve the overall model. Human interventions to the rules for the TACRED relation \texttt{org:parents} boost the performance on that relation by as much as 26\% relative improvement, without negatively impacting the other relations, and without retraining the semantic matching component. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_03305 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Best of Both Worlds: A Pliable and Generalizable Neuro-Symbolic Approach for Relation Classification Vacareanu, Robert Alam, Fahmida Islam, Md Asiful Riaz, Haris Surdeanu, Mihai Computation and Language Artificial Intelligence This paper introduces a novel neuro-symbolic architecture for relation classification (RC) that combines rule-based methods with contemporary deep learning techniques. This approach capitalizes on the strengths of both paradigms: the adaptability of rule-based systems and the generalization power of neural networks. Our architecture consists of two components: a declarative rule-based model for transparent classification and a neural component to enhance rule generalizability through semantic text matching. Notably, our semantic matcher is trained in an unsupervised domain-agnostic way, solely with synthetic data. Further, these components are loosely coupled, allowing for rule modifications without retraining the semantic matcher. In our evaluation, we focused on two few-shot relation classification datasets: Few-Shot TACRED and a Few-Shot version of NYT29. We show that our proposed method outperforms previous state-of-the-art models in three out of four settings, despite not seeing any human-annotated training data. Further, we show that our approach remains modular and pliable, i.e., the corresponding rules can be locally modified to improve the overall model. Human interventions to the rules for the TACRED relation \texttt{org:parents} boost the performance on that relation by as much as 26\% relative improvement, without negatively impacting the other relations, and without retraining the semantic matching component. |
| title | Best of Both Worlds: A Pliable and Generalizable Neuro-Symbolic Approach for Relation Classification |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2403.03305 |