Abstract
Identifying unexpected domain-shifted instances in natural language processing is crucial in real-world applications. Previous works identify the out-of-distribution (OOD) instance by leveraging a single global feature embedding to represent the sentence, which cannot characterize subtle OOD patterns well. Another major challenge current OOD methods face is learning effective low-dimensional sentence representations to identify the hard OOD instances that are semantically similar to the in-distribution (ID) data. In this paper, we propose a new unsupervised OOD detection method, namely Semantic Role Labeling Guided Out-of-distribution Detection (SRLOOD), that separates, extracts, and learns the semantic role labeling (SRL) guided fine-grained local feature representations from different arguments of a sentence and the global feature representations of the full sentence using a margin-based contrastive loss. A novel self-supervised approach is also introduced to enhance such global-local feature learning by predicting the SRL extracted role. The resulting model achieves SOTA performance on four OOD benchmarks, indicating the effectiveness of our approach. The code is publicly accessible via https://github.com/cytai/SRLOOD.
Original language | English |
---|---|
Title of host publication | The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024) - Main Conference Proceedings |
Editors | Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue |
Place of Publication | Paris France |
Publisher | European Language Resources Association (ELRA) |
Pages | 14641-14651 |
Number of pages | 11 |
ISBN (Electronic) | 9782493814104 |
Publication status | Published - 2024 |
Externally published | Yes |
Event | Joint International Conference on Computational Linguistics and International Conference on Language Resources and Evaluation 2024 - Hybrid, Torino, Italy Duration: 20 May 2024 → 25 May 2024 https://aclanthology.org/volumes/2024.lrec-main/ (Proceedings) https://lrec-coling-2024.org/ (Website) |
Conference
Conference | Joint International Conference on Computational Linguistics and International Conference on Language Resources and Evaluation 2024 |
---|---|
Abbreviated title | LREC-COLING 2024 |
Country/Territory | Italy |
City | Hybrid, Torino |
Period | 20/05/24 → 25/05/24 |
Internet address |
|
Keywords
- Domain Shift
- Out-of-distribution Detection
- Semantic Role Labeling